Elsevier

Biological Conservation

Volume 199, July 2016, Pages 123-131
Biological Conservation

Adaptive management for improving species conservation across the captive-wild spectrum

https://doi.org/10.1016/j.biocon.2016.04.026Get rights and content

Highlights

  • We provide a practical guide to adaptive management (AM) of threatened species.

  • AM is useful for iterative decisions where reducing uncertainty improves outcomes.

  • We illustrate how to set up AM for four case studies of captive-wild management.

  • Difficulties in monitoring and poor institutional support are the most common challenges.

  • Clear objectives and hypotheses about uncertainty are the key to successful AM.

Abstract

Conservation of endangered species increasingly envisages complex strategies that integrate captive and wild management actions. Management decisions in this context must be made in the face of uncertainty, often with limited capacity to collect information. Adaptive management (AM) combines management and monitoring, with the aim of updating knowledge and improving decision-making over time. We provide a guide for managers who may realize the potential of AM, but are unsure where to start. The urgent need for iterative management decisions, the existence of uncertainty, and the opportunity for learning offered by often highly-controlled captive environments create favorable conditions for AM. However, experiments and monitoring may be complicated by small sample sizes, and the ability to control the system, including stochasticity and observability, may be limited toward the wild end of the spectrum. We illustrate the key steps to implementing AM in threatened species management using four case studies, including the management of captive programs for cheetah (Acinonyx jubatus) and whooping cranes (Grus americana), of a translocation protocol for Arizona cliffroses Purshia subintegra and of ongoing supplementary feeding of reintroduced hihi (Notiomystis cincta) populations. For each case study, we explain (1) how to clarify whether the decision can be improved by learning (i.e. it is iterative and complicated by uncertainty) and what the management objectives are; (2) how to articulate uncertainty via alternative, testable hypotheses such as competing models or parameter distributions; (3) how to formally define how additional information can be collected and incorporated in future management decisions.

Introduction

Conservation biologists increasingly recognize that successful management of threatened species requires the integration of diverse management techniques (IUCN/SSC, 2008). While conservation approaches are often categorized as focusing on the “wild” or in situ environment versus its “captive” or ex situ counterpart, in reality they span a spectrum of management intensity; few programs involve completely unmanaged wild populations or complete control over captive populations (Redford et al., 2012). For simplicity, in this paper we refer to this spectrum as the captive-wild spectrum.

Along this spectrum, conservation management requires making decisions about which actions to apply. Decisions include whether to establish new populations in breeding centers or via translocations among wild populations, how and when to translocate individuals, and which methods to use to manage wild populations. Incomplete knowledge of the biological system results in uncertainty about how to manage most effectively (Burgman, 2005). On the other hand, threatened species management often requires immediate decisions, limiting the time available for traditional research (Martin et al., 2012b).

Still, management itself can provide opportunities to learn. By monitoring the outcomes of implemented actions, managers can improve their understanding of the system and inform future decisions. This process represents the essence of adaptive management (AM; Holling, 1978, Walters, 1986), which has been increasingly advocated for conservation in recent years (McCarthy and Possingham, 2007, Runge, 2011). With its focus on objectives and uncertainty, AM lies within the more general framework of structured decision making, the process of rationally analyzing decisions (Gregory et al., 2012). AM has been explicitly highlighted as an important tool in comprehensive species conservation strategies (IUCN/SSC, 2008), as well as in guidelines for reintroductions (IUCN/SSC, 2013) and ex situ programs (IUCN/SSC, 2014).

Despite its potential advantages, implementation of AM in conservation is infrequent and often incomplete or unsatisfactory (Westgate et al., 2013). This implementation gap may result from confusion surrounding key concepts and definitions, misunderstanding of the practical barriers to implementation, and inadequate institutional structures and support (Allen and Gunderson, 2011, Gregory et al., 2006). Rather than reviewing those challenges again, with this contribution we seek to assist managers of threatened species programs who understand the potential benefits of AM but are unsure of how to apply it to their specific decision problems. We interpret the conditions and challenges to AM implementation identified by previous studies in the practical context of threatened species management. We then illustrate the process of AM implementation using four case studies along the captive-wild spectrum.

Section snippets

How to get started in adaptive management

Management is adaptive when it explicitly recognizes the effect of uncertainty on decisions, and it seeks to reduce that uncertainty to improve management outcomes. This reduction can be “passive”, where managers make the decision that is considered best under the current knowledge, but apply adequate monitoring to collect specific information that will allow a subsequent re-evaluation of the management decision (Walters, 1986). Alternatively, “active” AM seeks to solve a "dual control"

Conditions and challenges for adaptive management across the captive-wild spectrum

In spite of its intuitive appeal, AM is not suitable for every type of decision problem. Williams et al. (2009) listed the following conditions for the application of AM: (1) the need for immediate action under uncertainty; (2) explicit and measurable objectives; (3) a real choice between alternative actions, which can influence management outcomes; (4) the ability to formulate uncertainty as a set of testable hypotheses; (5) adequate stakeholder support and institutional capacity to sustain an

Adaptive management in practice

In this section, we illustrate the application of AM using four case studies across the captive-wild spectrum, ranging from captive management for reintroduction to ongoing management of reintroduced populations in the wild. The case studies were selected for their illustrative potential, and reflect the experiences of the co-authors. As such, they do not represent an exhaustive review of the conditions and challenges in implementation of AM. Rather, we use them as a basis to describe the

Conclusions

The concept of learning while managing and then using new information to improve outcomes is intuitively appealing for threatened species management, where immediate decisions are often required in the face of incomplete knowledge. However, there is more to AM than just “learning by doing”. Attempting to apply AM where the necessary conditions do not exist, for example where the time horizon does not allow for the application of new information, can represent a poor allocation of resources. On

Acknowledgments

This work was initiated during a workshop funded by the National Environmental Research Program (NERP) at the University of Melbourne in 2014. Manuscript preparation was supported by the University of Melbourne and the ARC Centre of Excellence for Environmental Decisions.

References (60)

  • R.E. Bellman

    Dynamic Programming

    (1957)
  • M. Bloomsmith et al.

    Successful feeding enrichment for captive chimpanzees

    Am. J. Primatol.

    (1988)
  • C. Brownie et al.

    Statistical Inference from Band Recovery Data — A Handbook

    (1985)
  • M. Burgman

    Risks and Decisions for Conservation and Environmental Management

    (2005)
  • S. Canessa et al.

    When do we need more data? A primer on calculating the value of information for applied ecologists

    Methods Ecol. Evol.

    (2015)
  • I. Chadès et al.

    MDPtoolbox: a multi-platform toolbox to solve stochastic dynamic programming problems

    Ecography

    (2014)
  • S.J. Converse et al.

    Demographics of reintroduced populations: estimation, modeling, and decision analysis

    J. Wildl. Manag.

    (2013)
  • S.J. Converse et al.

    A matter of tradeoffs: reintroduction as a multiple objective decision

    J. Wildl. Manag.

    (2013)
  • D.H. Ellis et al.

    Post-release survival of hand-reared and parent-reared Mississippi sandhill cranes

    Condor

    (2000)
  • D.H. Ellis et al.

    Minimum survival rates for Mississippi sandhill cranes: a comparison of hand-rearing and parent-rearing

  • J.G. Ewen et al.

    New Zealand species recovery groups and their role in evidence-based conservation

    J. Appl. Ecol.

    (2013)
  • J.G. Ewen et al.

    Reintroduction objectives, decisions and outcomes: global perspectives from the herpetofauna

    Anim. Conserv.

    (2014)
  • P. Fackler

    MDPSolve user's guide

  • M.J. Folk et al.

    Current status of nonmigratory whooping cranes in Florida

  • P. Goodwin et al.

    Decision Analysis for Management Judgement

    (2004)
  • R. Gregory et al.

    Using structured decision making to help implement a precautionary approach to endangered species management

    Risk Analysis

    (2009)
  • R. Gregory et al.

    Deconstructing adaptive management: criteria for applications to environmental management

    Ecol. Appl.

    (2006)
  • R. Gregory et al.

    Structured Decision Making: A Practical Guide to Environmental Management Choices

    (2012)
  • C.E. Hauser et al.

    Experimental or precautionary? Adaptive management over a range of time horizons

    J. Appl. Ecol.

    (2008)
  • Cited by (42)

    • Shifting precipitation regimes alter the phenology and population dynamics of low latitude ectotherms

      2022, Climate Change Ecology
      Citation Excerpt :

      There are clear places where further study has the potential to reduce uncertainty, such as measuring diapause survival in the field. Our modeling effort is the first step in an iterative adaptive management process [69–71]. More broadly, including both biological and climate uncertainty in an adaptive management framework provides land managers with the information necessary to broadly assess the risk that climate change poses to key resources or populations and to efficiently allocate resources to develop responsive management actions.

    • Adaptive management of species recovery programs: A real-world application for an endangered amphibian

      2019, Biological Conservation
      Citation Excerpt :

      If dealing with larger sites or a more cryptic species, we would have had to account for imperfect detection in the learning process. The already small differences between the alternative learning approaches might have become negligible once this additional layer of uncertainty was taken into account (Canessa et al., 2016b). Imperfect detection may apply to most amphibian species (Schmidt, 2005), which can make the actual benefits of learning much smaller than the theoretical ones (Canessa et al., 2015; Runge et al., 2011).

    • A three-pipe problem: dealing with complexity to halt amphibian declines

      2019, Biological Conservation
      Citation Excerpt :

      However, the analysis also suggested that action should not be delayed until more research is completed. These two findings together support development of an adaptive management program (Canessa et al., 2016; Runge, 2013; Walters, 1986) to guide management of this system. Finally, identification of a management alternative that was robust to a range of objective weights was powerful because it allowed for each of the Boreal Toad Team member agencies to be reasonably accommodated in the identification of a solution.

    • Do substitute species help or hinder endangered species management?

      2019, Biological Conservation
      Citation Excerpt :

      Our experimental test revealed that restoration actions improved endangered species habitat through its effects on key resources, and that differences in resource requirements between the two butterflies resulted in the failure of our substitute species. When restoration experiments are incorporated into an adaptive management framework, each experiment will update knowledge, reduce uncertainty, and refine management actions to those that best promote the endangered species (Canessa et al., 2016). Third, center data collection on small-scale habitat selection, movement, and demographic parameters.

    View all citing articles on Scopus
    View full text