Scanning the Conservation Horizon A Guide to Climate Change Vulnerability Assessment Copyright © 2011 by National Wildlife Federation
Suggested citation: Glick, P., B.A. Stein, and N.A. Edelson, editors. 2011. Scanning the Conservation Horizon: A Guide to Climate Change Vulnerability Assessment. National Wildlife Federation, Washington, D.C.
ISBN 978-0-615-40233-8
Financial support for this publication was provided by the Doris Duke Charitable Foundation, Department of Defense Legacy Resource Management Program, U.S. Fish and Wildlife Service, U.S. Forest Service Rocky Mountain Research Station, U.S. Geological Survey, National Park Service, and National Oceanic and Atmospheric Administration. Note: Financial support does not imply endorsement of this document; use
U.S. Government. of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the
with U.S. Geological Survey Fundamental Science Practices. This publication has met scientific peer review standards and been approved for publication in accordance
Scanning the Conservation Horizon is available online at: www.nwf.org/vulnerabilityguide
Cover: Photo of grizzly bears in Alaska by Michio Hoshino/Minden Pictures.
National Wildlife Federation 901 E Street, Suite 400 Washington, DC 20004 202.797.6000 www.nwf.org
i Scanning the Conservation Horizon A Guide to Climate Change Vulnerability Assessment Scanning the Conservation Horizon A Guide to Climate Change Vulnerability Assessment Edited by Patty Glick, Bruce A. Stein, and Naomi A. Edelson
Dan Gifford
A Guide to Climate Change Vulnerability Assessment ii Lazareva Evgeniya Vulnerability Assessment Workgroup Members
This guidance document is a product of an expert workgroup on climate change vulnerability assessment convened by the National Wildlife Federation in collaboration with the U.S. Fish and Wildlife Service.
Naomi Edelson, National Wildlife Federation (Chair) Nancy Green, U.S. Fish and Wildlife Service (Co-Chair) Rocky Beach, Washington Department of Fish and Wildlife Molly Cross, Wildlife Conservation Society Carolyn Enquist, The Nature Conservancy Deborah Finch, U.S. Forest Service Hector Galbraith, Manomet Center for Conservation Sciences Evan Girvetz, The Nature Conservancy Patty Glick, National Wildlife Federation John Gross, National Park Service Katharine Hayhoe, Texas Tech University; ATMOS Research and Consulting Jennie Hoffman, EcoAdapt Doug Inkley, National Wildlife Federation Bruce Jones, U.S. Geological Survey Linda Joyce, U.S. Forest Service Josh Lawler, University of Washington Dennis Ojima, The Heinz Center for Science, Economics, and the Environment John O’Leary, Massachusetts Division of Fisheries and Wildlife Bruce Stein, National Wildlife Federation Bruce Young, NatureServe
iii Scanning the Conservation Horizon A Guide to Climate Change Vulnerability Assessment Contents Executive Summary ...... 1 Preface ...... 4 I. Introduction ...... 6 (Bruce Stein and Patty Glick) Our Rapidly Changing World ...... 6 Climate Change Adaptation—Putting Vulnerability Assessment in Context ...... 8 Why Assess Vulnerability? ...... 13 II. Vulnerability Assessment Basics ...... 19 (Bruce Stein, Patty Glick, and Jennie Hoffman) Components of Vulnerability ...... 19 Components of Biodiversity ...... 22 Setting Goals and Engaging Stakeholders ...... 27 Selecting Assessment Targets ...... 33 Space and Time: Selecting the Right Scales ...... 35 Complexity: More Isn’t Always Better ...... 37 Embracing Uncertainty ...... 38 III. Assessing the Components of Vulnerability...... 39 (Josh Lawler, Carolyn Enquist, and Evan Girvetz) Assessing Sensitivity ...... 39 Assessing Exposure ...... 44 Assessing Adaptive Capacity ...... 48 IV. Peering into the Future: Climate and Ecological Models ...... 51 (Katharine Hayhoe, Bruce Jones, and John Gross) Climate Models ...... 51 Ecological Response Models ...... 61 V. Addressing Uncertainty in Vulnerability Assessments...... 68 (Linda Joyce, Molly Cross, and Evan Girvetz) The Language of Uncertainty ...... 69 Assessing and Understanding Uncertainty ...... 72 VI. Using Vulnerability Assessment Results ...... 74 (Doug Inkley, Molly Cross, Jennie Hoffman, and John O’Leary) Vulnerability Assessment Outputs ...... 74 Dealing with Uncertainty in Adaptation Planning ...... 79 Looking Ahead ...... 82
A Guide to Climate Change Vulnerability Assessment iv VII. Case Studies ...... 83 Case Study 1. NatureServe’s Climate Change Vulnerability Index for Species in Nevada ...... 85 (Bruce Young, Jennifer Newmark, and Kristin Szabo) Case Study 2. U.S. EPA’s Threatened and Endangered Species Vulnerability Framework ...... 90 (Hector Galbraith and Jeff Price) Case Study 3. Species Vulnerability Assessment for the Middle Rio Grande, New Mexico ...... 96 (Deborah M. Finch, Megan M. Friggens, and Karen E. Bagne) Case Study 4. Vulnerability of Massachusetts Fish and Wildlife Habitats to Climate Change ...... 103 (Hector Galbraith and John O’Leary) Case Study 5. Vulnerability to Sea-Level Rise in the Chesapeake Bay ...... 114 (Patty Glick and Michael Wilson) Case Study 6. An Integrated Climate Change Assessment Framework in the Four Corners Region ...... 122 (Patrick McCarthy and Carolyn Enquist) 129 (Michael Case and Josh Lawler) Case Study 7. Pacific Northwest Climate Change Vulnerability Assessment ...... Glossary ...... 135 Acronyms ...... 138 Climate Change Vulnerability Assessment Resources ...... 139 Publications ...... 139 Web-Based Tools ...... 141 Literature Cited ...... 142
Michael D. Peterson
v Scanning the Conservation Horizon Executive Summary Executive Summary
apid climate change is the Vulnerability assessments are a key tool for informing adaptation planning and our generation. The effects of enabling resource managers R defining conservation issue of climate change are increasingly apparent, to make such judgments. from drowned coastal marshes and drying Scanning the Conservation prairie potholes to melting glaciers. These Horizon is designed to assist climate-driven changes will profoundly and other conservation and wildlife and the habitats on which they resourcefish and wildlife professionals managers to better depend.affect our Indeed, ability preparingto conserve for fish and and coping plan, execute, and interpret with the effects of climate change—an climate change vulnerability endeavor referred to as climate change assessments. Alan D. Wilson adaptation—is emerging as the overarching framework for conservation and natural Climate change vulnerability assessments resource management. provide two essential contributions to
The ecological impacts associated with help in: climate change do not exist in isolation, adaptation planning. Specifically, they but combine with and exacerbate Identifying which species or systems existing stresses on our natural systems. are likely to be most strongly affected by Understanding those interactions will be projected• changes; and critical to designing effective conservation measures. Conservation in an era of Understanding why these resources climate change will require that we are likely to be vulnerable, including the not only acknowledge and address the •interaction between climate shifts and environmental problems of the past but existing stressors. also anticipate and prepare for those of an increasingly uncertain future. Determining which resources are most vulnerable enables managers to better set Developing and implementing effective priorities for conservation action, while understanding why they are vulnerable understanding of the potential impacts provides a basis for developing appropriate ofadaptation climate change strategies on our first natural requires world. an management and conservation responses. To provide the best possible chance for conserving species and ecosystems in a Vulnerability to climate change, as rapidly changing climate, it is essential that managers have the ability to both identify the term is used in this guide, has what we need to do differently in the future, three principle components: sensitivity, as well as which existing strategies and exposure, and adaptive capacity. activities continue to make sense from a climate adaptation perspective.
Executive Summary 1 Vulnerability to climate change, as the term and rate of change the species or system is used in this guide, has three principal is likely to experience. Adaptive capacity components: sensitivity, exposure, addresses the ability of a species or system and adaptive capacity. Vulnerability to accommodate or cope with climate assessments are, therefore, structured change impacts with minimal disruption.
Although climate change Key Steps for Assessing Vulnerability to Climate Change vulnerability assessments can be Determine objectives and scope applied to human infrastructure as well as natural systems, our focus • Identify audience, user requirements, and needed products here is on approaches designed Engage key internal and external stakeholders • to support wildlife conservation Establish and agree on goals and objectives • and ecosystem-based adaptation. Identify suitable assessment targets • Such assessments can target • Determine appropriate spatial and temporal scales various levels of ecological or • Select assessment approach based on targets, user needs, and available resources biological diversity. Because of Gather relevant data and expertise their relevance to most wildlife • Review existing literature on assessment targets and climate impacts management and conservation • Reach out to subject experts on target species or systems practitioners, this guidance • Obtain or develop climatic projections, focusing on ecologically relevant variables focuses on assessments of species, and suitable spatial and temporal scales habitats, and ecosystems, • Obtain or develop ecological response projections detailing approaches for assessing Assess components of vulnerability sensitivity, exposure, and adaptive capacity at each of these biological Evaluate climate sensitivity of assessment targets • levels. Understanding likely Determine likely exposure of targets to climatic/ecological change • future change is central to these Consider adaptive capacity of targets that can moderate potential impact • assessments, and we also provide • Estimate overall vulnerability of targets an overview and guidance for • Document level of confidence or uncertainty in assessments the use of climate and ecological Apply assessment in adaptation planning response models relevant to • Explore why specific targets are vulnerable to inform possible adaptation responses • Consider how targets might fare under various management and climatic scenarios vulnerability assessments. conducting fish and wildlife • Share assessment results with stakeholders and decision-makers • Use results to advance development of adaptation strategies and plans Climate change vulnerability
foremost, intended to support around assessments of these distinct decision-making,assessments and are,as such first they and should components. Sensitivity generally refers to be designed from the start with an eye innate characteristics of a species or system toward the needs of the end users, whether and considers tolerance to changes in such they be on-the-ground managers, policy- makers, or others in the management or regimes, or other key processes. Exposure, inthings contrast, as temperature, refers to extrinsic precipitation, factors, fire is to identify the scope and objectives of focusing on the character, magnitude, thescientific assessment communities. based on A thecritical intended first step
2 Scanning the Conservation Horizon Executive Summary user, their information needs, and existing decision processes. We also provide guidance on successful approaches for engaging stakeholders. Designing assessments requires attention to several other key considerations, including selection of the appropriate geographic and temporal scales, the features to be assessed (e.g., species or ecosystems), and level of detail and complexity. Given the inherent uncertainties associated with various aspects of climate projections and vulnerability assessments, we provide addressing, and documenting uncertainty. Mike Brake Finally,specific climateguidance change on understanding, is not occurring in a scope and complexity. These examples vacuum, and assessments must be carried include assessments that employ different out in the context of existing stresses analytical approaches (e.g., expert opinion on our species and systems—from the vs. computer models), conservation targets fragmentation and loss of habitat to the on- (e.g., species vs. habitats), and spatial going deluge of invasive species. scales (e.g., states vs. regions) among other variables. Collectively, these case studies Vulnerability assessments can provide a represent many of the leading examples of factual underpinning for differentiating wildlife and ecosystem-oriented climate between species and systems likely to change vulnerability assessments. decline and those likely to thrive, but do not in themselves dictate adaptation There is no single right approach to strategies and management responses. vulnerability assessment that applies Indeed, a continuum of possible adaptation to all situations. Rather, the design and approaches exists ranging from: (1) execution of an assessment must be based building resistance to climate-related stressors as a way of maintaining high- the decision processes into which it will priority species or systems; (2) enhancing feed,on a firm and understandingthe availability of theresources user needs, resilience in order to provide species such as time, money, data, and expertise. and systems with a better chance for Scanning the Conservation Horizon is accommodating and weathering changes; intended to provide resource managers and (3) anticipating and facilitating and conservationists with much-needed ecological transitions guidance for understanding the basic changing environmental conditions. concepts behind vulnerability assessments, that reflect the and for identifying which approaches may To help bring the concepts behind vulnerability assessment alive, the we rise to the challenge of conserving our guide concludes with a series of seven best serve their specific needs as together climate change. fish and wildlife resources in an era of rapid case studies, profiling efforts of varying
Executive Summary 3 Preface “I skate to where the puck is going to be, not where it has been.” – Hockey great, Wayne Gretzky.
reparing for and coping with Provide an overview of the general the effects of a changing principles of climate change vulnerability P climate—known as climate as• it relates to species, habitats, and change adaptation—rapidly is becoming ecosystems the dominant framework for conservation and natural resource management. Describe the various approaches Developing sound adaptation strategies available for assessing the components of requires that managers understand which vulnerability• and address key issues and of the resources they are managing are considerations related to these tools and most likely to be affected, and what options practices may be available to sustain them into the future. Climate change vulnerability Highlight examples of climate change assessments provide an essential tool vulnerability assessment in practice among for informing the development of such government• agencies, non-governmental adaptation plans, and a variety of organizations, academic institutions, and approaches for assessing vulnerability are other stakeholders now in use or are under development. Because the needs and challenges facing Scanning the Conservation Horizon conservation and resource management agencies and organizations are so variable, professionals and other conservation this document offers a framework and is designed to help fish and wildlife practitioners understand how vulnerability general guidelines for assessing climate assessments can help them in responding change vulnerability rather than provide to the challenges of managing natural a step-by-step “cookbook” for conducting resources in an era of rapid climate change. assessments. Similarly, the intent is not Developed by a collaborative working to identify and promote a single “best” group of conservation professionals and approach for assessing vulnerability, but conservation scientists (see below), the rather to help readers understand the document provides guidance for agencies range of approaches available and enable and organizations to consider in developing them to identify the best match for and conducting vulnerability assessments their particular conservation requirements, in support of their conservation and decision processes, and available resources. management missions and as a tool in the Guidance documents, no matter how development of climate change adaptation well written, are no substitute for strategies. The guidance document has in-person training and hands-on three primary objectives: experience, and this guide is designed to support future training sessions to be held on the topic of vulnerability assessment and adaptation planning.
4 Scanning the Conservation Horizon Preface Acknowledgements Wildlife Federation). In addition, we thank Krista Galley (Galley Proofs Editorial Services) for editorial assistance and Maja This guidance document is a product of Smith (MajaDesign, Inc.) for design and an expert workgroup on vulnerability production assistance. assessment convened by the National Wildlife Federation in collaboration This publication has undergone with the U.S. Fish and Wildlife Service. This workgroup includes many of the with U.S. Geological Survey guidelines leading thinkers and practitioners in forscientific Fundamental peer review Science in accordance Practices (SM 502.3). We thank the U.S. Geological from state and federal agencies, non- Survey for coordinating the formal peer- governmentalthis rapidly evolving conservation field and organizations, draws review process for this publication, and and universities. We are grateful to the are grateful to Lawrence Buja (National workgroup members listed on page iii who Center for Atmospheric Research), Harold have given unstintingly of their time Mooney (Stanford University), David and expertise to participate in this Peterson (U.S. Forest Service and University workgroup and collaboratively develop of Washington), and J. Michael Scott (U.S. this guidance document. Geological Survey and University of Idaho) for their careful and insightful reviews. We are also grateful to the following We also appreciate the thoughtful individuals who collaborated with comments offered by U.S. Fish and workgroup members to co-author case Wildlife Service staff. studies: Jennifer Newmark and Kristin Szabo (Nevada Natural Heritage Program), We are especially grateful to the following Jeff Price (World Wildlife Fund), Megan organizations and agencies for providing Friggens and Karen Bagne (U.S. Forest Service), Michael Wilson (Center for Charitable Foundation, Department of Conservation Biology), Patrick McCarthy Defensefinancial Legacy support Resource to this effort: Management Doris Duke (The Nature Conservancy), and Michael Program, National Park Service, U.S. Fish Case (University of Washington). and Wildlife Service, U.S. Forest Service Rocky Mountain Research Station, U.S. We would like to acknowledge the Geological Survey, and National Oceanic following individuals for assisting the and Atmospheric Administration Climate workgroup in various ways: Dan Ashe, Kurt Johnson, and Eleanora Babij (U.S. Fish and Wildlife Service); Arpita Choudhury Program Office, Coastal Service Center, and (Association of Fish and Wildlife Agencies); Finally,Office of we Habitat dedicate Conservation. this publication to Craig Groves and Chris Zganjar (The the memory of Dr. Stephen H. Schneider, Nature Conservancy); Mike Harris (Georgia Department of Natural Resources, knowledge and steadfast commitment Wildlife Resources Division); and John towhose effective lifelong communication pursuit of scientific set the stage Kostyack, Natalie Flynn, Helen Chmura, for the tremendous contribution that the Maggie Germano, Nicole Rousmaniere, science of climate change has made for the Erin Morgan, Austin Kane, Melinda public good. Koslow, and Amanda Staudt (National
Preface 5 I. Introduction
NOAA Our Rapidly Since the release of the Intergovernmental Panel on Climate Change (IPCC) Fourth Changing World Assessment Report (AR4) in 2007 (IPCC 2007a, 2007b, 2007c, 2007d), new apid changes in the earth’s evidence that our planet is experiencing climate* are well underway, and more and larger shifts underscored reasons for concern (Smith, R significant and irreversible changes has are expected, even under the best-case et al. 2009). In the United States, we are scenarios for greenhouse gas emissions seeing a multitude of changes consistent reductions. It is clear from current trends with a rapidly warming climate. Climate and future projections that the planet’s change impacts in the United States living resources—humans, plants, and summarized by the U.S. Global Change animals alike—will exist in an environment Research Program in Global Change in the future that will be vastly different Impacts in the United States (USGCRP from the one we have experienced over 2009, p. 27) include: the past century, during which our conservation traditions evolved.
Lead authors: Bruce A. Stein and Patty Glick. *Terms highlighted in blue are defined in the Glossary.
6 Scanning the Conservation Horizon Introduction U.S. average temperature has risen more Arctic sea ice is declining rapidly and this than 2 degrees Fahrenheit over the past 50 is very likely to continue. years• and is projected to rise more in the • future; how much more depends primarily These changes are already having a on the amount of heat-trapping gases considerable impact on species and natural emitted globally and how sensitive the systems, including changes in the timing climate is to those emissions. of biological events (i.e., phenological changes), such as the onset and end of Precipitation has increased an average of about 5 percent over the past 50 years. shifts in geographic ranges; and changes in Projections• of future precipitation generally communitybreeding seasons, dynamics migration, and populations and flowering; (U.S. indicate that northern areas will become CCSP 2008a). For example: wetter, and southern areas, particularly in the West, will become drier.
The amount of rain falling in the heaviest amphibians,• Across North and America, other wildlife plants are are leafing- downpours has increased approximately 20 breedingout and blooming or migrating earlier; earlier; birds, and butterflies, species percent• on average in the past century, and are shifting or expanding their ranges, this trend is very likely to continue, with often northward and to higher elevations the largest increases in the wettest places. (Parmesan and Galbraith 2004; Kelly and Goulden 2008; Root et al. 2005). Many types of extreme weather events, such as heat waves and regional droughts, Increased water temperatures in coral have• become more frequent and intense reefs in southern Florida, the Caribbean, during the past 40 to 50 years. • contributed to unprecedented The destructive energy of Atlantic bleachingand Pacific and Islands disease have outbreaks hurricanes has increased in recent decades. (Donner et al. 2006; Harvell et The• intensity of these storms is likely to al. 2007). increase in this century. Severe storm events, sea-level rise, and saltwater intrusion hurricanes have become stronger since have• led to a decline in coastal the• In 1980s, the eastern even while Pacific, the the total strongest number of wetland habitats from the storms has decreased. Atlantic Coast to the Gulf of Mexico (Janetos et al. 2008; Sea level has risen along most of the U.S. Kennedy et al. 2002). coast over the last 50 years, and will rise more• in the future. Salmonids throughout the USFWS Cold-season storm tracks are shifting challenged• by global warming– northward and the strongest storms inducedPacific Northwest alteration are of habitat now conditions are• likely to become stronger and more throughout their complex life cycles frequent. (ISAB 2007).
Introduction 7 Forest and grassland systems throughout time frames (e.g., over several decades) the West have been stressed by drought, than we have traditionally considered. • Addressing climate change will also and expansion of invasive species (Ryan et require us to design and implement al.catastrophic 2008). wildfires, insect outbreaks, research and conservation efforts at larger landscape and biogeographical scales, These and other changes often spanning multiple are bellwethers for what Adaptation is rapidly institutional and political jurisdictions (Opdam and scientists project will becoming the primary be even more dramatic Wascher 2004). Further impacts for many species, lens for conservation and complicating matters, habitats, and ecosystems in natural resource planning climate change does the decades to come. Even not occur in a vacuum. with the acknowledgement and management. Indeed, it is the combined that there is considerable effects of climate change uncertainty in climate change projections, and existing problems such as habitat the underlying message is clear: fragmentation that ultimately pose the widespread changes already are occurring, greatest threat to our natural systems and they will continue, they will expand in scope and scale in the next few decades (Root and Schneider 2002). the fish, wildlife, and people they support due to greenhouse gases already in the atmosphere, and they will expand Climate change adaptation is the even more over longer time horizons emerging discipline that focuses on helping if greenhouse gas emissions continue people and natural systems prepare for and unabated or increase. cope with the impacts of climate change (Glick et al. 2009). Indeed, adaptation is rapidly becoming the primary lens for Climate Change conservation and natural resource planning Adaptation—Putting and management.
Vulnerability Until recently the human response to Assessment in Context climate change has focused largely on efforts to reduce the greenhouse gas The potential for far-reaching impacts of emissions that are the underlying driver climate change are driving a fundamental of climate change and global warming. shift in conservation and natural resource Adaptation efforts serve as an essential management. Managers can no longer complement to such climate change look exclusively to the past to guide their “mitigation” efforts. Adaptation, however, conservation and restoration goals, but has only recently begun to be widely instead must anticipate an increasingly acknowledged and embraced as a response different and uncertain future (Milly to the challenges of climate change. As a et al. 2008). We will need to make result, the adaptation science and practice conservation decisions based on longer is still in an early developmental stage and is evolving rapidly (Heller and Zavaleta
8 Scanning the Conservation Horizon Introduction 2009). Additionally, much of the early stressors as a way of thinking and work on adaptation has maintaining high-priority been targeted, understandably, toward species or systems; (2) protecting human communities and enhancing resilience infrastructure from climate impacts, with in order to provide limited attention to date on safeguarding species and systems the natural systems that sustain both with a better chance people and wildlife. for accommodating and weathering changes; Developing meaningful adaptation and (3) anticipating and strategies requires an understanding of, facilitating ecological Susan Stein transitions associated with climate change, and environmental conditions. In the climate second,first, the the impacts, vulnerability risks, and of theuncertainties different change adaptation that reflect literature, the changing resistance components of our natural world to those typically refers to the ability of a system changes. In this context, vulnerability (e.g., and ecosystem, species, population, to climate change refers to the extent to etc.) to withstand a disturbance or change which a species, habitat, or ecosystem is susceptible to harm from climate change structure or function (U.S. CCSP 2008b; impacts (Schneider et al. 2007). More Hellerwithout and significant Zavaleta loss2009; of Nyströmecological et al. vulnerable species and systems are likely to 2008; Williams et al. 2008; Walker et al. experience greater impacts 2004; Easterling et al. 2004; from climate change, while Adaptation refers to Hansen and Biringer 2003). less vulnerable species In other words, the species and systems will be less measures designed to or ecosystem can tolerate reduce the vulnerability or avoid the impacts Accordingly, climate change of systems to the effects of altered air or water affected, or may even benefit. temperatures, extreme as “initiatives and measures of climate change. events, and/or other designedadaptation to can reduce be defined the climate change variables vulnerability of natural systems to actual altogether. Resilience, in an adaptation or expected climate change effects” context, generally refers to the ability of (IPCC 2007d). a system to recover from a disturbance
Key Adaptation Concepts function or structure, and to return to a givenor change ecological without state, significant rather than loss shiftof to a A considerable body of knowledge is now different state (Gunderson 2000). emerging focusing on ecosystem or natural resource-based adaptation (Groves et Coral reefs provide a useful illustration of al. 2010; West et al. 2009; Lawler 2009; these concepts. One of the primary ways Mawdsley et al. 2009; Glick et al. 2009). in which climate change is affecting coral Adaptation efforts generally fall under reefs is through higher average sea surface one or more of the following approaches: temperatures, which is contributing to an (1) building resistance to climate-related increase in the frequency and extent of
Introduction 9 of coral-dominated reefs to algal- dominated reefs in some areas following mass bleaching and mortality is a strong indication of decreased resilience of these systems (Hughes et al. 2003).
While efforts to promote or maintain ecosystem resilience are among the most commonly recommended strategies for climate change adaptation, it will also be important to develop strategies that actually enable or facilitate the ability of a species or ecosystem to change in response to global warming, not just avoid or bounce back from the impacts (Heller and Zavaleta 2009; Galatowitsch et al. 2009).). NOAA In all likelihood, measures to manage for ecological transitions are going to be coral bleaching events around the world (Hoegh-Guldberg et al. 2007). A coral reef conservation agenda. may be able to avoid bleaching and its an increasingly significant part of our associated mortality if, for example, local Although relevant adaptation strategies upwelling draws cooler water to the surface where that reef is located (Grimsditch circumstances, several general adaptation and Salm 2006). Similarly, a coral reef principleswill vary considerably are broadly applicable:based on specific may be resilient to a coral bleaching event if, after experiencing bleaching during a Reduce existing stressors. Climate period of high ocean temperatures, the change will exacerbate many existing coral ecosystem recovers and continues threats to our wildlife and natural to function as a coral-dominated system. • ecosystems, such as the loss of habitat and On the other hand, conditions may be spread of invasive species. Reducing those such that the reef system may not be able existing stressors that interact negatively to withstand or recover from a major with climate change will often be key to bleaching event (e.g., adverse temperature promoting ecosystem resilience. conditions may be prolonged and/or multiple climate and non-climate stressors Manage for ecosystem function. may be at play). Recently, the conversion Healthy and biologically diverse ecosystems will• be better able to withstand or bounce Managing for ecological transitions back from the impacts of climate change. will be an increasingly significant part of our conservation agenda. Protect refugia and improve habitat connectivity. Identifying and •protecting both existing and possible future strongholds of wildlife populations and wildlife corridors will be important for
10 Scanning the Conservation Horizon Introduction helping sustain the full array of species, 1. Identifying which species or systems are likely to be most strongly affected by Ensuring connectivity among these core projected changes habitatecosystems, areas and will their facilitate human the benefits. ability of species to shift ranges in response to 2. Understanding why they are likely to be changing climates. vulnerable
Implement proactive management Determining which resources are most and restoration. Efforts that actively vulnerable enables managers to better set facilitate• the ability of species, habitats, priorities for conservation action, while and ecosystems to accommodate climate understanding why they are vulnerable change—for example, beach nourishment, provides a basis for developing appropriate enhancing marsh accretion, and planting management and conservation responses. climate change-resistant species—may be necessary to protect highly valued species Figure 1.1 offers an overall framework or ecosystems when other options are for adaptation planning, indicating how
and support that process. Elements of insufficient.Vulnerability Assessment: thisvulnerability framework assessments should look can familiar fit into to A Tool for Adaptation many conservationists because it draws Planning from a number of existing conservation planning frameworks, such as The Nature The conservation and resource Conservancy’s Conservation by Design management community is now being (TNC 2006) and the U.S. Fish and Wildlife challenged to take the type of general Service’s Strategic Habitat Conservation principles described above and develop framework (U.S. FWS 2009a). climate change adaptation plans that
Ensuring that these plans are truly “climate-smart”address specific andon-the-ground do not simply needs. represent relabeled business-as-usual will require that managers go through an explicit process for bringing climate data and ecological understanding to bear on their planning.
Climate change vulnerability assessment represents a key tool for providing adaptation planning efforts with such explicit climate input. Vulnerability assessments can provide two essential types of information needed for adaptation planning: Kim Matticks
Introduction 11 Elementidentified 4:and Selected evaluated management based on technical, strategiesfinancial, and can legalthen considerations.be implemented, with the activities and outcomes subject to monitoring in order to feed into a regular cycle of evaluation, correction, and revision. Climate change is not occurring in a vacuum, and the elements of the adaptation planning process must also take existing Kyle Barrett stressors into consideration as well as Element 1: The framework starts with other relevant factors affecting the system. identifying conservation targets, whether they be species, habitats, ecosystems, This guide focuses on how vulnerability or some other unit. Element 2: These assessment (Element 2) can support conservation targets are then assessed conservationists and natural resource for their vulnerability to climate change managers as we move into a future that in order to determine which are likely does not necessarily have past analogs. to be most at risk and which are more For although these assessments must be likely to persist. Element 3: Based on strongly science based, they are not an understanding of why the species or systems are regarded as vulnerable to they must be viewed as an integral part climate change and other stressors, an ofsimply a broader scientific adaptation assessments; planning rather, and array of management options can be implementation framework.
Overarching Conservation Goal(s)
• Species 1. Identify 2. Assess • Sensitivity • Habitats Conservation Vulnerability • Exposure • Ecosystems Target(s) to Climate • Adaptive Capacity Change
Monitor, Review, Revise
• Changes in Policy 4. Implement 3. Identify • Reduce Sensitivity • Changes in Practice Management Management • Reduce Exposure • Institutional Changes Options Options • Increase Adaptive Capacity
Figure 1.1. Framework for Developing Climate Change Adaptation Strategies
12 Scanning the Conservation Horizon Introduction
Box 1.1 “Top-Down” vs. “Bottom-Up” Approaches to Adaptation Planning
The process of developing a climate change adaptation strategy can be approached from either a “top-down” or “bottom-up” perspective, or some combination of these. The most appropriate approach will depend on the scale and goals of your strategy, which in turn will help guide the design of your vulnerability assessment (Hansen and Hoffman 2011). A top-down approach generally starts with looking at one or more scenarios for shifts in climate (e.g., projections for sea-level rise, temperature changes, or extreme rainfall events); assessing what the future land- scape might look like under those scenarios (e.g., what are the plausible ecological effects of the projected physical changes); and finally setting specific conservation objectives and management priorities designed to address those projected future changes. This approach is particularly useful for broad-scale efforts, such as those conducted at regional or national levels, focused on regional ecosystem or biomes, or that have multiple species as conservation targets. A bottom-up approach, on the other hand, usually starts with an organization or agency’s specific con- servation or management goals (e.g., protecting critical habitat for a particular endangered species, managing a specific wildlife refuge, or setting maximum allowable pollutant levels); identifying how climatic variables influence those conservation goals (e.g., the influence of temperature on species’ health and reproduction or on the toxicity of pollutants); determining plausible physical and ecological changes under a range of climate scenarios; and finally identifying and evaluating options for reducing the vulnerability of the agency’s goals to those projected changes.
Why Assess Set Management and Vulnerability? Planning Priorities Vulnerability assessments help resource As described above, vulnerability managers better understand the relative assessments are key tools for the susceptibility of the species, habitats, development of climate change adaptation ecosystems, or special strategies. We would like places they are working to to highlight in particular protect to the likely future three key motivations for impacts of climate change. carrying out vulnerability They help answer two assessments: related questions regarding setting priorities. First, they Help in setting help us identify answers to management and the question: “What should planning priorities • we be doing differently in Mark Karrass light of climate change?” Assist in informing and Just as important, however, they also help crafting adaptation strategies clarify answers to the question: “Which of • our existing activities and management actions continue to make sense in a climate scarce resources change context?” Focusing our conservation • Enable more efficient allocation of
Introduction 13 efforts with an explicit climate perspective 2. Land managers are concerned about will give us a greater chance of success an invasive plant or insect species that in evaluating current conservation and has been spreading across areas to the management objectives to determine if they south of their current location. Model should be adjusted and if so, how, and in simulations project that these species will designing effective approaches for reaching expand into their region due to higher our objectives. In cases where the potential temperatures and increased disturbances impacts of climate change are highly uncertain, managers may initially focus devote additional resources toward halting on so-called “no regrets” strategies, which thefrom spread wildfires. of this They invasive decide before to proactively it arrives in the region. Such efforts may not have or not the projected magnitude of climate been viewed as a priority if those new areas changesprovide conservationactually occur. benefits whether which the particular invasive species mightwere not thrive. identified In other as areas,a viable land habitat managers in of how climate change vulnerability may decide to lessen or abandon efforts assessmentsThe following might are simplified help inform examples conservation plans: suggest climate change may do the job for them—forto fight invasive example, species as models where projectstudies 1. A coastal organization concerned about drier conditions that will no longer preserving an important sea turtle nesting support the invader. site commissions a study that shows that the region is at substantial risk of being Inform and Craft inundated due to rising sea levels. Although Adaptation Strategies there is uncertainty about how much sea-level rise will occur and when at their Vulnerability assessments can also inform site, loss of most or all of the nesting site is the development of effective management considered highly likely. The organization strategies for meeting a conservation can then plan to acquire or secure a long- goal that considers climate change as term easement for land inland of the an added stressor. As will be elaborated current site to provide an additional habitat on later, vulnerability consists of three “buffer” (i.e., protect a greater amount of components—sensitivity, exposure, existing habitat area than is considered and adaptive capacity—and adaptation strategies can be designed either to reduce perhaps accommodate potential habitat the sensitivity and/or exposure of a species migrationsufficient under (i.e., the current transformation conditions) of or “new” or system, or to increase its adaptive areas inland into habitat with suitable capacity. For example: conditions for nesting). Without an understanding of the potential impacts of 1. Climate change may be contributing to sea-level rise, the organization’s resources an increase in average water temperatures might have been spent in other directions, in an important trout stream. Targeted and the option of conserving habitat for a measures to help moderate those new nesting site may have been ultimately temperatures, such as expanding riparian lost to development or other uses.
14 Scanning the Conservation Horizon Introduction Box 1.2. Adaptation and Adaptive Management: Complementary but Distinct Concepts
Adaptation and adaptive management are distinct concepts that are frequently confused with one another. As described earlier, adaptation refers to strategies designed to prepare for and cope with the effects of climate change. Because of the uncertainties associated with predicting the effects of future climates on species and ecosystems, flexible management will almost certainly be a component of well-designed adaptation strategies.
In contrast, adaptive management is one particular approach to management in the face of uncertainty, and is not necessarily tied to climate change. Adaptive management has been described as an iterative learning process producing improved understanding and management over time (Williams et al. 2007). Most portrayals of adaptive management describe a cyclical process in which: management goals are defined based on current understanding and predictive models but with key uncertainties explicitly highlighted; management actions are carried out and monitored, and outcomes are compared to predictions; and refinements are made to goals and actions based on real-time learning and knowledge generation.
While it is a common complaint that current environmental rules and regulations lack the flexibility needed for true adaptive management, the Department of the Interior’s technical guide to adaptive management (Williams et al. 2007) provides both suggestions for and examples of effective adaptive management in the federal context.
Adaptation to climate change is characterized by making decisions in the face of uncertainty. While the adaptive management framework is structured to enable managers to act in the face of uncertainty, other management approaches and philosophies, as discussed in Chapters V and VI, are also designed to address different levels of uncertainty.
To summarize, adaptive management can be an important component of adaptation efforts, but not all adaptive management is climate change adaptation, nor is all climate change adaptation necessarily adaptive management.
vegetation, protecting cold-water refugia, use of proactive measures to assist in the or increasing cold-water spill from existing accretion of sediments as a means for the reservoirs, could become an important marsh to keep up with rising waters. part of trout conservation in the area. Such Chapter VI provides more detail about actions would help reduce that species’ how to use the results of vulnerability exposure to adverse conditions. assessments in the context of developing climate change adaptation strategies. 2. Coastal marshes may be in danger Allocate Scarce Resources conservation action that may not have beenof being considered flooded bywithout rising knowledgesea levels. A of It follows from the aforementioned likely impacts of climate change is the reasons that the results of vulnerability
Introduction 15 2. Managers may decide to spend more of their budget on increased and well- designed monitoring efforts, which will
knowledge gaps and reduce uncertainty aboutbe particularly climate change important impacts to help over fill time. Long-term, appropriately designed monitoring is a critical component of adaptive management, which is likely to play an important role in the development and implementation of climate change adaptation strategies (see Box 1.2).
What Vulnerability Assessments Won’t Do
It is equally important to understand what climate change vulnerability assessments will not do. Although these assessments can provide information about the levels and sources of vulnerability of species or systems to help in setting priorities, the assessments alone do not dictate what those priorities should be. Managers increasingly will be faced with the dilemma
Cheryl Empey of deciding how to invest scarce resources to address various conservation needs. assessments can help wildlife managers Vulnerability assessments can provide a allocate scarce conservation resources factual underpinning for differentiating between species and al. 2007). For example: The choice of whether to systems likely to decline more efficiently (Marsh et focus conservation efforts and those likely to thrive. 1. Vulnerability assessments The choice of whether to may steer managers away on the most vulnerable or focus conservation efforts from potentially costly most viable will be based on the most vulnerable, conservation measures that not only on science, but the most viable, or a may have a low likelihood combination of the also on social, economic, two, will of necessity to climate change, such as and legal values. be based not only on restorationof being efficacious of a particular due habitat type in an area where assessments also social, economic, and legal values. indicate continued habitat suitability is Although uncomfortablescientific to factors, consider, but policy- highly unlikely. makers, managers, and society as a whole increasingly will be called upon to make
16 Scanning the Conservation Horizon Introduction Box 1.3. The Evolution of Climate Change Vulnerability Assessments
Vulnerability assessments have been used for decades in a wide range of sectors to address a wide range of risks. They may target a single risk (e.g., terrorism) or multiple risks (e.g., assessing all sources of vulnerability for an endangered species). The development of climate change vulnerability assessments is part of this ongoing history, adding a new suite of risks for regulators, managers, businesses, and others to consider. Vulnerability to climate change may be investigated in a stand-alone assessment, but in many cases it will be more effective to include it as part of broader vulnerability assessments addressing a range of risks.
As the scientific understanding of the potential and observed impacts of climate change has grown over the past two decades, so too has the interest in developing useful definitions and frameworks for conducting climate change vulnerability assessments (Füssel and Klein 2005). Earlier efforts tended to focus on developing frameworks for assessing the vulnerability of agriculture, public health, and other human systems to climate change, building on approaches used in addressing problems such as poverty, famine, and natural hazards (e.g., Bohle et al. 1994; Handmer et al. 1999; Kelly and Adger 2000; Downing and Patwardhan 2003). More recently, attention also has been placed on assessing the vulnerability of natural systems (species, habitats, and ecosystems) to climate change (Nitschke and Innes 2008; Zhao et al. 2007), as well as multi-disciplinary efforts to assess the vulnerability of ecosystem services to humans (Metzger et al. 2005) and the interactions between multiple stressors (Turner et al. 2003).
Within each of these areas, however, different definitions and concepts for climate change vulnerability have emerged, which often has led to misunderstandings and challenges in assessment efforts (Füssel 2007). In this guide, we followed the general framework adopted by the IPCC (2001a, 2007c), and subsequently by many others, in which vulnerability assessments are founded on evaluations of exposure, sensitivity, and adaptability to climate changes. The information in this guide provides a general framework for assessing vulnerability of natural systems to climate change, drawing from and building on some of the major concepts gleaned from the literature and attained in practice.
determining whether a species ought to has been described as a marriage of art receive protection under the Endangered anddifficult science triage and choices. that will Conservation continue to longhold Species Act (ESA). The types of information true. Making decisions in the face of climate used in climate change vulnerability change will depend on a combination of assessments can, however, provide sound science and practical experience information useful in considering the modulated by societal values. status of a species in relation to the ESA’s requirements. For example, information Climate change vulnerability assessments about vulnerability of species and their will not provide an estimate of extinction habitats to climate change, including risk or provide the sole basis for uncertainty, has been one of the key
Introduction 17 as a candidate of listing (U.S. FWS 2008b); revising critical habitat designated for the
2009b); and determining that the American pika,Quino both checkerspot at the species butterfly and (U.S.subspecies FWS levels, does not warrant listing under the ESA (U.S. FWS 2010).
Finally, there is a permeable boundary between where climate change vulnerability assessments stop and where later components of adaptation planning begin. In this document we focus on the role of vulnerability assessments in providing insights into the relative vulnerabilities of species, habitats, and ecosystems, and understanding the factors involved in those vulnerabilities and other stressors, some of which may be exacerbated by climate change. Adaptation
evaluation, and selection of potential managementplanning also responsesrequires the to identification,address those vulnerabilities. In practice, some vulnerability assessment efforts go to this next level to identify management responses (e.g., Case Study 6), while others do not. This guidance document does not attempt to address detailed techniques and approaches for identifying,
Mary Graham evaluating, and selecting such adaptation responses. However, one increasingly elements considered in several U.S. Fish common technique for taking the and Wildlife Service decisions recently process to the next step is the use of under the ESA. These have included: scenario-based management planning, listing the polar bear under the ESA as a technique for decision-making in the a threatened species (U.S. FWS 2008a); face of high uncertainty, which is discussed identifying the Rio Grande cutthroat trout in Chapter VI.
18 Scanning the Conservation Horizon Vulnerability Assessment Basics II. Vulnerability Assessment Basics
his chapter highlights the measure of whether and how a species or overarching principles of system is likely to be affected by a given T climate change vulnerability change in climate. Exposure is a measure of how much of a change in climate and wildlife management and discussesassessments general in the considerations context of fish and in the design of an assessment, Box 2.1. Key Steps for Assessing Vulnerability to Climate Change Determine objectives and scope determining scope and objectives. Theincluding next chapter the critical (Chapter first step III) of • Identify audience, user requirements, and needed products Engage key internal and external stakeholders provides more detailed guidance • Establish and agree on goals and objectives on how to conduct a vulnerability • Identify suitable assessment targets assessment once those goals and • • Determine appropriate spatial and temporal scales objectives have been established. • Select assessment approach based on targets, user needs, and available resources Box 2.1 summarizes the key Gather relevant data and expertise Although the specifics may vary, steps to carrying out a climate • Review existing literature on assessment targets and climate impacts change vulnerability assessment • Reach out to subject experts on target species or systems as: (1) determining objectives • Obtain or develop climatic projections, focusing on ecologically relevant variables and scope, (2) gathering relevant and suitable spatial and temporal scales data and expertise, (3) assessing • Obtain or develop ecological response projections the various components of Assess components of vulnerability vulnerability, and (4) applying the Evaluate climate sensitivity of assessment targets assessment in adaptation planning • Determine likely exposure of targets to climatic/ecological change and resource management. • • Consider adaptive capacity of targets that can moderate potential impact • Estimate overall vulnerability of targets Components of • Document level of confidence or uncertainty in assessments Vulnerability Apply assessment in adaptation planning • Explore why specific targets are vulnerable to inform possible adaptation responses • Consider how targets might fare under various management and climatic scenarios as a function of the sensitivity of • Share assessment results with stakeholders and decision-makers aThe particular IPCC defines system vulnerability to climate • Use results to advance development of adaptation strategies and plans changes, its exposure to those changes, and its capacity to adapt to those changes (IPCC 2007c). Sensitivity is a
Lead authors: Bruce A. Stein, Patty Glick, and Jennie Hoffman.
Vulnerability Assessment Basics 19 associated problems a species or system physiological or biological variables. For is likely to experience. Adaptive capacity example, a species that is already living at refers to the opportunities that may exist the upper end of its biological temperature to ameliorate the sensitivity or exposure range may not be able to tolerate increases of that species or system. The relationship in the average temperature in its habitat among these three components is outlined due to climate change. That species is schematically in Figure 2.1. Considering therefore considered to be “sensitive” to the degree of change (i.e., exposure) that a at least one element of climate change, species or system is projected to experience higher average temperatures. Conversely, a along with its likely response (i.e., population already living in hot conditions sensitivity) to those changes determines may have adapted evolutionarily to high the potential impact. Understanding the temperatures, and may be less vulnerable likely consequences (i.e., vulnerability), to warming than other populations of that however, requires further consideration species adapted to cooler conditions. of the ability for the species or system to reduce or moderate those potential impacts (i.e., its adaptive capacity). physical or ecological factors. For example, aSensitivity local river also habitat may thatbe a depends factor of onspecific
Exposure Sensitivity sensitivesnowmelt to to projected maintain reductions sufficient instream in average snowpackflows for fish due and to climatewildlife change, is likely as to wellbe as to changes in the timing and intensity of precipitation. Finally, sensitivity to climate Potential Adaptive Impact Capacity by the existence and extent of other human-relatedchange impacts stressors, may be highly such influencedas habitat fragmentation due to roads and other development, which can limit the ability of a species to shift ranges in response to Vulnerability changing climate conditions and associated shifts in habitats or ecosystem processes Figure 2.1. Key components of vulnerability, illustrating important for the life cycle of the species. In the relationship among exposure, sensitivity, and addition, a problem such as unsustainable adaptive capacity. harvest may increase the sensitivity of a species to climate change by reducing the Sensitivity genetic diversity of individuals within that population. Some of these factors may be The sensitivity of a species, habitat, or considered part of the adaptive capacity of a species or system, rather than an element degree to which that system is or is likely of sensitivity (see below). Additional details toecosystem be affected to climateby or responsive change reflects to those the on aspects of sensitivity and methods for changes. Sensitivity may depend on innate assessing it are provided in Chapter III.
fifl�������fi ����fi��fi
20 Scanning the Conservation Horizon Vulnerability Assessment Basics Exposure Box 2.2. A Burning Example of Vulnerability Even if a particular species or Sunburn is an easily grasped (albeit sometimes painful) example of how the system is inherently sensitive to components of vulnerability relate to one another. climate change, its vulnerability also depends on the character, • Sensitivity. Fair-skinned individuals are usually more sensitive to sunburn than magnitude, and rate of changes those with deeper skin tones. This sensitivity has a clear biological basis: the skin to which it is exposed. This pigment melanin absorbs ultraviolet (UV) radiation, which is the primary cause of includes exposure to not only the sunburn. As a result, the skin of individuals with lower melanin levels is innately more physical climate changes (e.g., prone to burning than that of individuals with higher concentrations of melanin. temperature and precipitation) Exposure. but also to related factors such • . Depending on one’s exposure to UV rays, even individuals with high levels of melanin can burn. In this instance, exposure is related to both the strength of the sun’s rays, which varies by latitude, season, and weather conditions, as well as the in vegetation types, increased number of hours in the sun. salinityas altered due fire to regimes,sea-level shifts rise, location of the species or • Adaptive Capacity. A variety of intrinsic and extrinsic means exist for system on the landscape (e.g., ameliorating a person’s likelihood of burning, and therefore reducing vulnerability. latitude and elevation), etc. For Options for reducing exposure to UV radiation range from protective clothing and sunscreen to remaining indoors and out of direct sunlight. A person’s intrinsic of a temperature-sensitive sensitivity to UV rays can also be reduced through graduated exposure to sunlight, speciesexample, may a specific inhabit population an area leading to a temporarily increased concentration of melanin – a process otherwise likely to be sheltered from rapid known as tanning. temperature increases, such as a north-facing, highly vegetated forest or a high-elevation a range of future scenarios. The climate headwater stream (i.e., refugia). In such system can be represented by models of instances, the population may have a lower varying complexity, that is, for any one vulnerability than others of its species component or combination of components given its lower level of exposure. a spectrum or hierarchy of models can be
Use of climate change projections at as the number of spatial dimensions, various scales can help managers get a theidentified. extent toModels which differ physical, in such chemical, aspects sense for where and how much change or biological processes are explicitly might be expected to affect a given represented, or the level at which empirical conservation target. Depending on parameterizations are involved. availability, vulnerability assessments can take advantage of regional climate It is also possible to identify the potential change projections (i.e., changes in ecological effects associated with climate average temperature or precipitation change through the use of so-called projected across an entire region) or more ecological response models, which geographically explicit (but not necessarily provide ways to assess the sensitivity more accurate) data from downscaled and potential adaptability or resilience of climate projections. Both originate from species, habitats, and ecosystems exposed simulations by climate models, driven by to climate change impacts (Wormworth
Vulnerability Assessment Basics 21 and Mallon 2007). There are numerous As mentioned above, some factors could types of response models, ranging from equally well be included as part of adaptive simple to complex. Some of the most capacity, sensitivity, and exposure, commonly used types of response models particularly in the case of species-based are the “habitat and occupancy” models, assessments. However, while there is no which can project changes in habitat hard-and-fast rule about where each of suitability for one or more species over the overall vulnerability assessment, the habitat criteria (e.g., optimal temperature distinctionthese elements may shouldbe useful fit infor as informing part of regimes)large geographic and biophysical areas based attributes on specific that management responses. For example, a a species or community species that is highly can occupy. Other types Adaptive capacity is sensitive to climate include conceptual models, different from specific change but also has a general characterization high adaptive capacity models, expert opinion adaptation measures; it may be considered models, vegetation/ can be considered a less vulnerable than a habitat response models, moderately sensitive physiologically based “pre-existing condition.” species with little or models, and ecological no adaptive capacity. models. Chapter IV provides a more It is important to recognize, as well, detailed discussion of climate and response that the adaptive capacity of a given models and how they may be used in conservation target is different from vulnerability assessments. reduce vulnerability. Essentially, it can be Adaptive Capacity consideredthe specific asadaptation a “pre-existing measures condition” to of that species or system that subsequent The adaptive capacity of a species, habitat, adaptation measures can address. For or ecosystem refers to the ability of that example, some adaptation measures, particular system to accommodate or cope such as removal of seawalls, may serve to with climate change impacts with minimal enhance the adaptive capacity of a coastal disruption. Broadly, adaptive capacity may habitat, thereby reducing its vulnerability be considered a factor of particular internal to sea-level rise. traits, such as the ability of a species to physically move in search of more favorable habitat conditions, adapt evolutionarily, Components of or modify its behavior as climate changes. Biodiversity Adaptive capacity may also be a factor of external conditions such as the existence Devising a useful vulnerability assessment of a structural barrier such as urban areas, not only requires an understanding of the seawalls, or dikes that may limit the ability components of vulnerability, but also the of that species or habitat to move, or components of biodiversity and natural overharvest that limits the genetic diversity systems so that the most appropriate available for evolutionary adaptation. features can serve as targets of the
22 Scanning the Conservation Horizon Vulnerability Assessment Basics assessment. Such targets can include based on regional variations species, habitats, or ecosystems, and several sections of this guidance document mapping efforts. are structured around those biological in ecological classification and Each biological level, in turn, for these biological units, however, is often can be viewed as having three thelevels. subject The definitionsof considerable of and discussion terminology and primary attributes: composition, debate, and terms like “habitat” can have structure, and function (Noss multiple meanings. For that reason, this section provides a brief summary of the forest type can be viewed in various components of biodiversity and terms1990). of As its an composition example, a specific (the Bruce Stein discusses how these concepts and terms different species of plants and animals are used in the context of vulnerability making up and inhabiting the forest), its assessments in this guidance document. structure (e.g., overstory trees, midstory shrubs, understory forbs), and its functions Levels of (e.g., key ecological processes such as Biological Diversity periodic fire or nutrient cycles). The concept of biological diversity—or Distinguishing among these three attributes biodiversity—has become an overarching may seem an abstract exercise, but can be framework for characterizing the full important for distinguishing among the variety of life on earth (Wilson 1992; climate impacts to a particular species or Stein et al. 2000). Although many people habitat type. In a particular forest type, for think of biodiversity in terms of the array instance, shifting climate may eliminate or of species that exist in a particular place, decrease the frequency of certain species, the concept is considerably broader and translating into an change in composition. includes at least three biological levels of Depending upon the affected species, organization—genes, however, that change can species, and ecosystems. Each biological level can also represent a shift in Most vulnerability be viewed as having three ecosystem structure or assessments focus function. Pine rocklands at either species or attributes— composition, in the lower Florida ecosystem levels, structure, and function. Keys, for example, or include some are characterized by combination of the two (although genetic open stands of slash pine with a scrubby factors can come into play in assessing understory of palms and shrubs. In 2005 species vulnerabilities). Terminology and saltwater inundation from hurricane- application is often widely divergent, associated storm surge covered large however, especially for ecologically portions of this habitat on the National Key Deer Refuge on Big Pine Key, causing community, vegetation type, habitat mortality of the overstory pines (Sah et al. type).defined Usage features often (e.g., differs ecosystem, markedly natural 2010). As a result, this portion of the refuge between academic researchers and land has been converted from an open woodland or wildlife managers, and also differs to a scrubland, with consequent affects on wildlife values and ecological functioning.
Vulnerability Assessment Basics 23 Box 2.3. How Does “Resilience” Fit In?
As discussed in Chapter I, one of the most prominent concepts in the field of climate change adaptation today is resilience. A number of factors can determine whether and to what extent a particular species or ecosystem is resilient to climate change. For example, studies show that diversity at multiple levels (i.e., among different functional groups, species within functional groups, and within species and populations of those species, in addition to species richness itself) is particularly critical for ecosystem resilience (Kareiva et al. 2008; Worm et al. 2006; Folke et al. 2004; Luck et al. 2003; Elmqvist et al. 2003). Essentially, such diversity is like climate “insurance”—if one element of a system is compromised, it is more likely that other elements will still be available to support key ecological processes (Peterson et al. 1998). However, while a more resilient ecosystem might be considered less vulnerable to climate change, where and how to incorporate the concept into a vulnerability assessment is not necessarily clear cut (Gallopin 2007). For example, a system that is considered sensitive to climate change, such as a coral reef, may or may not be resilient (e.g., return to a coral-dominant system after a major bleaching event) (Nyström et al. 2000). It is likely that, in most cases, the concept of resilience in a climate change vulnerability assessment will be considered an element of the adaptive capacity of an ecosystem.
Species and Populations than the species as a whole. Common exceptions include assessments mandated by federal statutes such as the ESA. plants often constitute the focus of conservationIndividual species efforts, of fish,and similarlywildlife, andare frequent targets for climate change assessing the individual components of The implication of this is significant in vulnerability assessments. Such vulnerability with respect to a species. assessments can consider a species at the Many aspects of sensitivity relate to innate “full taxon” level, that is, across its entire characteristics of a species, and would be expected to hold relatively constant across portion of the species. The its full range. These geographicrange, or focus subsets on a maygeographically definedIn practice, most might includes factors simply be that portion vulnerability assessments such as reproductive rate or physiological of a species that exists are geographically limited within the area of interest thresholds. On the for the assessment, or and will consider one or other hand, exposure is more populations, rather depending on location. by definition variable (includingmay reflect subpopulations biologically than a species as a whole. Given the same level ordefined metapopulations). populations of innate sensitivity a Most vulnerability assessments are species may be exposed to more change geographically limited in scope (e.g., a state, in some portions of its range than others. region, or place) and will therefore usually For example, a temperature-sensitive consider one or more populations, rather species may be at risk of exceeding its
24 Scanning the Conservation Horizon Vulnerability Assessment Basics temperature threshold in the southern Defining “Habitat” portion of its range but not along the northern range boundary. As a result, Fish and wildlife managers are accustomed its overall vulnerability may differ to thinking about habitat in relation to their work, and many if not most conservation northern populations. It is also possible activities focus on habitat protection, significantly between southern and that adaptive capacity can vary across a management, or restoration. In practice, species geographic range. In this instance, habitat generally refers either to the place genetic variation across the species’ range in which an organism exists, or more may render the plant or animal more or less capable of dealing with climate or provide such things as food, water, and ecosystem variability and perturbations. shelterspecifically, necessary to the tobiophysical sustain an features organism. that In
Habitats and Ecosystems That is, habitat is viewed through the prism ofa strict a particular sense, habitats organism, are constituting species specific. those Terminology related to ecological things that are needed by and used by that levels of biodiversity is complex and particular species. Different organisms contentious and tends may have similar or to provoke interminable In this document, the overlapping habitat discussions and debates term “habitat” should be requirements, but about appropriate these requirements will usage. Among the many interpreted in its virtually always differ terms and concepts most inclusive and either subtly or more involved are: habitat, general sense. conspicuously. natural community, biotic community, biological assemblage, Notwithstanding this organism-centric ecological community, ecological system, view of habitat, the term is perhaps even ecosystem, ecoregion, biome, and more commonly used to describe and landscape. It is not the purpose of this communicate about natural ecosystems and landscapes more generally. In this and distinguish among these various terms andguidance concepts, document and there to attempt are many to definearticles and texts in ecology, wildlife biology, and conservation biology that address aspects of this topic (e.g., Bailey 2009; Jax 2006).
assessment,Because of the though, significance it is important of ecologically to draw adefined few key units distinctions, to the practice as well of as vulnerability to clarify the sense in which key terms are used in this guidance document.
Tom Nebel
Vulnerability Assessment Basics 25 Carl Heilman
sense, the usage can be extremely broad— Despite variability in usage and meaning, referring for instance to natural cover habitat is such a central concept in the practice of conservation—and to key audiences for this guide—that we use the providing some wildlife benefit—or very term extensively throughout this guidance innarrow, terrestrial applying systems to a specificand for terrestrialand precisely document. Unless otherwise noted, in this organismsdefined vegetation most commonly type. Usage is based of the on term document the term should be interpreted a combination of vegetation cover and in its most inclusive and general sense. physical features (e.g., cliff faces, soil types). Habitat-oriented vulnerability assessments In aquatic systems the term commonly can be very powerful tools, but given the is based on physical features such as varied usage and interpretations of this geomorphology, bottom substrate, and term, it is essential that when they are used water current velocity. as targets of assessments the basis for the habitats (both in concept and execution) be
highly variable and give rise to an exceptionalHabitat classifications range of habitat are, not mapping surprisingly, Definingclearly identified “Ecosystem” and documented. efforts based on different attributes and Just as the term habitat has multiple mapping have been standardized in some meanings, so too does the term ecosystem. disciplinesstandards. Habitatand in some classifications states or regions,and In its classical sense, the term refers but not in others. For example, in the to a natural unit consisting of the northeastern United States, the states interaction of living organisms and the have collaborated on the development of physical environment (Odum 1953). This traditional concept of an ecosystem is scaleless in the sense that it can refer to typesa regional that habitatwere the classification focus of individual designed state the interaction among biotic and abiotic wildlifeto cross-walk action the plans state-specific (Gawler et habitatal. 2008).
26 Scanning the Conservation Horizon Vulnerability Assessment Basics elements contained within a tiny water- et al. 2003). Another promising approach from a climate adaptation standpoint is a acre landscape. focus on conserving the ecological “arena” filled depression, or across a million- As noted above, however, there is a host through the use of “land facets”—recurring landscaperather than units specific with biological uniform topographic“actors” that refer to different types of ecological and soil attributes (Beier and Brost 2010; units.of terms Some of varying focus on technical the interactions specificity Anderson and Ferree 2010). that exist among organisms themselves (e.g., biological communities), some on particular classes of organisms (e.g., Setting Goals and vegetation types), while others take a Engaging Stakeholders more geographic or landscape-level perspective (e.g., Greater Yellowstone Climate change vulnerability assessments Ecosystem). It is not our intent to descend into the bottomless pit of debating decision-making, and as such they should the appropriateness of one set of terms beare, designed first and from foremost, the start intended with an to eyesupport over another. In this document, where the toward the needs of the end users, whether term “ecosystem” is used, it can be taken they be on-the-ground managers, policy- to refer in a general sense to ecological makers, or others in the management features or units, and indeed, we often simply refer to “systems.” is so important that or scientific communities. This concept Assessments should be the National Research There are, however, Council (2009) lists several ecosystem-related designed from the start “begin with users’ needs” concepts that have with an eye toward the great applicability for needs of the end users. for effective decision adaptation planning and supportas its first in principle the face of vulnerability assessment. a changing climate. First, there is a wide gradation in spatial scales for different types of units. As an vulnerability assessment is to identify the example, the U.S. National Vegetation scopeA critical and first objectives step then of thein conducting assessment a based on the intended audience and uses of characterizing and mapping vegetation of the assessment. More than anything Classification provides a fine-scale means types in a nationally consistent way based else, the audience and decision process the primarily on vegetation structure and assessment is intended to inform will help composition (Grossman et al. 1998). At a shape the contours of the analysis. somewhat coarser scale, the “Ecological In this section we discuss the importance U.S. Geological Survey’s Gap Analysis of identifying the audience for and Systems” classification that supports Program (GAP) and the U.S. Forest Service’s stakeholders in the assessment, LANDFIRE effort are based on vegetation determining the appropriate level of structure and composition, as well as stakeholder engagement for your particular underlying ecological processes (Comer assessment, clarifying up-front goals
Vulnerability Assessment Basics 27 and objectives, and addressing some key assessment results will be necessary. considerations that logically relate to For example, creating simplistic “bathtub” meeting those objectives within available models of sea-level rise, which are based time and resources. There are a number of primarily on coastal land elevation data, different approaches to assessing climate can be enormously effective in raising change vulnerability, which vary in the awareness of the potential impacts of sea- input requirements and type of outputs. level rise. Such simple models, however, Some approaches are more quantitative are not likely to be particularly informative and other more qualitative, some are modeling-intensive while others rely management actions, since they don’t more on expert knowledge. There is no for targeting specific on-the-ground one single best approach for conducting a processes, such as the effects of tides or vulnerability assessment. Rather, the right sedimenttake into account accretion. important On the other fine-scale hand, approach for any particular effort will while conducting a more sophisticated and depend on user goals and requirements, including the question being asked and the require additional time and resources, it level of resources—data, expertise, time, canfine-scale ultimately analysis produce and assessment a more actionable may and funding—available. places. Similarly, if your target audience Who Is Your Audience? isset a offederal results or for state managers agency developingof specific an adaptation plan that aimed at conserving Execution of a climate change vulnerability a particular endangered species, more assessment should be geared toward complex assessments that consider detailed the particular user (which we refer to biological information about the species as the audience) who will be using the and involve projecting ecosystem-level results. Different audiences will likely changes to its habitat might be the most warrant different assessment targets, valuable approach if resources allow. levels of complexity, and approaches What Are Your Objectives? primary goal of conducting a vulnerability assessmentto communicate is to theraise findings. greater Ifpublic the Clearly establishing the goals and awareness of the threat that climate change objectives is an essential step in designing a successful vulnerability assessment. First, consider relevant mandates, goals, conductposes to afish review and wildlifeof existing at a literature regional and objectives that already exist for your onor nationalclimate change level, it impacts may be sufficientor conduct to organization, agency, refuge, or other such relatively broad and general assessments unit. Particularly for state and federal and then synthesize that information in actors, these may constrain the degree understandable and accessible outreach tools. On the other hand, if the intended to the vulnerability assessment itself. audience is a refuge or park manager who However,of flexibility how they those have goals when and it comesobjectives are described is important from the acquisitions and restoration investments, standpoint of ensuring they are framed will be using the data to target specific land in ways that are clear and meaningful to
then much more fine-scale data and
28 Scanning the Conservation Horizon Vulnerability Assessment Basics those who will conduct the assessment. Although vulnerability assessments can Consequently, the description of the goals feed directly into stand-alone climate and objectives should be a collaborative adaptation planning efforts, there will endeavor that includes the prospective end be other times when this information will need to inform existing agency and involved in carrying out the assessment. organizational planning or decision Allusers too as often, well managersas scientific and and researchers technical staff processes. Indeed, in many instances it will speak in different terms and have different be more important to get climate change expectations and understandings. Time adaptation principles embedded into spent at the beginning of a project to established planning and decision-making ensure that all participants have a common processes, many of which have the force understanding of intended outcomes, of the law. technical requirements, resource needs, and timelines will maximize the likelihood In some cases, the goals of a vulnerability of the assessment helping achieve the assessment may depend on factors such as conservation goals. (See National Research the management jurisdiction or mandate Council [2009] for a detailed discussion of of the agency or agencies conducting the linking information producers and users.) analysis. Many state wildlife agencies, for example, are focused on managing “species Ultimately, the purpose of conducting a of greatest conservation need” (SGCN) as vulnerability assessment in support of adaptation planning is to help increase plans. While they may also be interested in the likelihood that you can achieve your assessingdefined under the vulnerability their state wildlife of habitats action and conservation goals and ecosystems, targeting objectives given the added It will be important to get efforts toward those impacts and complexities species will likely be of climate change in climate change adaptation important to inform conjunction with other principles embedded into the agency’s relevant stressors. The objective adaptation decisions. may be to restore and established planning and Federal agencies are protect populations of decision-making processes. required to utilize a particular species or their programs in group of species. Or, it may be to ensure furtherance of achieving the conservation that a given ecosystem will continue to of species under the ESA. Some agencies support sustainable levels of a natural or organizations may be responsible resource such as timber, or provide certain for managing a particular park or other ecosystem services such as clean water. In protected area, or an area available for use some cases, the goal may be to facilitate a for various purposes of high interest to substantial change in conditions, including changes in habitat and in the composition information about climate change will be of of plant and animal species, so that as greatestthe public—for interest them, and importance. regionally specific much “naturalness” as possible can be maintained. Consider the vulnerability of your goal itself to climate change.
Vulnerability Assessment Basics 29 Regardless of the application and focus, First, consider what you hope to gain coping with climate change will require from stakeholder engagement. Engaging fundamental shifts in the way conservation and informing stakeholders can help to and natural resource management are accomplish the following: carried out. The traditional approach of using past conditions and trends as 1. Provide Data. While there is a large a benchmark and goal for conservation amount of relevant data available in public will become increasingly problematic in a contexts such as on-line databases and the rapidly changing climate. While many of published literature, there are even more our conservation tools and principles data available from less easily accessible will remain the same, it is likely that some of our goals and priorities will need to programs or long-term citizen science change as we look at protecting native projects.sources such Less as formal site-specific or accessible monitoring data sources can be particularly useful for changing environment. understanding local or regional climatic species of fish, wildlife, and plants in a or ecological systems and patterns and
than is available elsewhere. For example, localfor providing observers information and resource at a users finer canscale help to identify which particular climate
annual temperature, etc.) are likely to be mostvariables important (timing for of thefirst ecosystem rainfall, minimum under consideration. The number and type of stakeholders that need to be engaged as providers of climatic or ecological information depends on various factors such as how well characterized is the Gary Tischer/USFWS system being assessed, the quality, size, Why Engage Stakeholders? and availability of existing data sets, and
Engaging the right stakeholders in the the system. right way and at the right times can be the the degree of finer scale variation within critical factor in determining the success of 2. Refine Scope and Focus. To an assessment under some circumstances. maximize the usefulness of a vulnerability We address three important categories for assessment, it is also important to engage decisions about stakeholder engagement in stakeholders in determining the scope or a climate change vulnerability assessment: focus of the assessment. If the goal is to why, who, and how. The goals and context inform resource management over a wide of a particular assessment will, in turn, area involving multiple jurisdictions, for determine the kind and amount of effort instance, you need the input of a broad directed to involve stakeholders. array of resource managers as to how they make decisions—the timing of decision
cycles, the variables they use, etc. Defining
30 Scanning the Conservation Horizon Vulnerability Assessment Basics the scope or focus may happen in two important for developing and implementing stages. A smaller group of individuals the adaptation plan (Vogel et al. 2007). may conduct an exploratory vulnerability They may not need to be full participants assessment that is used to inform decisions in the vulnerability assessment, but they about who needs to be engaged at a may need to know that it is happening broader level. Again, the approach taken and understand how it will feed into the adaptation planning process. This is particularly relevant if one’s goal is to 3.will Provide depend Sociopoliticalon the specific circumstances.Context. support adaptive management plans that Because the sociopolitical setting accompany or are part of climate change adaptation plans or other conservation of natural systems, it is important to plans, since such plans may require broad engageinfluences stakeholders the climate who vulnerability can explain and public support to achieve the needed integrate important components of the sociopolitical system into the level of flexibility. assessment. These components may include national, regional, Box 2.4. Steps to Identify the Appropriate Scope of or local laws, regulations, rules, Stakeholder Participation and plans; important subsistence or cultural uses of the natural • Create an initial list of organizations, interest groups, and individuals who may environment; and value systems wish to be involved in the process or whose buy-in may contribute to project that may determine how human success or failure. systems in the region in question • Meet with representatives of these groups separately in informal settings that are respond to climate change. While familiar to the people with whom you are meeting. ecological and sociopolitical • Explain clearly the principles of vulnerability assessment and adaptation and the elements of vulnerability are often goals of the project with which you are asking them to engage. considered separately, some level • Emphasize the importance of public participation, and that you are asking them of integration is likely to produce to decide among a range of options for engagement, both in terms of the level of more robust and useful results. involvement and the mechanism. Ask group members to express their interests or concerns, and request the selection Local stakeholder engagement is • of a group representative to participate in an initial joint meeting of all the groups. especially important when there • Ask these interested parties if they know of others who should be involved in are ethnographic considerations. the process. Cultural and/or spiritual • Once all interested groups, sectors, and individuals have been approached information is often poorly individually, hold an initial meeting with representatives from all interested groups documented or resides entirely in and sectors to agree on the details of the participation process. Depending on oral histories and traditions. funding and the degree of trust among participants, it may be useful for participants to select a mediator for the stakeholder engagement process, someone 4. Build Support for who is widely respected and viewed as neutral. It may also be necessary to Adaptation. Finally, if the provide some background information or training for stakeholder groups, for goal is to use the vulnerability instance if they will be asked to interpret the results of climate models. assessment for climate change adaptation planning, it is Source: Integrated Resource Planning Committee (1993). worth engaging individuals and organizations that will be
Vulnerability Assessment Basics 31 Whom to Engage? involve simply providing information along the way, while at the other end of the The variety of individuals and organizations spectrum it can involve guiding the entire that may need to be involved is as great as process. It is generally the case that the the variety of reasons to engage them at all. more deeply engaged stakeholders are, the Categories to consider include: more committed they will be to a climate change vulnerability assessment and to Decision-makers (e.g., regulators and using the results in subsequent adaptation managers), in addition to those who may planning and projects. The expected be• requesting or directing scale of the assessment that a vulnerability The more deeply engaged and of the subsequent assessment be conducted stakeholders are, the more adaptation planning will help determine the Decision implementers committed they will be to most desirable level of (e.g., managers) using the results. stakeholders. Engaging • involvement by specific End users of resources/lands (e.g., too many stakeholders or engaging hunters, birders, oil and gas developers) stakeholders too intimately can lead to a • quagmire in which little is accomplished; engaging too few stakeholders or respected individuals within the region or engaging stakeholders too shallowly sector• Opinion of interest) leaders (influential and can lead to inaccurate or incomplete assessments and lack of buy-in for Climate change adaptation planners subsequent adaptation projects.
• Providers of information (e.g., scientists, One important element of engaging holders of traditional knowledge, stakeholders is to be clear with them about sociologists,• etc.; will usually overlap with their role. This may vary depending on other groups) the circumstances and on the stakeholders involved (e.g., in some circumstances you Time allocated to thoughtfully identifying may want the selection of assessment and engaging stakeholders in the targets to be determined entirely or vulnerability assessment will usually largely by stakeholders, while in other be more than worth the effort if the circumstances the selection of targets vulnerability assessment is to be part of may be dictated by the organization a longer-term engagement on climate or agency conducting or commissioning change issues. the assessment).
How to Engage Another important element is to let stakeholders know about any decisions Stakeholders? that already have been made about assessment targets and processes. For The degree of stakeholder engagement example, there may be situations in which in a vulnerability assessment may vary resource managers identify target species widely. At one end of the spectrum, it may
32 Scanning the Conservation Horizon Vulnerability Assessment Basics or habitats for climate change vulnerability Narrow environmental tolerances or assessments due to legal or policy thresholds that are likely to be exceeded considerations, and that while stakeholders under• climate change may be asked for input about additional species to assess, some targets may be set a priori. triggers or cues that are likely to be• Dependence disrupted by on climate specific change environmental Finally, it is important to acknowledge (phenological responses—e.g., rainfall the value of stakeholders’ time and offer or temperature cues for migration, constructive ways to ensure that both you breeding, or hibernation) the assessment process. Dependence on interactions between and they benefit from their engagement in species that are likely to be disrupted Selecting • Inability or poor ability to disperse Assessment Targets quickly or to colonize a new, more suitable• range Species Target species may be selected for a wide array of reasons. Some species may not have any of the biological traits that match conservation work at the state and federal the list above, but an assessment of their levelsGiven thatare focused a significant on individual portion ofspecies the of vulnerability to climate change may be of plants and animals, species are and will interest for other reasons. For example, likely continue to be one of the primary the vulnerability of species that are of targets for climate change vulnerability high economic, social, or cultural value assessments. A wide variety of traits and in an area may be of interest to resource processes can make a species more or less managers, business people, and others vulnerable to climate change. The effects of who want projections to help them gauge a changing climate tend to exacerbate the whether regional populations are likely effects of other threats, such as habitat loss to be sustained or to move elsewhere as a or pressure from invasive species that may result of a changing climate, even though have already made a species susceptible to they are not at risk of becoming extinct. The population declines or even extinction.
examples of climate change vulnerability The World Conservation Union (IUCN) has assessmentsfirst three case targeted studies to in species. Chapter VII are that can make species more vulnerable to Habitats climatedescribed change five categories (Foden et ofal. biological 2008): traits
As described earlier, the term habitat is Specialized habitat or microhabitat used in a variety of ways. Nonetheless, requirements because many wildlife conservation actions • are delivered on the ground based on a habitat framework, using habitats as the
Vulnerability Assessment Basics 33 target of a vulnerability assessment can be climate conditions (Botkin et al. 2007). a helpful way to ensure that the results will It is also possible to apply more complex support the needs of managers. Focusing on models that can simulate habitat responses and project potential changes in ecosystem assessments may occur as an objective structure and function. inspecific and of habitats itself, or as may a target be in for tandem vulnerability with efforts to assess species vulnerability. For aquatic habitats, wider availability of spatially and temporally downscaled Climate change can affect habitats in a climate models have allowed for more number of ways (e.g., it can alter their localized projections on likely changes in species composition, their location and/ temperatures and precipitation to a scale or their size, or their functioning). For relevant for hydrological impact studies, example, areas that are currently managed which can help inform watershed planning as important shrub–steppe habitat may and other management efforts under become more suitable for piñon–juniper climate change (Wood et al. 2004). habitat or may be likely to undergo changes Ecosystems future climate conditions. Further analysis (bothdue to quantitative fire and invasive and qualitative) species under can help The use of ecosystems as the basis for determine how these habitat changes climate change vulnerability assessment might affect associated species, such as will depend largely on the availability of greater sage-grouse, various species of ecological characterization and mapping migratory songbirds, and numerous efforts in the region of concern, and on other animals and plants associated with the way in which ecosystems (or related shrub–steppe habitat. and planning regimes. Some assessments As with species analysis, climate change mayconcepts) focus fitentirely into prevailing on a single management large- vulnerability assessments for habitats landscape “ecosystem,” in which case the can range in levels of complexity. There assessment targets will not actually be the are a number of modeling tools and ecosystem itself, but rather subcomponents resources that can assist habitat managers such as species or particular biological in conducting vulnerability studies. For communities or habitats, or an examination terrestrial systems, scientists frequently of ecological processes. rely on models that can project shifts in the range of vegetation or other organisms due Of particular concern in assessing to changes in climatic variables, usually ecosystem vulnerabilities are the potential at relatively large regional scales. Some of for disruptions in ecological interactions the more basic models project vegetation and compromises to key ecosystem changes under steady-state conditions. functions and processes (Shaver et al. 2000). In turn, impacts to ecosystem distribution of a species to current climate functions can have profound consequences conditions,Specifically, suchthey asrelate temperature the current and for the services that are provided by precipitation, and then project a potential the particular system. The concept of future range under scenarios of future ecosystem services (e.g., water production,
34 Scanning the Conservation Horizon Vulnerability Assessment Basics carbon sequestration) increasingly is such assessments more accessible. For serving as an important framework for example, some dynamic global vegetation human valuation of natural systems models (DGVM) can simulate ecosystem (Millennium Ecosystem Assessment 2005). processes such as carbon dioxide (CO2) Because many ecological assemblages (e.g., the connection between pollinators and (Bachelet et al. 2001). uptake and fluxes in nutrients and water and the insects on which they feed) will Chapter IV provides more detail likelythe flowers be disassembled they fertilize, under or breeding future climate birds about the use of these and other models change as their component species respond in conducting a climate change to changes differently, a combined strategy vulnerability assessment for species, of targeting both species and ecosystems habitats, and ecosystems. may be desirable in many situations (Root and Schneider 2002). Space and Time: Further complicating matters is the fact Selecting the that the ecological impacts of climate Right Scales change do not occur in isolation, but combine with and exacerbate other stresses Setting the appropriate geographic scale on our natural systems. Leading threats to for your vulnerability assessment and biodiversity include habitat destruction, determining over what time scale the alteration of key ecological processes analysis should cover are two key factors in designing a successful assessment. species, and the emergence of new such as fire, the spread of harmful invasive pathogens and diseases. The health and Geographic Extent resilience of many of our species and natural systems are already seriously Climate change vulnerability assessments compromised by these “traditional” can be done at local, regional, and national stressors and changes in climate will have the effect of increasing their impact, relevant assessment targets, a number of often in unpredictable ways. As noted factorsscales. Ascan with determine the identification the spatial of scale the on earlier in this document, some aspects of which you will focus. By its very nature, sensitivity, exposure, and adaptive capacity however, climate change will require that take other stressors into account to we think and plan within the context some degree. For some systems and of larger landscapes, even when our situations (again, depending on users’ management needs are very local. For needs), it may be important to take an example, many species are expected to shift ranges in response to shifting climates, integrates the intersecting effects of all the importantassessment stressors. approach that more specifically An inverse relationship exists between Although assessing the vulnerability of the geographic scale of an assessment ecosystems to climate change is inherently complex, advances in modeling have made and the certainty of projections.
Vulnerability Assessment Basics 35 assessment early can help keep the process Clearly defining the spatial scale of the conducted at the state level, it is important toas considerefficient ashow possible. it will takeIf an into assessment account is species that cross state boundaries, including species that may move into or out of the state or region under future climate conditions. In some cases, conducting a multi-state vulnerability assessment or coordinating with neighboring states can help resolve these problems (see, for example, Case Studies 6 and 7).
Much adaptation planning and implementation will, of necessity, be conducted at the level of individual land management units, whether parks, preserves, military installations, national forests, or other managed landscapes. Ideally, such local-scale planning will be able to draw from vulnerability assessments conducted at broader geographic scales. Nonetheless, some local-scale managers will be interested in conducting their own vulnerability assessments. To the extent possible, these should be structured to build from and J&K Hollingsworth/USFWS take advantage of assessments covering and as a result, our existing portfolio of the state or multi-state region in which the protected areas and wildlife management landscape rests. areas may no longer support the suite of species for which they had originally been Vulnerability assessments for individual established (Hannah et al. 2007). This is protected areas should identify the likely especially true for migratory species, whose effectiveness of those areas to support habitat range may span several states, a given species, habitat, or ecosystem countries, or even continents. Accordingly, under scenarios of climate change. Beyond selecting an appropriate geographic considering the species or habitats that scale for an assessment must consider may be lost from an area, however, they not only the organization’s management should consider what species or habitats jurisdiction, but also the geographic may be likely to move into the area requirements of the species or ecosystems that may be of management interest. In that are the target of the assessment. general, it is important to consider the scale of projections from climate models and the scales desired for projections of
36 Scanning the Conservation Horizon Vulnerability Assessment Basics biological responses, to ensure they match Complexity: More Isn’t appropriately (Wiens and Bachelet 2009). There is often an inverse relationship Always Better between the geographic scale of an assessment and the level of certainty Climate change vulnerability assessments regarding projections of both climate and for species and ecosystems use a range ecological response. Climate projections, of methodologies, from qualitative for instance, are most robust at coarser assessments based on expert knowledge scales, and even with the availability of to highly detailed, quantitative analysis downscaled climate projections, less so using ecological models. Selecting an approach may depend on a host of factors, out vulnerability assessments at local including the availability of already scales,at finer it scales. is particularly As a result, important in carrying to existing information, the level of expertise, understand uncertainties and refrain from time and budget constraints, and so on. For example, while there are a growing number of models available that can Timeoverinterpreting Frame fine-scale projections. project the impacts of climate change on plant and animal ranges, the availability Another key consideration is which to conduct more detailed analyses such as climate change scenarios to use, and over modeling the dynamic ecological responses what time frame. As described in detail in among diverse species within and among Chapter IV, there are multiple scenarios ecosystems is still relatively limited. available based on a range of assumptions, In some cases, focusing quantitative including future emissions trends, levels assessments more broadly on habitat of economic activity, and other factors. changes and then applying qualitative Identifying the potential impacts of climate assessments of potential species responses change under multiple scenarios and may be the best approach given existing time steps (e.g., 10 years, 25 years, 100 information. Additional studies can then years) will be important to inform a range be undertaken as information and of possible management strategies. In resources allow. determining the appropriate time frame for an assessment, consider that near-term projections of climate change scenarios tend to have a higher degree of certainty than those that look farther out. This is the greenhouse gas emissions might change incase the because future, whereasit is difficult the toclimate anticipate change how we experience over the next few decades will be primarily caused by past emissions. However, it may be appropriate for some vulnerability assessments to consider a longer time frame, acknowledging the higher level of uncertainty in long-term Jerry Seagraves climate projections.
Vulnerability Assessment Basics 37 Embracing Uncertainty
Assessing the vulnerability of species, habitats, or ecosystems to most stressors, and certainly to climate change, is complex, and there are different levels of certainty
information and expert knowledge that areand integrated confidence together in each pieceto produce of scientific a vulnerability assessment. Uncertainty is a reality: No one knows exactly how climate may change or how ecological or human systems may respond to change, in any particular location.
Management decisions can proceed in the face of uncertainty. A useful way to characterize uncertainty in the assessment
input or outcome. In some instances we will process is the level of confidence in a given all of the parts determining climate change vulnerability,have a high level and of in confidence other cases in we some may or be less certain in one or more vulnerability factors. It is important to understand the level of certainty about the different components of vulnerability, to identify the range of potential vulnerability given the uncertainties, and to determine what we can and cannot say about the vulnerability of the system. At the same time, lingering uncertainty about climate change need not paralyze us in making decisions and
Brandi Korte developing strategies for adapting to climate change. Chapter V provides a more detailed discussion of the nature of uncertainty, presents a language for addressing certainties and uncertainties, and provides methods for incorporating uncertainty into vulnerability assessments.
38 Scanning the Conservation Horizon Assessing Components of Vulnerability III. Assessing the Components of Vulnerability
his chapter provides a more sensitivity factors that affect all three levels detailed treatment of how of ecological organization, they tend to do T to apply the sensitivity, exposure, so through their effects on individuals or and adaptive capacity framework presented species. Thus, most sensitivity factors are in the previous chapter for conducting a described in the species subsection, below. vulnerability assessment. Depending on the objective of the analysis, whether it Hydrology
ecosystem, or geographic place—different Both terrestrial and aquatic species, componentsfocuses on a singlewill be species, more or specific less useful. habitat, habitats, and ecosystems can be sensitive to Some elements will be useful for assessing changes in hydrology. For example, salmon the vulnerability of a spawning and migration wide range of targets Many of the sensitivity are sensitive to the timing (species, habitats, or and the volume of stream ecosystems). Below, elements that apply across biological levels are and structure of forest flows; the composition “universal” elements, associated with physical stands are sensitive to thatwe first is, aspects describe that these are the availability of relevant to most any systems and processes. ground water; and vulnerability analysis. We dissolved oxygen levels, then provide descriptions of some of the water temperatures, decomposition rates, elements that are better suited to assessing and other wetland attributes are sensitive the vulnerability of particular targets. out of the wetland. Other species are sensitiveto the amount to reductions of water inflowing snowpack, into and Assessing Sensitivity such as the American wolverine, which requires persistent spring snow cover Universal Elements of for its natal dens (Copeland et al. 2010; Sensitivity Aubry et al. 2007).
Many of the critical sensitivity elements Fire that apply across biological levels—that is, to species, habitats, and ecosystems— Individual species as well as habitats and are associated with the earth’s physical ecosystems can be sensitive to changes in systems and processes such as hydrology, For example, plant species that depend the frequency, severity, and extent of fires. fire, and wind. Although there are other
Lead authors: Josh Lawler, Carolyn Enquist, and Evan Girvetz.
Assessing Components of Vulnerability 39 snapping midstem or becoming completely uprooted, as their large taproot and areon fire sensitive for germination (albeit in different and conversely ways) to widespread lateral root system provide species that are intolerant of intense fires them with greater stability. For another which a plant community type is sensitive example, wind can transport dust from changes in fire regimes. The degree to lower-elevation deserts to higher-elevation not only on the individual sensitivities ofto itschanges component in fire species,regimes butwill also depend on (Steltzer et al. 2009). This can, in turn, any synergistic effects that particular leadsnowfields, to early increasing germination the in rate many of snowmelt plant combination of species and their spatial species, leaving them susceptible to frost damage and, in some cases, mortality structure of the vegetation, the ground- (Inouye 2008). arrangement has on fire behavior (e.g., the invasion of habitat by plant species, such Species-Level Sensitivities aslevel cheatgrass accumulation or buffelgrass, of fine fuels, that and can the alter Physiological Factors 1995). Similarly, an ecosystem’s sensitivity a system’s fire regime) (Young and Blank Species-level sensitivities often are of its component species and habitats, characterized by physiological factors butto fire also will by be topography, affected by hydrology, the sensitivities and such as changes in temperature, moisture, potentially by the spatial arrangement CO2 concentrations, pH, or salinity. These of habitat types and the sensitivities of physiological sensitivities can be thought neighboring ecosystems. of as direct sensitivities to climate change. Examples of sensitivities to changes in Wind with maximum temperature tolerances Sensitivities to changes in wind and storm (e.g.,temperature bull trout), are turtlescold-water with fish temperature- species events also may apply across multiple dependent sex ratios, and trees with frost biological levels. For example, individual tolerances or required growing-season tree species will be more or less sensitive to lengths (Dunham et al. 2003; Janzen 1994; wind based on their physiologies. Habitats Luedeling et al. 2009). Examples of direct and ecosystems will be more or less physiological sensitivities to changes in sensitive depending on their component moisture include germination requirements species but also the arrangement and in plants, moisture requirements for composition of those species and potential some amphibians, nest microclimate interactions between wind events, insect requirements for incubation or nestling survival in birds, and snowpack-insulation for example, are considered to be less effects on high-elevation plants and animals sensitiveoutbreaks, to and wind fire. storms Longleaf than pine are loblollytrees, (e.g., the American pika, which exhibits pine and slash pine (McNulty 2002). high temperature sensitivity during the When encountering hurricane-force summer months) (Fay et al. 2009; Blaustein winds, longleaf pine trees are more likely et al. 2010; Rauter et al. 2002; Smith and than these other species to have minimal Weston 1990). Increases in atmospheric damage or be blown over, rather than CO2 concentrations can increase water use
40 Scanning the Conservation Horizon Assessing Components of Vulnerability 2005). There may also be changes in the growth through “CO2 fertilization” (Belote efficiency in plants and therefore increase et al. 2003). Increasing CO2 concentrations CO2-driven increases in water use are also affecting the pH of marine and, interspecific relationships themselves. in some cases, freshwater systems. For competitive relationship among species. example, many corals and other species Likewise,efficiency anhave increase the potential in temperature to change could the with calcareous exoskeletons or shells will be sensitive to changes in pH (Kuffner and Tihansky 2008). competeincrease thewith feeding a cool-water efficiency species. of a warm- water fish species, allowing it to better Dependence on Sensitive Habitats Other key examples of changes in ecological linkages might include changes in the In many cases, individual species’ community of invasive species (e.g., new invaders), changes in the frequency or sensitivity of its habitat to climate change intensity of pest outbreaks, and changes in (McCartysensitivity 2001). is likely For to example, be influenced species by thatthe the prevalence, spread, and susceptibility breed in vernal pools, ephemeral wetlands, to disease. For example, increases in intermittent streams, and species that live temperature have allowed mountain in alpine environments or low-lying coastal pine beetles to complete their life cycles zones are all likely to be highly susceptible within a single year at higher elevations, to climate impacts such as rising exposing previously isolated whitebark temperature regimes; winter precipitation pine to beetle outbreaks (Logan et al. 2003; arriving more frequently as rain than snow; Logan and Powell 2001). Similarly, changes shifts in the timing of snowmelt, runoff, and in climate have been implicated in the range shift of the disease-causing chytrid fungus, potentially affecting previously speciespeak flows that in are streams; linked andto highly sea-level sensitive rise. isolated amphibian populations and species habitatsIn some cases,such as it thosemay suffice listed above.to highlight In other (Pounds et al. 2006; Bosch et al. 2007). cases, one might want to conduct a detailed habitat sensitivity assessment (based on Phenological Changes some of the factors listed below in the Phenology refers to recurring plant and
Ecologicalsection on habitat-specific Linkages sensitivities). agriculturalanimal life-cycle crops, stages, emergence such as of leafing insects, Species’ sensitivities also likely depend on and migrationflowering ofof plants,birds. Many maturation of these of the effects of climate change on predators, events are sensitive to climatic variation competitors, prey, forage, host plants, and change (Parmesan and Yohe 2003), diseases, parasites, and other groups and when the timing and sequence of of species that affect the focal species these events are altered, loss of species (Parmesan 2006). For example, there may and certain ecological functions (e.g., be changes in the occurrence or abundance pollination) can occur (Root et al. 2003; of predators or prey that subsequently Visser and Both 2005). For example, affect the focal species (Visser and Both climate change can reduce the amount of
Assessing Components of Vulnerability 41 food available to birds during migration Degree of Specialization and the breeding season (Both et al. 2006; Leech and Crick 2007). It can also alter Generalist species are likely to be less fundamental interactions between species sensitive to climate change than are that affect competition and survivorship specialists (Brown 1995). Species that (Root and Hughes 2005). However, the use multiple habitats, for example, have scale of response can depend on the life multiple prey or forage species, or have history of the individual species. For multiple host plants are likely to be less example, sea birds with high dispersal sensitive to climate change than are species capacity have been shown to respond to with very narrow habitat needs, single- broader-scale cues (e.g., the North Atlantic forage or -prey species, or single-host-plant Oscillation) whereas permanent residents species (Thuiller et al. 2005). In addition, tend to response to local-scale cues (e.g., sea surface temperature) (Frederikson et evolutionary factors (e.g., low genetic al. 2004). Similarly, the timing of spring variationspecialist forspecies heat areresistance) likely to thathave limit specific migration by long-distance migrants is their ability to adapt to changing conditions more attuned to regional- to continental- over time (Kellermann et al. 2009). scale climates while timing of spring migration by short-distance migrants Reproductive Strategy responds much more strongly to the local climate (MacMynowski and Root 2007). The reproductive strategy of species may Information on phenology can be found at also make them sensitive to climate change. http://www.usanpn.org/. Some studies suggest that species with long generation times and fewer offspring Population Growth Rates (e.g., “K-selected” species) are likely to be at greater risk of extinction under long- Species that can quickly recover from low term climate change than those whose life population numbers are more likely to be history is characterized by short generation able to withstand rapidly changing climates times and many offspring (e.g., “r-selected” as well as colonize new locations following species) (Isaac 2009; Chiba 1998). For climate disruption (Kinnison and Hairston example, more opportunistic, rapidly 2007). In addition, rapid population growth reproducing species may be better able to can also help maintain genetic variability. take advantage of major climate change– This trait favors colonization of extant and novel habitats by early-successional and hurricanes. invasive species (Cole 2010). As a result, related disturbances such as wildfires and conservation practitioners and managers Interactions with Other Stressors will be challenged by deciding which species have conservation value as species The existence of other stressors has reassemble and populations grow. the potential to exacerbate the effects of climate change on individuals and populations. For example, research suggests that exposure to pollutants such as heavy metals, oil, and pesticides may
42 Scanning the Conservation Horizon Assessing Components of Vulnerability act synergistically with ocean warming example, research suggests that restoring to induce coral bleaching events in some heterogeneity in vegetation structure, instances (Brown 2000). Studies have also composition, density, and biomass to found that the existence of pollution can rangelands such as the tallgrass prairie of the Great Plains is likely to be an important corals after major bleaching events and strategy to improve the resilience of these diseasesignificantly outbreaks reduce (Carilli the recovery et al. 2009). rate of systems to climate change (Fuhlendorf and Engle 2001). Increased homogeneity in Habitat-Level Sensitivities some rangeland systems due to livestock production has made these systems much Sensitivity of Component Species more vulnerable to disturbances such as
The sensitivity of a given habitat type will from the more patchwork-type burn largely be determined by the sensitivities patternswidespread and wildfires, associated which grazing are patternsdifferent of of its component species. The sensitivity a more diverse tallgrass prairie system. of dominant species, ecosystem engineers, keystone species, and “strong interactors” Ecosystem-Level Sensitivities sensitivity of a habitat type. are likely to have large influences on the Sensitivity of Component Species Community Structure The sensitivity of a given ecosystem Plant and animal communities depend will largely be determined by the in part on a delicate balance of multiple, sensitivities of the ecological function and biological diversity of that system, as communities will be more sensitive to well as the sensitivities of the component climateinterspecific change interactions. than others. Some For example, species/habitats. the presence of algae-grazing species of As with habitats, the sensitivities of overgrowth of harmful, opportunistic dominant, keystone, and indicator species algaefish and on invertebratescoral reef habitats can help damaged limit the by coral bleaching, facilitating their ability sensitivity of an ecosystem. For example, to recover (Nyström et al. 2000). Coral theare likelyochre tosea have star large is a keystone influences species on the in reefs in regions where problems such as Northwest, in that it maintains community of algae-grazing species are therefore diversityrocky intertidal through ecosystems predation ofon the mussels. Pacific likelyoverfishing to be morehave reducedsensitive the to climatepopulation Research has found that such predation change than those with such functional is sharply reduced by decreases in water communities intact. temperature (such as during the seasonal upwelling of cold, nutrient-rich water from The level of diversity of component species and functional groups in a habitat is sensitive to climate, and its frequency also may affect the sensitivity of that andthe ocean intensity floor) may (Sanford be reduced 2002). by Upwellingfuture habitat to climate change impacts. For climate change. As a result, predation
Assessing Components of Vulnerability 43 by the ochre sea star may become more organization. This section provides a regular, transforming the community general overview of the various elements of dynamics by reducing mussel populations. exposure. The following chapter (Chapter IV) offers more detailed information about Sensitivity of Ecosystem Processes to Temperature or Precipitation in climate change vulnerability assessment. specific tools available to evaluate exposure Many ecosystem processes, such as Historic versus Future decomposition, nutrient transport, Projected Change
changes in temperature or precipitation. Vulnerability assessment can be conducted sedimentation, fire, etc., are sensitive to either based on historic observed changes temperatures, for instance, are likely to in climate (retrospective assessment), haveChanges an impact in river on flow eutrophication and water and future modeled projections (prospective oxygen depletion in estuarine systems such assessment), or a combination of the two. as the Chesapeake Bay. If climate change Historic changes will generally indicate the leads to an increase in the frequency current vulnerability as compared with the or extent of heavy downpours, as some past, while the future climate projections will give an assessment of future greater amounts of nutrients and other vulnerability. Depending on the objectives pollutantsmodels suggest, into coastal increased waters runoff (Hagy will et flush al. of the assessment, one or the other may 2004). Heavy runoff also decreases water be more appropriate. If the resources and mixing as less dense fresh water rides over data are available, a combination of both the top of denser salt water, inhibiting the retrospective and prospective assessments mixing of water and the replenishment provides the most complete picture in of oxygen in deep waters. Higher water terms of the current status and the likely temperatures also may affect oxygen levels future status. in some systems because warm water holds less dissolved oxygen than cool water Basic Climate (Najjar et al. 2000). Further, higher water temperatures can accelerate the bacterial The most basic and direct types of exposure decay of organic matter present in the are from changes in climate: temperature, water, thereby consuming oxygen and precipitation, wind, humidity, cloud exacerbating hypoxia (Varekamp cover, and solar radiation. Although these et al. 2004). variables often increase vulnerability indirectly—such as by changing hydrology,
Assessing Exposure (e.g., competitors, predators, prey, etc.)— theyfire, orcan distribution also directly of increaseinteracting vulnerability. species Universal Elements The change in the mean values of these of Exposure basic climate variables can be used in vulnerability analyses (e.g., changes in Most of the following exposure elements average annual temperature or total annual apply to all three levels of ecological precipitation). However, it may be changes
44 Scanning the Conservation Horizon Assessing Components of Vulnerability to the extreme values of these variables Program (http://drought.unl.edu/DM/ (e.g., daily minimum or daily maximum MONITOR.html) provides current and temperatures) that are most important recent historic maps of drought severity for determining vulnerability. The ability in the United States. The National Oceanic for plant species to exist in a certain area and Atmospheric Administration provides is often related to the temperature of the current drought maps at: http://lwf.ncdc. coldest day of the year in that location. noaa.gov/oa/climate/research/prelim/ drought/palmer.html. These basic climate variables can be measured for different time periods— Hydrologic Changes or even within a day (e.g., nocturnal/ There are many hydrologic changes diurnalannually, extremes). seasonally, Understanding within specific months, that may occur to both terrestrial and which of these time periods and climate aquatic systems in a changing climate. For measures are biologically relevant is key terrestrial systems, the types of hydrologic for determining climate vulnerability. exposure will generally relate to the For example, changes to springtime amount of available soil moisture through temperatures may be the most important changes in precipitation, water runoff, factor determining climate sensitivity for and evapotranspiration (ET). In general, organisms or ecosystems dependent on increasing temperatures will result in increased rates of ET causing decreases in soil moisture. Increases in precipitation wintertimethe timing of temperature specific spring changes events might such can offset this increase in ET, but it must beas floweringmost important or hatching. for a species In contrast, with chilling requirements for seed production right time of year. If temperature increases (Luedeling et al. 2009). andcome precipitation in sufficient doesamounts not change and at theor decreases, it can be fairly safely assumed Drought
Changes in temperature and precipitation severity. In general, it is thought that under climatecan influence change drought there will frequency be an increase and/or in the incidence, intensity, and duration of droughts, but this will certainly differ by location. There are drought indices available for quantifying the exposure to drought (Trenberth et al. 2003). Two of the most commonly used indices are the Palmer Drought Severity Index (PDSI) and the Standardized Precipitation Index (SPI). For a good overview of drought indices see: http://drought.unl.edu/whatis/ indices.htm. The U.S. Drought Monitor USFWS
Assessing Components of Vulnerability 45 that soil moisture will decrease. There are Changes in Fire Regimes various methods of calculating changes to ET. The simplest methods are based only Climate change is expected to contribute on temperature and number of daylight hours (e.g., the Hamon method) (Hamon in some regions, including shifts in 1961). The assessment approach used in theto significant timing, intensity, changes and in fire frequency regimes of the Four Corners case study (Case Study 6) included application of the Hamon method. More complex methods for computing ET inwildfire western events forests (Flannigan have become et al. 2000).more For need information about temperature, wind frequentexample, andresearch larger shows since thatthe mid-1980s, wildfires speed, relative humidity, and solar radiation a trend that corresponds with warmer (e.g., Penman-Monteith) (Allen et al. 1998). springs and an expansion of summer dry Once ET is estimated, various related periods (Westerling et al. 2006). Studies metrics can be calculated by subtracting project that the overall acreage burned it from ET from precipitation: actual ET, could double in size across parts of the west by mid- to late century as average temperatures continue to rise (Spracklen et Macroscalemoisture deficit, hydrologic and moisture models providesurplus. an al. 2009; McKenzie et al. 2004). even more complex and generally more
accurate but computationally intensive Changes in CO2 Concentrations method for estimating hydrologic
responses to climate change. One such There is no doubt that atmospheric CO2 concentrations have already increased (VIC) model developed by researchers from approximately 280 parts per million atmodel the University is the Variable of Washington Infiltration (http:// Capacity (ppm) in the recent historic past to around www.hydro.washington.edu/Lettenmaier/ 385 ppm today. Future atmospheric CO2 Models/VIC/). These models explicitly concentrations could range from 500 ppm track the movement of water through to close to 1000 ppm based on emissions the landscape, and so are generally more scenarios used in the IPCC AR4 (2007b). accurate and complete than the other The concentrations will continue to rise methods for estimating ET. In addition to through at least the late part of this century, the metrics mentioned above, these models and depending on the emissions scenario can also model other hydrologic changes, considered, concentrations may continue such as snowpack depth and water runoff. to increase beyond the end of the century. The changes in water runoff can be used to These changes in CO2 concentrations can have physiological effects on plant species. For example, increases in atmospheric estimate changes in river flow. Depending change in different ways, but with warming CO2 concentrations can result in increased temperatureon how climate there changes, are some river changes flow may that
are more likely to occur, such as earlier an atmosphere with enriched CO2, these spring runoff and lower summertime base waterplants usemay efficiency be able to in grow some in plants. drier climates In than they currently occupy. Changes
flows in snow-fed rivers. higher-level ecological impacts such as in water use efficiency can also lead to
46 Scanning the Conservation Horizon Assessing Components of Vulnerability changes in species competitive interactions and 50 degrees north (Stott et al. 2008), and changes to community composition while increasing water runoff from melting (Cramer et al. 2001). glaciers and polar ice caps are causing a decrease in salinity in oceans near the poles Changes in Vegetation (Curry et al. 2003). For a good overview of ocean salinity see: http://nasascience.nasa. Changes in climate have the potential to gov/earth-science/oceanography/physical- alter the distribution of plant species and ocean/salinity. Data and maps on changes hence alter plant associations and plant to salinity can be found at: http://aquarius. communities. Dynamic global vegetation jpl.nasa.gov/AQUARIUS/index.jsp. models and other less complex models can be used to project how “plant functional Changes in pH types” such as conifers, broad-leafed deciduous trees, and grasses are likely to As the oceans absorb atmospheric CO2, they change in the future (Bachelet et al. become more acidic. If CO2 concentrations 2001; Cramer et al. 2001; Sitch et al. in the atmosphere continue to increase 2003). Projected shifts in vegetation at the current rate, then the oceans will types or biomes can provide an idea of become relatively more acidic (i.e., will have a lower pH) than they have been in millions might change. of years (Caldeira and Wickett 2003). This how much specific plant community types lower pH will erode the basic mineral Changes in Species Distributions building blocks for the shells and skeletons of calcareous, reef-building organisms Species distributions will shift as climate changes. In some cases, it will be useful to number of important microorganisms that aresuch a asfoundation shellfish andfor the corals, marine as well food as web a move in response to climate change. For (Kuffner and Tihansky 2008; Orr et al. example,understand maps how of a projected specific species range shiftsmight 2005). for invasive species, keystone species or ecosystem engineers, or predators, Changes in Storm Frequency competitors, or diseases of a focal species and Intensity may serve as useful exposure elements for individual species, habitats, or ecosystems. In general, the frequency and magnitude Species distribution modeling can be used of intense storms is projected to increase. to project how species’ ranges will shift due This is at least in part due to increased to the many different factors affected by temperatures causing greater evaporation. climate change (Lawler et al. 2006, 2009). For example, modeling studies have projected an increase in tropical cyclone Changes in Salinity (hurricane) intensity, and there is evidence that the number of Category 4 and 5 Climate change is altering salinity hurricanes has increased over the past 30 concentrations in the world’s oceans. In the years (Trenberth 2007; Webster et al. 2005; Atlantic Ocean, there has been an increase Emanuel 2005). in salinity observed between latitude 20
Assessing Components of Vulnerability 47 As mentioned earlier in this report, it is important to note that some of these examples may be considered as factors that contribute to a species or system’s sensitivity to climate change, rather than as adaptive capacity.
Species-Level Adaptive Capacity
Richard Maack/Fountain Studio Plasticity
Assessing The ability for a species to modify its Adaptive Capacity physiology or behavior to synchronize with changing environmental conditions or coexist with different competitors, predators, and food sources (a potential, capability, or ability of a system characteristic called plasticity) can be toThe adjust IPCC todefines climate adaptive change, capacity to moderate as “the considered a factor of adaptive capacity potential damages, to take advantage (Running and Mills 2009; Nylin and of opportunities, or to cope with the Gotthard 1998; Gotthard and Nylin consequences” (IPCC 2007c). In the context 1995). In general, plasticity increases the of assessing the vulnerability of human likelihood that a species will be able to communities, adaptive capacity often refers respond effectively to both climate change to the potential to implement planned itself and to effects of climate change, such adaptation measures to cope with change, as phenological mismatch (Parmesan 2005; including factors such as economic wealth, Parmesan and Galbraith 2004; Parmesan institutional capacity, and equity (Metzger et al. 1999). There is evidence that recent et al. 2005). For natural systems, adaptive climate change has already elicited these capacity is often considered to be an types of adaptive responses across a wide intrinsic trait that may include evolutionary range of plant and animal species (Walther changes as well as “plastic” ecological, et al. 2002). Over time, it is possible behavioral, or physiological responses that these traits may become a genetic, (Williams et al. 2008). This is not to say, evolutionary component (see below). however, that only the intrinsic factors of adaptive capacity are relevant in assessing Dispersal Abilities the vulnerability of species, habitats, or ecosystems to climate change. Certainly, Dispersal refers to the movement of a there are likely to be a number of external species away from an existing, typically factors (both natural and anthropogenic) natal, population (Fahrig 2007). Some species may be able to disperse over long or system to adjust to or cope with climate distances (e.g., seeds may be carried to changethat will (see influence Box 3.1). the This ability section of a species provides different areas by birds or other hosts). some examples of adaptive capacity within both of these contexts.
48 Scanning the Conservation Horizon Assessing Components of Vulnerability Other species, such as those that have Maintaining the evolutionary potential for evolved in patchy or rare habitats, may species to adapt will be key in designing have lower dispersal ability. In general, climate change adaptation strategies. species that are poorer dispersers may be Among the best approaches for retaining more susceptible to climate change as they this potential is ensuring that protected will be less able to move from areas that area networks harbor a well-distributed climate change renders unsuitable and into representation of species found in a region. areas that become newly suitable. Berg et al. (2010) reviewed dispersal distances Habitat-Level across broad taxonomic groups, noting Adaptive Capacity especially that below-ground organisms tend to have an extremely limited ability to Permeability of the Landscape disperse. Barriers to dispersal may increase the vulnerability of some species with high The degree to which species, propagules, innate dispersal ability. and processes can move through the landscape will affect the sensitivity of Evolutionary Potential species, habitats, and ecosystems to climate change. More permeable landscapes Some species and some populations will with fewer barriers to dispersal and/ be better able to adapt (evolutionarily) or seasonal migration will likely result to climate change. Relevant traits include in greater adaptive capacity for species, generation time, genetic diversity, and habitats, and ecosystems. However, the population size (Skelly et al. 2007; degree to which a landscape is permeable Bradshaw and Holzapfel 2006). For example, species with shorter generation times, in general, have faster evolutionary rates than species with longer generation times, and may be able to evolve behavioral or physiological traits that allow them to withstand climatic changes more rapidly than will than long-lived species with long generation times. Likewise, populations with high genetic diversity for traits related to climate tolerance are more likely to contain individuals with heritable traits that increase the tolerance of the species to climate change. Several recent studies have already discovered heritable, genetic changes in populations of some animals, including the Yukon red squirrel (Réale, et USFWS al. 2003), the European blackcap (Bearhop et al. 2005), and the great tit (Nussey et al. 2005), in response to climate change, most often associated with adaptation to the timing of seasonal events or season length.
Assessing Components of Vulnerability 49 depends on the process or organism that Box 3.1. Assessing Adaptive Capacity: Insights is being considered, and thus a permeable from a Coastal System landscape for one species may not be very permeable for another species. The relative Both natural and anthropogenic factors can affect the adaptive permeability of a landscape may depend capacity of a system, as illustrated in the case of coastal vulnerability on both natural and anthropogenic factors. to sea-level rise (Klein and Nicholls 1999). The impacts of sea-level In particular, fragmentation of habitat rise on a coastal system depends on the global rate of eustatic sea- due to urban development, agriculture, level rise, which refers to the change in volume of the oceans due dams, and other human activities is likely to be an important factor in reducing the to thermal expansion and the addition of water from land-based ice adaptive capacity of some otherwise highly melt, as well as localized factors that affect the relative amount of dispersible and migratory species (Vos et sea-level rise in a particular area. Relative sea level rise is affected al. 2002). by such variables as rates of geological uplift and deposition of sediments: marsh sediment accretion, for instance, can lessen Ecosystem-Level the amount of localized sea-level rise, while land subsidence can Adaptive Capacity exacerbate the problem. In deltaic systems, for example, the release of river sediments downstream can help habitats such as coastal Redundancy and Response Diversity wetlands keep pace with sea-level rise (Reed 2002; Morris et al. within Functional Groups 2002). Similarly, coastal habitats such as wetlands and beaches might be able to occupy new areas farther inland as rising sea levels Within any community, there is a range of inundate or erode those habitats along the shore. Essentially, these functional groups present. In ecological communities, this includes groups such as variables can be considered elements of the adaptive capacity of a primary producers, herbivores, carnivores, coastal ecosystem. and decomposers. In systems where each functional group is represented by multiple A number of factors can either enhance or reduce this adaptive species and the response to any given capacity. For example, altered river flows (due to climate change, upstream water uses, and/or other stressors) or the existence of among the species that make up the environmental change varies significantly dams or levees can reduce or eliminate the amount of sediments functional group, system resilience to that reach the coast, contributing to a higher rate of relative environmental change is likely to be higher (Nystrom et al. 2008; Naeem 1998; Petchey sea-level rise. Similarly, the existence of upland barriers, either and Gaston 2009). In other words, if a natural (e.g., rocky cliffs) or anthropogenic (e.g., seawalls), can particular species or decomposer responds limit or prevent the ability of coastal habitats to migrate inland. negatively to a climate change but others Ultimately, understanding the multiple factors that can affect the respond positively, decomposition function adaptive capacity of a coastal ecosystem can help inform relevant within the system may not be disrupted. management decisions, such as finding ways to restore the deposition of sediments or removing coastal barriers.
50 Scanning the Conservation Horizon Climate and Ecological Models IV. Peering into the Future: Climate and Ecological Models
ssessing the exposure of species be involved in running these models: to climate change requires the rather, it is to ensure that assessors are A ability to peer into the future knowledgeable about the range of models and identify likely or potential changes available, and can be well-informed in ecologically relevant variables. These consumers, understanding the basis for variables can be both direct climatic and assumptions underlying widely used factors, such as changes in temperature models. In particular, most vulnerability or precipitation, or indirect factors, assessments will not involve running such as shifts in ecosystem processes sophisticated and complex global climate or interactions with other species. change models, but will instead rely Models provide an important means for on existing scenarios and make use of forecasting possible future conditions. available downscaled climate projections. A model constitutes a representation of Assessments more often will rely on a system, which enables researchers to application of ecological response models, investigate and understand the properties although even those models may be of that system. Depending on their design, supplanted or bolstered by existing studies models can also be used to simulate future conditions and outcomes. Although models than modeling (e.g., expert elicitation). can be powerful, they also have limitations. in the scientific literature or by means other The statistician George Box famously has been quoted as saying “all models are Climate Models wrong, but some are useful” (Box and Increasing emissions of CO , methane, and Draper 1987). Considerable progress has 2 other heat-trapping greenhouse gases are been made over the past few decades, perturbing average climate conditions at however, in developing robust and useful local to global scales in ways that cannot be models for understanding both the earth’s predicted by the past. Instead, projections climate systems, as well as the ecological of future climate conditions rely on climate responses to climate. model simulations driven by assumptions about how population, energy use, and This chapter provides an overview technology are likely to develop in the of the types of climate and ecological future. These assumptions are collectively response models that are relevant to known as emission scenarios, as they serve vulnerability assessments of species, as the basis to estimate the emissions of habitats, or ecosystems. The purpose of greenhouse gases, particulates, and other reviewing these models is not to suggest pollutants that would result. that all vulnerability assessments will
Lead authors: Katharine Hayhoe, Bruce Jones, and John Gross.
Climate and Ecological Models 51 Emission scenarios are then used as important to understand the limitations of inputs to global climate models in order these data as well. Downscaled projections to simulate the changes in temperature, are uncertain for the same reasons as precipitation, and other aspects of climate global projections: the range in plausible likely to result from that set of assumptions. future scenarios; the sensitivity of the Global climate models represent climate at climate system to those emissions; our a relatively coarse resolution and they do imperfect understanding of and ability not resolve differences in climate variables to model the climate system; the natural variability of the climate system; and the kilometers. The “basic” global climate degree to which the simulations are able modelsat scales are finer the than General several Circulation hundred Models to capture the relationship between local (GCM), which are mathematical models of climate and large-scale drivers. a planetary atmosphere or ocean based on Due to the uncertainty inherent in and chemistry. At the most comprehensive future projections, multiple future endgiven of equations the spectrum for physics, are the coupledfluid motion, scenarios should be considered in impact Atmosphere–Ocean General Circulation assessments. In an area where precipitation Models (AOGCM), which address additional trends are highly uncertain, for example, factors such as models for sea ice or a state might choose to consider two evapotranspiration over land. scenarios: one warmer and wetter, and one warmer and drier. Adaptation strategies that are robust to multiple likely climate change scenarios would be considered “no regrets” strategies. In some instances, expert opinion can be useful for assessing the likelihood of how future local climates
habitats will change. Expert opinion shouldwill reflect be accompanied regional projections by an estimate or how of certainty and description of the assumptions, evidence, or reasoning underlying the opinion.
Chuck Young/USFWS Historical and Future Output from global climate model Scenarios simulations can be used to calculate Control Scenarios incorporate directly into planning efforts. Forregional this reason, trends, abut range these of aredownscaling difficult to Climate models are not rigidly controlled techniques have been developed. Although by boundary conditions, as are response downscaled climate projections often models. Rather, they generate their own provide the spatial and temporal resolution internal natural variability. Before they can needed to assess the impacts of climate be run in transient mode (i.e., generating change on a given region or system, it is information for real calendar years),
52 Scanning the Conservation Horizon Climate and Ecological Models climate model simulations begin with Collectively, these historical total (human a long, time-independent control run. and natural) forcing scenarios are known External forcing mechanisms (e.g., the as the “20th Century Climate in Coupled strength of the sun and the concentrations Models” or 20C3M scenarios in the IPCC of key atmospheric gases and aerosols) AR4. These historical 20C3M scenarios are set to preindustrial conditions and the are essential as they provide a baseline model is run for several hundred years of observed climate against which in order to “spin up” to the equilibrium condition in which our planet exists. Future climate projections should neverfuture beclimate compared change directly can be to quantified. historical This is an essential step as climate observations in order to calculate the models are, at their most basic level, amount of change that may occur, as even simply numerical approximations of the the best models contain biases relative to fundamental laws of physics that govern observations. Rather, future projections nature at the scale of the planet, including must be compared to historical simulations, conservation of momentum, conservation as biases are assumed to remain relatively of mass and energy, and the four laws of constant over time. This change, or delta, thermodynamics. Control runs are not provides the most reliable information on intended to be used by anyone outside the future change. modeling community; their purpose is to establish a baseline set of model conditions that can be used to initiate a transient Scenarios simulation beginning in preindustrial times A1B 25 A1T and moving forward in to the future. A1F1 A2 Historical Scenarios 20 B1 B2 IS92a Once a preindustrial control run has been 15 completed for any given climate model,
a transient (time-dependent) simulation Emissions (GtC) 2 can be run, beginning in the year that was 10 used to set conditions for the control run. CO This initial year varies from about 1850 to 5 1890, depending on the global modeling group. Historical scenarios run from the 2000 2020 2040 2060 2080 2100 beginning year, in the 1800s, through 1999. Figure 4.1. Projected future carbon Each month, external forcing mechanisms emissions for the SRES emission scenarios. observed or measured for that month in Emissions for the highest scenario (A1FI) the past are input to the models. These correspond to the red dotted line at the top, while emissions for the lowest (B1) observed drivers can include changes in scenario are indicated by the solid green solar radiation, volcanic eruptions, human line (Nakicenovic et al. 2000). emissions of greenhouse gases and other radiatively active species, and secondary changes in ozone and water vapor.
Climate and Ecological Models 53 Emissions Scenarios CO2 concentrations (the amount of CO2 in the atmosphere as a result of emissions) To estimate potential climate changes reach 940 ppm by 2100—more than triple through 2100, we need to ask: preindustrial levels.
How will human societies and economies The lower-emissions scenario (B1) also evolve over the coming decades? represents a world with high economic • growth and a global population that peaks What technological advances are mid-century and then declines. However, expected, and how will they affect this scenario includes a shift to less emissions?• fossil fuel–intensive industries and the
Which energy sources will be used in technologies. Emissions of greenhouse the future to generate electricity, power gasesintroduction peak around of clean mid-century and resource-efficient and then transportation,• and serve industry? decline. Atmospheric CO2 concentrations reach 550 ppm by 2100—about double The answers to these questions will affect preindustrial levels. future emissions of greenhouse gases from human activities. And these emissions will Concentration Pathways in turn determine future climate change at both the global and the regional levels. New Representative Concentration Pathways (RCP) (Moss et al. 2010) are To address these questions, in 2000 the under development for the IPCC Fifth IPCC developed a set of future emissions Assessment Report (AR5). In contrast to the scenarios known as SRES (Special Report SRES scenarios used in the AR4, the RCPs on Emissions Scenarios) (Nakicenovic are expressed in terms of carbon dioxide et al. 2000). These scenarios use a wide equivalent (CO2-eq) concentrations in the range of projections for future population, atmosphere, rather than direct emissions. demographics, technology, and energy use to estimate the greenhouse gas emissions Although climate model simulations are that would result from a variety of possible not yet available for the RCPs, it is still futures. In doing so, they cover a wide possible to place SRES-based projections range of plausible futures that illustrate into the context of these new scenarios by differences in the extent and severity of the converting the SRES emission scenarios global warming that result from alternative to CO2-eq concentrations. When we do emissions choices (Figure 4.1). that using the simple energy-balance climate model MAGICC (Model for the For example, the SRES higher-emissions or Assessment of Greenhouse-gas Induced fossil-intensive scenario (A1FI) represents Climate Change), we see that the highest a world with fossil fuel–intensive economic RCP 8.5 corresponds closely to the higher growth and a global population that SRES A1FI emissions scenario, with end- peaks mid-century and then declines. of-century CO2-eq concentrations of 1465 ppm for RCP 8.5 as compared to 1360 are introduced toward the end of the ppm for A1FI (Figure 4.2) (Wigley 2008). century.New and In more this efficientscenario, technologies atmospheric In contrast, the lowest RCP 2.6 projects
54 Scanning the Conservation Horizon Climate and Ecological Models a future where emissions are reduced illustrate the potential range of changes that could be expected, and how these SRES scenarios, with CO2-eq concentrations depend on energy and related emission significantlyrising to nearly below 500 even ppm the then lowest falling of to the choices made over coming decades. 450 ppm by the end of the century. The mid-low RCP 4.5 corresponds most closely Uncertainties in Future Scenarios to SRES B1, with CO2-eq concentrations of nearly 600 ppm by the end of the century It is important to note that, as broadly as compared to 640 ppm for B1. separated as they are, neither the SRES nor the RCP scenarios cover the entire range of Figure 4.2 is an important comparison as possible futures. While the recent economic it enables the climate model projections decline slowed CO2 emissions growth rates currently available (and indeed, all that in comparison to previous years, actual will be available through 2012) to be emissions remain near the top of the range placed in the context of the next generation of IPCC scenarios for the period 2000 to of climate scenarios. It also reveals that 2010 (Manning et al. 2010). the substantial difference between the SRES A1FI and B1 scenarios, although conservative in comparison to the RCPs in in emissions, on the order of 80 percent its estimate of the lower end of the range belowOn the 1990other levels hand, or significant more, could reductions stabilize CO2 levels below the lowest SRES emission of future emissions, is still sufficient to Emissions Concentrations
40 1500 A1F1 B2 A1F1 B2 A2 A1T A2 A1T A1B B1 eq) A1B B1 30 2 1200
20 900
SRES (IPCC AR4) 10 600 GHG concentrations (CO greenhouse gas emissions (GtCeq) 0 300 2000 2020 2040 2060 2080 2100 2000 2025 2050 2075 2100
40 1500 RCP 8.5 RCP 8.5 RCP 4.5 RCP 4.5
eq) RCP 2.6
RCP 2.6 2 30 1200
20 900 RCP (IPCC AR5) 10 600 GHG concentrations (CO greenhouse gas emissions (GtCeq) 0 300 2000 2020 2040 2060 2080 2100 2000 2025 2050 2075 2100
Figure 4.2. Projected future CO2-equivalent emissions and concentrations for the SRES emission scenarios (IPCC AR4) and Representative Concentration Pathways (IPCC AR5).
Climate and Ecological Models 55 atmosphere, oceans, and earth’s 6.0 A2 surface. As output, AOGCMs produce A1B geographic grid-based projections of C)
O 5.0 B1 precipitation, temperature, pressure, Year 2000 Constant Concentrations cloud cover, humidity, and a host of other 4.0 20th Century climate variables at daily, monthly, and annual scales. 3.0 Because of the complexity of these models, 2.0 they are generally designed and run by large research teams at supercomputing 1.0 centers. Models are constantly being 0.0 Global Surface Warming ( climate improves and as computational enhanced as scientific understanding of -1.0 power increases. Over time, the number of global climate models has grown. By 1900 2000 2010 2008, 16 international climate modeling Year teams had submitted historical and future simulations from 25 different climate Figure 4.3. Projected future global models to the IPCC’s AR4. All future temperature change for the SRES emission simulations by these models agree that scenarios (degrees Celsius). The range for each individual emission scenario indicates both global and regional temperatures model uncertainty in simulating the response will increase over the coming century of the earth system to human emissions of in response to increasing emissions of greenhouse gases (IPCC 2007b). greenhouse gases from human activities (Figure 4.3). scenario (e.g., Meinshausen et al. 2006). Such policy options were not considered in Model Selection the SRES scenarios, although the new RCPs (Moss et al. 2008, 2010) currently under Some models are more successful than development for the IPCC Fifth Assessment others at reproducing observed climate Report at least partially address this issue. and trends over the past century in particular geographic regions. Inter-model Global Climate Models comparisons (e.g., Kunkel et al. 2006; Wang et al. 2007; Stoner et al. 2009) generally Description three broad categories: “good” models Emissions or concentration scenarios are thatfind thatare able climate to simulate models importanttend to fall climate into used as input to global climate models, features, from Arctic weather systems to which vary in complexity. The most natural variability such as El Niño, across complex are the AOGCMs. These are large, the globe; “fair” models that perform three-dimensional coupled models that well in some regions and at some tests, incorporate the latest understanding but poorly in others; and “poor” models, of the physical processes at work in the usually those in relatively early stages
56 Scanning the Conservation Horizon Climate and Ecological Models of development, that consistently fail to otherwise been evaluated and shown to reproduce fundamental aspects of the adequately reproduce key features of the earth’s climate system. atmosphere and ocean system. Key features include seasonal circulation patterns, For the purposes of global climate model selection for impact analyses, however, the and other teleconnection patterns affecting most relevant point is that multi-model climateatmosphere–ocean in the region heat (Covey fluxes, et al.El 2003;Niño, comparisons have shown that, for a given AchutaRao and Sperber 2002; Chapman emissions scenario, the average of multiple and Walsh 2007; DeGaetano et al. 2008; models generally provides a more robust Vrac et al. 2006). picture of future conditions than any one model (Tebaldi and Knutti 2007). Moreover, The models chosen should encompass evaluating which set of models may be the greater part of the range of uncertainty “best” at simulating future trends over in• climate sensitivity simulated by global any given region and for a certain variable climate models. Because many of the is a long and involved process, generally processes at work in the earth–atmosphere requiring a fundamental understanding of system are not yet fully understood, these climate dynamics that have contributed to are represented somewhat differently in climate variability and long-term trends different global climate models. A range in over the region of interest. (Although projected temperature change and other simple biases or differences between climate variables arise from the different models and observations have been climate sensitivity of the models. used in the past to judge which models are “better” than others, this practice is From a purely practical perspective, highly discouraged as the information the projections required for the impact derived from such a calculation bears little analysis• must be available from that model, relevance to actual model performance in preferably for multiple future scenarios simulating climate change and may mislead in order to encompass the uncertainty in more than it may guide.) So in evaluating predicting future drivers of change. the potential impacts of climate change on any given region, it is always best to use the average of multiple global climate models rather than to rely on one or two. As it may not be possible to use all global climate model simulations, some general criteria for model selection are the following:
Consider only well-established models, whose strengths and weaknesses are already• extensively described and evaluated
The models should have participated in the Coupledin the peer-reviewed Model Intercomparison scientific literature. Project USFWS (CMIP; http://cmip-pcmdi.llnl.gov/) or
Climate and Ecological Models 57 Box 4.1. How Much Sea-Level Rise Should We Plan For?
Asking how much sea-level rise to assume in coastal vulnerability assessments is a common question, but not necessarily straightforward to answer. As with modeling the climate system itself, projecting how much global sea levels will rise due to global warming is complex, and there are many sea-level rise scenarios in the literature. In general, global sea-level rise figures refer to changes in eustatic sea level, referring to changes in the volume of seawater. For planning purposes, it is important to recognize that a number of other factors affect the amount of relative sea-level change experienced in a given area. Local land subsidence, for example, will mean that the rate of relative sea-level rise in that area will be higher than the eustatic rate. Nevertheless, projections for eustatic sea level are an important baseline for assessing coastal vulnerability.
Modeling eustatic sea level requires integrating factors such as thermal expansion of the oceans as well as rates of change of land-based ice, including glaciers and continental ice sheets, all of which are subject to uncertainties. Some of the most widely used scenarios are from the IPCC, whose most recent estimates range from an additional 7 to 23 inch rise in global average sea level over 1990 levels by the 2090s (IPCC 2007b). There is compelling new evidence, however, that because these figures ignore recent dynamic changes in Greenland and Antarctica ice flow, they underestimate the rate of sea-level rise that we are likely to experience during this century (Overpeck and Weiss 2009; Pfeffer et al. 2008). Taking at least some of this accelerated melting into account, Vermeer and Rahmstorf (2009) suggest that a feasible range might be 30 to 75 inches for the period 1990-2100. Complicating matters, the magnitude of sea-level rise will not be globally uniform because of ocean circulation patterns and the earth’s rotation, gravitational differences, local and regional geological differences, and other factors (Bamber et al. 2009; Mitrovika et al. 2009; Church et al. 2004; Yin et al. 2009).
Ultimately, choosing which scenarios upon which to base vulnerability assessments and associated climate change adaptation strategies will depend on how much risk we are willing to accept. When relatively little is at stake in the way of infrastructure investment or public inconvenience, we could choose to design for a conservative or low-end sea-level rise scenario. Where more is at stake, such as the decimation of habitats critical to a region’s ecological and economic well-being, we might design for a mid-range or aggressive sea-level rise scenario. The bottom line, however, is that we may never know with certainty how much or how fast sea level will rise. Accordingly, using a range of scenarios in your assessment might be the optimal approach.
Finally, note that within any climate Downscaling of Global model, uncertainties will vary for different Model Simulations simulated• phenomena. For example, large-scale surface temperature is The geographic grid cells that form the generally trusted more than precipitation, basis of AOGCMs typically range in size which requires many physical processes from about 160 to 800 kilometers per to be simulated correctly at the same side. In general, this type of resolution time and place.
58 Scanning the Conservation Horizon Climate and Ecological Models (a) HCN Summertime (b) PCM Summertime Max Temp: 1990s Max Temp: 1990s 30.0 30.0
29.75 29.75
450 29.50 29.50 450 29.25 29.25
29.0 29.0
28.75 28.75 400 400 28.5 28.5 -75 -70
(c) RCM Summertime (d) SD Summertime Max Temp: 1990s Max Temp: 1990s 30.0 30.0
29.75 29.75
450 450 29.50 29.50
29.25 29.25
29.0 29.0
28.75 28.75 400 400 28.5 28.5
Figure 4.4. Average summer maximum temperature in the U.S. Northeast for the 1990s, based on (a) observations from weather stations in the Historical Climatology Network, (b) the global climate model PCM, (c) regional climate model simulations, and (d) statistically downscaled simulations. From Hayhoe et al. 2008.
Dynamical Downscaling changes experienced at the regional scale. Foris too this coarse reason, to capturea number the of fine-scale downscaling Dynamical downscaling, or regional techniques have been developed to climate modeling, uses a high-resolution transform global climate model output into climate model centered over a relatively higher-resolution projections capable of small region and driven by global climate resolving the impacts of climate change on local conditions. Model grid cells range from 10 to 50 model output fields at its boundaries.
Climate and Ecological Models 59 kilometers, and contain substantially Statistical Downscaling different physics than the global models in order to resolve the physical processes Statistical downscaling relies on historical that occur at spatial scales below those of instrumental data for calibration at the the global models. Regional climate models local scale. A statistical relationship is are able to simulate the dynamic changes in climate likely to occur as global climate for a past “training period,” and observed changes; however, regional climate model climatefirst established variables between of interest AOGCM (here, output daily simulations are expensive to run and maximum and minimum temperature and few global climate models save the high- precipitation). This relationship is averaged over at least two decades to remove required to drive the regional models. Hence,resolution regional temporal model fields simulations (at 3 or 6 tend hours) relationship between AOGCM output and monthlyyear-to-year or daily fluctuations. climate variables The historical at the future decades, rather than a continuous regional scale is then tested using a timeto be period,limited andto specific tend to timeslices be driven fromby second historical “evaluation period” to several global models rather than the full ensure the relationship is valid. If so, suite of IPCC global climate models. The most comprehensive set of regional model downscale future AOGCM simulations to simulations has been generated by the thatthen samethe relationship scale. is finally used to North American Regional Climate Change Assessment Program (NARCCAP), which Unlike regional climate modeling, statistical uses four regional model–global model downscaling assumes that the relationships pairs to simulate conditions for 2041 to between large- and small-scale processes 2070 compared to 1971 to 2000.
forremain precipitation. fixed over However, time. This analysis assumption of 37 stationsmay not inalways the state be justified, of Illinois particularly suggests that this relationship only breaks down for the most extreme precipitation events above the 99th percentile of the distribution (Vrac et al. 2006). Analyses for the Northeast (Hayhoe et al. 2008) further indicate that, in areas of variable topography such as mountains and coastlines, statistical methods trained to match historical spatial patterns may perform better than regional climate models that are limited by their convection schemes (see Figure 4.4). In addition, statistical downscaling has a substantial time and cost advantage;
Ron Schreiber hundreds of years of model simulations can be downscaled using the same computing resources required to run only a few
60 Scanning the Conservation Horizon Climate and Ecological Models years of regional-model downscaling. of species, habitats, and ecosystems. Three of the case studies in Chapter VII Moreover, response models help identify use statistical downscaling for climate indicators and potential thresholds change data. Case Studies 1, 3, and 6 rely or tipping points that can be used in on a tool called the ClimateWizard (www. vulnerability assessments (Bradley and climatewizard.org), which was developed Smith 2004; Groffman et al. 2006). by The Nature Conservancy, University of Washington, and University of Southern The decision on which model or Mississippi to enable technical and combinations of models to use depends nontechnical users to access historical and on the species, habitats, and ecosystems projected climate change information for of concern, the types of questions being a given area based on downscaled data asked, and the particular end-users’ needs. from a range of climate models, emissions The local geographic and biophysical scenarios, and time periods. In Case Study characteristics of habitats and ecosystems 7, researchers will be conducting their own require the use of response models that are statistical downscaling projections for at least six future climate projections. settings (Primack et al. 2009). Moreover, thewell type suited of questionsto the specific and needsenvironmental associated Ecological with the vulnerability assessment affect the scale and scope of the assessment. The Response Models objective of some vulnerability assessments is to target species or geographic areas Ecological response models are a critical where species, habitats, and/or ecosystems part of the overall vulnerability assessment are potentially most vulnerable to declines process. They provide a way to assess in conditions due to climate change the sensitivity and potential adaptability (Bradley and Smith 2004). These types or resilience of wildlife species, habitats, of vulnerability assessments are usually and ecosystems to climate change conducted at large basin or regional scales, (Wormworth and Mallon 2007). They also and the results are used to target species are fundamental to understanding the and geographic areas where further kinds of climate variables that are needed attention (data gathering, modeling, etc.) to conduct vulnerability assessments. As is needed. Other vulnerability assessments such, information from response models rely on more complex models involving enhances the iterative dialog between biologists/ecologists and climatologists geographies, as well as a range of species such that downscaled climate change traitsspecific and species processes or habitats (Martin in 2007). specific variables address the most appropriate scales (temporal and spatial) and scope An important part of model selection (types of variables that relate to sensitivity and/or development process is to and/or resilience) needed to conduct the clearly identify the endpoints of interest. vulnerability assessment. This dialog is critical in identifying the spatial extent aspects of the environment upon which of the downscaled climate information environmentalEndpoints are definable assessments and aremeasurable made. that will then be used to assess exposure Lack of clear endpoints often reduces the
Climate and Ecological Models 61 relevance and utility of the assessment and software can be found at http://www. results (Boyd 2007). In some cases, species’ vulnerabilities are assessed because they model.html. are the endpoint (e.g., a species listed fileheap.com/software/conceptual_data_ through the ESA) or because they affect General Characterization Models ecosystem functions (e.g., an invasive Characterization models usually represent regimes) (Hunter 2007). broad groups (e.g., amphibians, riparian species as it affects fire disturbance species) or generalized traits to identify Types of Response Models how groups of species might respond to climate and/or habitat change. These models can be fairly simple. For example, types of response models and how they are they can involve groupings of species generallyThis section used. briefly Different describes approaches different are based on their preference for certain often used in combination (Martinez-Meyer habitats. Or, they can involve species 2005), but for clarity we provide basic grouped by physiological, functional, descriptions of each. and/or other biological traits, and their
Conceptual Models aspects of climate change (Lavorel et al. 1997).potential In sensitivitymost cases, (response) spatial distribution to specific Conceptual models are qualitative of sensitivity is estimated by applying the descriptions and diagrams of key attributes classes of vulnerability (characterization part) to species distribution maps (Lavorel habitats, or ecosystems of concern (see et al. 1997). Another characterization Schlesingerand processes et al.related 1990 to for specific an example species, approach involves meta-analyses whereby existing studies are pooled together to illustrate or describe important linkages to estimate common responses of species stressors,related to suchdesertification). aspects of climate They also and land (Parmesan and Yohe 2003; Allen et al. use, and how changes in stressors affect 2010). These can be especially effective important attributes (e.g., soil texture) at validating responses over large areas. The NatureServe Climate Change etc.). These linkages to stressors provide Vulnerability Index is an example of an theand basic functions information (nutrient needed uptake, to water assess flux, easy-to-use characterization tool to assess vulnerability (e.g., sensitivity), and help vulnerability (http://www.natureserve. inform the spatial extent of the stressor org/prodServices/climatechange/ needed to analyze exposure. Nearly all ClimateChange.jsp) (see Case Study 1). types of other response models are built from well-conceived conceptual models Expert Opinion Models as they help identify key variables in models needed to assess sensitivity and These models are constructed from exposure. For a review of conceptual model the opinions of experts on a particular development see Heemskerk et al. (2003). species, habitat, or ecosystem. They are Various types of conceptual modeling tools often used when existing data preclude or
are insufficient to develop a quantitative
62 Scanning the Conservation Horizon Climate and Ecological Models model. A series of workshops and/or determine the spatial distribution of surveys are often used to gather data suitable habitat for individual species. In from experts that are then used in model a few cases, species ranges and suitable development. In some cases results from expert input are combined with other data variables (e.g., Climate Envelope models) (e.g., from existing publications). Statistical (Harrisonhabitat have et al.been 2006; defined Pearson largely and by Dawson climate approaches (e.g., Bayesian statistics) are 2003). In these cases, species sensitivities often used to combine these data from and vulnerability can be directly assessed different sources to produce estimates with climate data. In cases where suitability of potential responses and uncertainty (Berliner et al. 2000; Prato 2009). Expert attributes (including distribution), models opinion is often used in assembling ofis definedhabitat shiftsby vegetation associated or landwith coverclimate conceptual models (see previous change scenarios are needed to conduct the discussion). There are many software assessment (Johnson et al. 2005). modules that can be downloaded to assist in development of expert opinion models. One example of this type of model is the Sea Some examples include the Bayesian Level Affecting Marshes Model (SLAMM), in Analysis Toolkit (http://www.mppmu.mpg. de/bat/), Treeage Pro software (http:// incorporating geometric and qualitative www.treeage.com/products/index.html), relationshipswhich a flexible is usedand complex to represent decision transfers tree and the Delphi Decision Aid site (http:// among coastal habitat classes under armstrong.wharton.upenn.edu/delphi2/). various scenarios of sea-level rise (Clough et al. 2010). Case Study 5 uses SLAMM Habitat and Occupancy Models (Version 5.0) in an assessment of coastal habitat vulnerability in the Chesapeake Bay Habitat and occupancy models are perhaps region. Vegetation- and habitat-associated the most common models used to address response to climate change seem to be the potential vulnerability of species to climate most prevalent type of response model and land-use change. Some habitat models, used to evaluate potential effects of climate such as those that have been developed change on species. by the U.S. Geological Survey GAP model habitat suitability over large geographic Niche-based models are also used to areas based on the development of habitat estimate species distributions and habitat criteria (developed mostly from expert suitability, but they generally involve more opinion and published literature). These quantitative approaches with estimates of requirements (rules) are expressed as the probability of occurrence of a species. Approaches such as Genetic Algorithm for (e.g., climate, soils, vegetation or land cover, Rule-set Production (GARP) (Stockwell elevation,ranges in specificetc.) that biophysical a species will attributes occupy and Peters 1999) and Maximum Entropy (e.g., “suitable” habitat). The requirements modeling (Maxent) (Phillips et al. 2006) are (range of biophysical conditions) are then examples of niche-based models. Niche and applied to wall-to-wall biophysical data, occupancy models produce probabilities such as through Geographic Information of occurrence using locations records System (GIS) coverages or grids, to of species (e.g., from museum records,
Climate and Ecological Models 63 systematic surveys, and other databases) Software used to develop niche and and wall-to-wall biophysical data such occupancy models can be downloaded for as elevation, topography, temperature, free from the Internet, including GARP precipitation, soils, and geology to name (http://www.nhm.ku.edu/desktopgarp/), a few (Ballesteros-Barrera et al. 2007). Maxent (http://www.cs.princeton. edu/~schapire/maxent/), Regression of the species. Other statistical models Trees and Random Forests (http://rattle. incorporateAs such, they approaches define the “habitatlike Regression niche” togaware.com/rattle-download.html), and Tree and Random Forests (O’Connor et Bioclim (http://software.informer.com/ al. 1996; Lunetta et al. 2004; Garzón et al. getfree-bioclim-download-software/). 2006). These also can be used to access Additionally, many of the species models habitat suitability for species, but have the developed by the GAP program can also be advantage of being able to assess how the downloaded, including biophysical data importance of biophysical attributes change (http://www.nbii.gov/portal/server.pt/ among different geographies (O’Connor et al. 2006). modeling/7000). community/maps_and_data/1850/species_ When climate change variables are Vegetation/Habitat Response Models important attributes of these models, direct responses to different climate change Many of the approaches and models scenarios can be assessed statistically. described for animal species above are When biophysical attributes such as also used to estimate potential response vegetation and land cover are the most of plants to climate change (Lawler et al. important variables, species responses to 2006). However, because of greater depth climate change are evaluated by applying of historical records for plants (e.g., pollen models of vegetation and/or land cover responses to climate. Statistical functions and vegetation response models using generated from these models result from historicalrecords), itrecords is often (Cole possible 2010). to refineMoreover, plant spatial variation and not the responses of plant and vegetation distribution records species and habitats to climate change over are considerably more abundant than time. That is, they use spatial variability animal records and, therefore, are easier to determine how species and habitats to model using statistical approaches (Van might respond to climate change. However, Mantgem et al. 2009). Vegetation and plant recently, phylogenetic and phylogeographic community models are critically important analyses have been used in combination to assess animal species changes, especially with niche-based models to improve when animal distributions and habitat response models by adding a historical response component (Waltari et al. 2007). plant community variables. Robinson et These analyses use molecular markers al.suitability (2008) provideare defined an extensive by vegetation review and and tree-based statistical approaches to of vegetation climate models, broadly determine how species have responded grouped into statistical species distribution models, GAP models, landscape models, historical climate change. biogeochemical models, and dynamic global (diversification, range contraction, etc.) to vegetation models.
64 Scanning the Conservation Horizon Climate and Ecological Models Physiologically Based Models plant nutrient cycling (Christensen et al. Physiologically based response models 2008;nitrogen Tague fluxes, et al. evapotranspiration, 2009). The CENTURY and incorporate sensitive aspects of individual model is a general model of plant–soil nutrient cycling and it has been used to nesting/reproduction, thermoregulation, simulate carbon and nutrient dynamics for andspecies migration physiologies (Root 1988athat influence and 1988b; foraging, different types of ecosystems, including Martin 2001; Reist et al. 2006; Bernardo grasslands, agricultural lands, forests, and et al. 2007; Hunter 2007). Broad-scale savannas (Ojima et al. 1996). DayCent- changes in species distributions have been Chem, which was built off of the CENTURY tied closely to physiological constraints model, predicts carbon and nitrogen (Root et al. 2003). The aim is to relate dynamics within forests and leaching the physiological traits and processes to of different types of nitrogen cations climate change variables. These models can from the forests to streams (Hartman et be used as part of general characterization al. 2007). The MC1 model is a dynamic models or as part of habitat models (Root vegetation model that combines the et al. 2003; see earlier discussions). More CENTURY biogeochemical model with a complex models quantify interactions biogeographical model, MAPSS (Mapped between key physiological variables and Atmospheric-Plant Soil System) (Bachelet other variables, such as behavior, growth, et al. 2001). The Regional Hydro-Ecologic and survival, and how important climate Simulation System (RHESSys) is a variables such as temperature affect GIS–based hydro-ecological modeling interactions (Biro et al. 2007). However, framework, which simulates how water, to build over large areas because of the the environment on a watershed scale amountcomplex and models type of of these data neededtypes are (e.g., difficult data (Christensencarbon, and nutrients et al. 2008). fluctuate The PnET through is a on movement patterns, behavior, growth, suite of three nested computer models, survival, etc.). Therefore, these models are which provide a modular approach to simulating the carbon, water, and nitrogen dynamics of forest ecosystems ofmost any commonly off-the-shelf built software for specific or tools species that (Aber et al. 1995). All of these have permitin specific development geographies. of physiologicallyWe are unaware climate-related inputs that permit an based climate response models. analysis of the potential impacts of climate change on fundamental ecological and Ecological Models hydrological processes.
There are several ecological models that Some of these models can be downloaded can be used to assess sensitivity and from the Internet, including RHESsys vulnerability of important ecological processes to climate change. Ecological PnET (http://www.pnet.sr.unh.edu/), and response models evaluate how climate the(http://fiesta.bren.ucsb.edu/~rhessys/), CENTURY model (http://www.nrel. change variables affect fundamental colostate.edu/projects/century5/). ecological processes such as carbon and
Climate and Ecological Models 65 Box 4.2. Ecological Thresholds
Ecological thresholds are important when considering response models because they represent situations where a small change in a driving variable, such as temperature or precipitation, leads to a disproportionately large response. When a system crosses such a threshold, the fundamental drivers and important processes can change abruptly, and these abrupt departures are often very difficult to include in models. In biological systems, climate-related threshold events include death of corals as a result of high water temperatures (Ward et al. 2007), widespread loss of piñon pines due to drought and high temperatures (Breshears et al. 2005), and the sudden eruption of spruce beetle in Alaskan forests when recent temperature increases permitted reproduction in a single season (Werner et al. 2006).
At the level of a single species, threshold events may result from a simple, direct effect of climate, such as exceeding a lethal temperature. At the ecosystem level, threshold events may be very broadly distributed, and they frequently involve positive feedbacks that amplify effects that otherwise may been a smaller perturbation to the system (U.S. CCSP 2009a). Wildfire behavior is a good example of a physical process with multiple thresholds (Peters et al. 2004). When small, wildfire spread is determined largely by local fuel attributes. As fire extent and intensity increase, several thresholds are crossed and the processes that drive fire behavior are almost completely different. Very large wildfires generate their own surface winds that can drive a fire across areas with small fuel loads and little fuel connectivity. Models with fundamentally different structures, scales, and processes are needed to accurately simulate wildfires across this range of scales (Peters et al. 2004).
Thresholds pose special problems for conducting climate change vulnerability and risk assessments. We know little about the location, even approximately, of most ecological thresholds, while the consequences of crossing a threshold can be profound. A commonsense approach to thresholds is to focus on an opposing attribute—the resilience of an ecological system. Resilience is the ability of a system to retain characteristic processes and structures when subjected to change or disturbance. As mentioned previously, many recommendations for responding to climate change focus on increasing the resilience of systems (U.S. CCSP 2009a; Heller and Zavaleta 2009), and these actions will help avoid catastrophic thresholds.
Limitations of with climate change (e.g., temperature Response Models tolerance), or based on shifts in habitats associated with climate change. These approaches are used because data on species attributes (e.g., demography, habitat world and, as such, have limitations in preferences, etc.) and species and habitat theirAll models use and are applications. simplifications Most of responsethe real distributions are available. However, these models used in vulnerability assessments models tend to ignore other factors and are simple models involving shifts in species traits that can affect vulnerability to species ranges based on direct interactions climate change. These include:
66 Scanning the Conservation Horizon Climate and Ecological Models Changes in interactions between species, Potential responses of species to climate including competitive interactions and change also may be affected by landscape disease• (Peterson et al. 2002; Murray et context. Landscape context has been al. 2006). shown to be an important factor in species survivorship and response to stressors Nonlinear, complex responses (e.g., such as land use (Ricketts 2001; Baum et al. thresholds or tipping points) associated 2004). Most species response models lack with• indirect interactions (Burkett et al. factors associated with landscape context 2005) (see Box 4.2). and pattern. One exception is the PATCH model (Schumaker et al. 2004), which Interactions between climate change and incorporates aspects of landscape pattern other important drivers or stressors such and species traits in assessing vulnerability. as• land-use and land-cover change (Root This model can be downloaded from the and Schneider 2002). Internet (http://www.epa.gov/wed/pages/ news/03June/schumaker.htm). as species migration (immigration/ Horizontal flow processes such Availability of emigration)• that can determine a species’ Response Models ability to move across an area, either in direct response to climate change or as an We highlighted a variety of models that can indirect response to habitat shifts (McRae be downloaded from the Internet. Some of et al. 2008). these models are relatively simple to use, while others require extensive data sets and Another issue regarding response models is adjustments to the models for individual transferability to different geographies and applications. However, even simple models scales. General relationships established should be used carefully, with close using broad-scale response models may examination of the caveats involved on any not hold up when assessing vulnerability particular modeling approach. There is a of a species at a local scale (Frederiksen et need for a Web portal that provides links al. 2004; Torti and Dunn 2005). Moreover, to the range of existing response models, and for enhanced simulation frameworks models developed in one biophysical that facilitate linking (e.g., climate models settingit may be to difficultanother tobecause transfer of responsedifferences to ecological response models). The in local responses to climate variables Terrestrial Observation and Prediction (Primack et al. 2009). System is an example of a sophisticated simulation framework that links historical Species selection may also affect the climate data, remotely sensed data, climate outcome of vulnerability assessments. projections, and response models (Nemani Species with limited distributions have et al. 2009). greater uncertainty in their response to climate change than species with broader ranges (Schwartz et al. 2006). Therefore, decisions on which species to include in a vulnerability assessment can affect the outcome and accuracy of the assessment.
Climate and Ecological Models 67 V. Addressing Uncertainty in Vulnerability Assessments
his chapter addresses issues and approaches for dealing with inclusion into a risk assessment or analysis. T RiskQuantification assessment of involvesuncertainty estimating can allow both for the context of conducting climate the probability of an event occurring, and change vulnerability uncertainty assessments specifically within(i.e., the severity of the impacts or consequences uncertainties related of that event. Analyses to identifying and Dealing with uncertainty of risk, therefore, modeling the sensitivities, is nothing new in natural provide an opportunity levels of exposure, and adaptive capacity of the resource management. uncertainties through assessment targets). The probabilisticto address quantifiable calculations. following chapter (Chapter VI) discusses Not all uncertainties can be addressed in how to develop adaptation management a risk assessment, as there are unknowns and planning activities in light of those uncertainties. While risk assessment may allow for the inclusionthat in many of some cases types cannot of beuncertainty, quantified. it uncertainty as it applies is also important to communicate those in both of these contexts. According to the uncertainties that cannot be handled IPCCTo begin, (2007a), we define uncertainty is:
“An expression of the degree to which a Managersthrough exact have quantification. always made decisions value (e.g., the future state of the climate even though there are uncertainties that system) is unknown. Uncertainty can result from lack of information or from or eliminated. Dealing with uncertainty disagreement about what is known or iscannot nothing be quantified,new in natural much resource less reduced even knowable. It may have many types management. Being transparent about the general magnitude of uncertainty and understanding the range of possibilities orof sources,terminology, from or quantifiable uncertain projectionserrors in given the uncertainty allows managers ofthe human data to behavior. ambiguously Uncertainty defined can concepts to articulate the reasoning for making a therefore be represented by quantitative measures (e.g., a range of values calculated change, managers may be seeking “bet by various models) and/or by qualitative hedging”specific decision. strategies With that regard make tosense climate under a number of plausible future scenarios or a team of experts).” that are generally robust to uncertainty. statements (e.g., reflecting the judgment of
Lead authors: Linda Joyce, Molly Cross, and Evan Girvetz.
68 Scanning the Conservation Horizon Addressing Uncertainty With respect to vulnerability assessments, Uncertainty in the the goal should be to use the best Scientific Literature available information on the uncertainties involved in estimating vulnerability, while Methods to quantify uncertainty or recognizing that it may be necessary to reassess vulnerability and the associated have been the subject of much study (e.g., uncertainties in an iterative fashion as uncertaintyconfidence in can assessments arise in a variety and analyses of ways in new information becomes available. The an analysis or an assessment). Quantitative assessment of uncertainty should identify analyses of species vulnerability to climate both the best estimate of how vulnerable change use mathematics or statistics to a system is to climate change but also describe the relationship between climate the potential range of vulnerability given and the species of interest. In these uncertainties. quantitative analyses, uncertainty can arise in the structure of the mathematics used to The Language of data used to parameterize the equations. Uncertainty Fordescribe example, the phenomenadifferent vegetation as well as models the field may describe quantitatively the growth of vegetation using different mathematical assessing uncertainty, it is important to expressions, which can mean slight acknowledgeBefore diving theinto need specific for “amethods language for of differences in the results under climate uncertainty.” The vulnerability assessment change. Uncertainty is also found in what is not known about the phenomena. For example, how elevated CO2 andwill synthesizemodeling experiments, scientific information as well as from field studies, experimental studies, plant growth is not known precisely, and the vegetation models differ will influence in their pulling the information together. Users of expression of this process. the vulnerabilityscientific knowledge assessment of the will experts want to know what the authors conclude about the assessment results based on the variety of different sources of information. have in the results of the vulnerability assessment?How much confidence There is a do need the to authors be consistent in describing the uncertainty so that the degree of uncertainty can be clearly communicated across vulnerability assessments (U.S. CCSP 2009c). We present several methods that can be used to describe the certainty of the assessment.
USFWS
Addressing Uncertainty 69 When the assessment is based on quantitative and qualitative analyses and modeling studies. Fewer techniques are availableon quantifying to structure uncertainty the uncertainty in the field in or by experts, then uncertainty is found in assessments where information from a a synthesis of the scientific literature variety of sources is synthesized to assess to the question of interest in addition the vulnerability of species or communities tothe the lack uncertainty of available in information the quantitative specific or ecosystems to climate change. It should studies. The breadth and nature of the also be noted that this discussion focuses authors’ background and experience also only on the assessment of vulnerability. contributes uncertainty. For example, These results must be cast against the assessing the climate impacts on a broader background on which the particular bird species may be limited by decision is made—and there will be what is known today about relationships uncertainty in those other factors between the bird’s biology and climate metrics such as temperature and species, habitat, or ecosystem. precipitation. However, there may also be influencing the management of the more complex interactions with climate, IPCC Approach to such as soil moisture affecting the habitat Uncertainty the birds use. The IPCC represents the longest focused Many techniques are available to attempt to describe uncertainty in the quantify and communicate uncertainty in context of climate change and has been mathematical, computer, and statistical evolutionary in the development of the models, many of which are summarized in methodology used (Risbey and Kandlikar the U.S. Climate Change Science Program’s 2007). The 2007 IPCC reports were explicit recent publication Best Practice Approaches about the language they used in describing for Characterizing, Communicating, and Incorporating Scientific Uncertainty in climate change. Yet, even though a common Climate Decision Making (U.S. CCSP 2009c). setuncertainty of guidance and was levels given of confidence to the authors, in Risk analysis methods, for example, can be the uncertainty language in the IPCC used to assess uncertainty when the range of all possible events is known, and Quantitatively Calibrated Levels of Confidence objective probabilities can Terminology Degree of Confidence in Being Correct be assigned to these events. Very high confidence At least 9 out of 10 chance of being correct The challenge in assessing High confidence About 8 out of 10 chance climate change impacts is in Medium confidence About 5 out of 10 chance quantifying the range of all possible events. In reviews Low confidence About 2 out of 10 chance of the attempts to quantify Very low confidence Less than 1 out of 10 chance uncertainty in the IPCC reports, reviewers note that the process was limited by the amount of work done in the primary literature
70 Scanning the Conservation Horizon Addressing Uncertainty A key observation about the language of of the subjects—for the physical sciences, uncertainty in the IPCC reports is that it has thereports uncertainty reflected language the disciplinary can build nature on evolved as scientists learn how to further the quantitative analyses in the sciences; for the socio-economic analyses, a more related to their analyses (Risbey and qualitative approach was taken given the Kandlikarrefine their 2007). science One and criticism the uncertainty of the nature of the primary analysis literature is the literature being less Some uncertainties can absence of a thorough quantitative. be quantified using assessment of the uncertainty. In some cases, “Where uncertainty statistics and modeling exploring the uncertainty is assessed more approaches, while in quantitative analyses can quantitatively using expert be expensive in terms of judgment of the correctness others may require more time and effort and hence of underlying data, models qualitative assessment. are not done as extensively or analyses, then the as someone applying that analysis might prefer. Further, the lack of a consistent approach to describing beingfollowing correct” scale (IPCC of confidence 2007a): levels is used uncertainty in the IPCC reports has been to express the assessed chance of a finding seen as inevitable given the very different nature of the science and the role of human is assessed using expert judgment and decisions in the potential responses to statistical“Where uncertainty analysis of in a specificbody of outcomesevidence climate change. (e.g. observations or model results), then the following likelihood ranges are used Likelihood Scale to express the assessed probability of occurrence” (IPCC 2007a): Terminology Likelihood of the Occurrence/Outcome Virtually certain >99 percent probability of occurrence “Where uncertainty is assessed Very likely >90 percent probability qualitatively, it is characterized by Likely >66 percent probability providing a relative sense of the amount About as likely as not 33 to 66 percent probability and quality of evidence (that is, information Unlikely <33 percent probability from theory, observations or models indicating whether a belief or proposition is Very unlikely <10 percent probability true or valid) and the degree of agreement Exceptionally unlikely <1 percent probability (that is, the level of concurrence in the approach is used by Working Group III throughliterature a onseries a particular of self-explanatory finding). This terms such as: high agreement, much evidence; high agreement, medium evidence; medium agreement, medium evidence; etc.” (IPCC 2007d).
Addressing Uncertainty 71 Assessing and The following are a few examples of methods available to address uncertainty in Understanding climate change vulnerability assessments: Uncertainty Monte Carlo Simulation. One common Identifying Sources quantitative approach for measuring uncertainty• is Monte Carlo analysis, as of Uncertainty Chapter VII (Case Study 1). Put simply, a Uncertainty in climate change vulnerability Monteexemplified Carlo insimulation the Nevada is acase computer- study in assessments can be rooted in a number based statistical technique that uses of stages in the process, including: random sampling to convert uncertainties in the input variables of a model (e.g., unknown interactions with non-climate incomplete knowledge of the climate limited/unreliable data; unidentified or sensitivity of a particular species) into interactions among different elements of probability distributions over output stressors; unidentified or unknown variables (Park 2008; Refsgaard et al. 2007; New and Hulme 2000; U.S. EPA 1997). orclimate unknown change; thresholds; unidentified ambiguously or unknown interspecific interactions; unidentified Expert Elicitation. The vulnerability disagreements about what is known; or defined concepts or terminology; scientific uncertain projections of human behavior. wildlife• habitats (Case Study 4) addressed Some of these uncertainties can be uncertaintyassessment forvia Massachusetts“expert elicitation,” fish andwhich is a formal, systematic process to determine approaches, while others may require more subjective judgments about uncertainties quantified using statistics and modeling qualitative assessment. A combination of from relevant experts (Refsgaard et these different methods can be used to al. 2007). Expert elicitations are often bound the uncertainty and understand warranted in cases where there are many the range of possibility for vulnerability to sources of uncertainty and where critical climate change (Refsgaard et al. 2007). information may be unavailable. The results of expert elicitation are often characterized quantitatively as probabilities that represent their levels
to include documentation of the evidence andof confidence. criteria used However, by the expertsit is also to important support their decisions.
Scenario Analysis. A relatively straightforward way to address uncertainties• inherent in projecting the
USFWS future is to base assessments on multiple scenarios (Walker et al. 2003). Scenario uncertainty implies that there is a range of possible outcomes, but the mechanisms
72 Scanning the Conservation Horizon Addressing Uncertainty leading to these outcomes are not well or reduce the overall uncertainty. For understood and it is, therefore, not possible example, there may be a species in the to formulate the probability of any one arid southwestern United States that is particular outcome occurring. For example, moisture limited, and global climate models if downscaled climate models are unable range from projecting small increases to to determine whether future conditions fairly large decreases in precipitation, with in a particular area will be warmer and temperatures increasing 2.5 to 4.5 degrees wetter or warmer and drier, assessing the Celsius by the 2080s. Although there is vulnerability of species or systems under a degree of uncertainty in the different both possible scenarios may be warranted. components, combining the temperature Similarly, given the currently wide range of and precipitation information shows that possible scenarios for eustatic sea-level rise the increase in temperature will offset and the numerous factors that can affect any projected increase in precipitation relative sea level at a local or regional level, resulting in a moderate to large decrease projecting future impacts and vulnerability in water availability, and thus a consistent based on a number of scenarios and increase in that species’ vulnerability. On the other hand, if a different species in this for determining possible management same area has a springtime temperature strategiesassumptions (see may Case offer Study the 5). most flexibility time, which is known within a range of Combining Multiple approximatelythreshold that cues4 degrees it to flower Celsius, at thea certain Sources of Uncertainty vulnerability of the species may range from very little and unlikely to fairly Understanding the degree of uncertainty great and likely. in each of the components that make up a vulnerability assessment is useful in The key to combining multiple sources understanding the overall vulnerability of of uncertainty is to identify interactions a system to climate change. However, one between the different components, such of the major goals of doing a vulnerability as how temperature and precipitation assessment should be to combine the interact to affect soil moisture and river different components of the assessment through conceptual models, diagrams, and in a way that provides an understanding flows. This can be done qualitatively of the range of vulnerability given the narratives, or more quantitatively through uncertainties in each of the components. algorithms. The method used for combining Being able to combine these multiple scientific models and computational sources of uncertainty is really the glue uncertainty should be chosen based on that brings vulnerability assessments the methods used to assess uncertainty together into a synthetic product that of the components (e.g., qualitative vs. can be used for decision-making and quantitative), the degree of understanding adaptation planning. about the interactions between the components and the resources available Combining the various sources of for combining the data (e.g., technological uncertainty is important because the capacity and budget). uncertainties may interact to magnify
Addressing Uncertainty 73 VI. Using Vulnerability Assessment Results
Vulnerability expected winners and losers under altered climate conditions. The vulnerability Assessment Outputs assessment case studies provided in Chapter VII offer examples of a wide range onducting a vulnerability of outputs from vulnerability assessments. assessment is not an endpoint. These assessments were conducted for C Vulnerability assessments are locations throughout the United States, an intermediate step, and results provide at various spatial scales, and for targets information used to develop adaptation strategies and inform management habitat types. ranging from species to broadly defined from vulnerability assessments are Vulnerability assessments should also determinedplanning. The by specific factors usessuch ofas results the selection of conservation targets, management with their relative vulnerability ranking provide a confidence value associated scale, tolerance for risk, and management as an output. This facilitates a transparent approaches. Two common outputs from consideration of uncertainty in subsequent vulnerability assessments are a ranking of conservation and management decision- the relative vulnerability of target species making. However, one must be careful or habitats, and an assessment of the not to allow uncertainty to preclude consideration of climate impacts on or habitats. species and habitats. As discussed specific factors that pose threats to species previously, uncertainty is inherent in all Relative vulnerability rankings may be projections, whether or not climate is displayed in tables or spatially through one of the factors considered in making maps. These rankings can range from the projections. In fact, simply because expected complete loss of the target, to a climate change is having and will continue ranking of greatly increased abundance to have major impacts on species and or distribution. In this way, conducting a habitats, a greater degree of certainty is vulnerability assessment helps to identify inherent in assessments accounting for climate change than those that do not Vulnerability assessments on species and habitats. Although the provide information to develop magnituderecognize the of climateinfluence change’s of climate impacts change on adaptation strategies and inform species and habitats may be uncertain, it is important to understand that climate management planning. change vulnerability assessments can, at
Lead authors: Doug Inkley, Molly Cross, Jennie Hoffman, and John O’Leary.
74 Scanning the Conservation Horizon Using Assessment Results a minimum, reveal information about the it easier to integrate the results of the direction of species and habitat changes in climate change vulnerability assessment response to climate change (e.g., whether into the existing planning framework used populations or habitat area are likely to by agencies. increase or decrease). Vulnerability assessments can be used Another important output is the wildlife conservation and management, contributing to a species’ or habitat’s includingto help inform selecting many which aspects species of fish or and vulnerability.identification Whenof the assessingspecific factors vulnerability, habitats should be the focus of conservation non-climate factors contributing to efforts, identifying priority areas for vulnerability (e.g., habitat fragmentation; land acquisition, informing management the extent of watershed covered by decision-making, and directing monitoring impervious surfaces; impacts from efforts. Box 6.1 describes an example invasive plants and animals; pest and of how the Massachusetts Division of pathogen outbreaks; and impacts from Fisheries and Wildlife is integrating water withdrawals and aquifer depletion) habitat vulnerability assessment into its land protection prioritization process and management decision-making. climateshould be and included. non-climate Identification sources, as of well Similarly, Case Study 6 describes a two- asspecific their interactions,vulnerability isfactors, key to fromdeveloping both pronged assessment approach that builds potential adaptation strategies. Knowing on vulnerability assessment to identify the factors or combination of factors that potential management options. make a species or habitat vulnerable allows or conservation strategies that can help reducemanagers those to developvulnerabilities. specific management
Informing Existing Planning Efforts
Most state, federal, and tribal natural resource agencies, as well as non- governmental organizations engaged in natural resource protection, are guided by well-established planning processes. A useful aspect of vulnerability assessments is that they can help inform the management process regardless of the administrative structure, function, and operating procedures of different management agencies. Carefully choosing Eric Engbretson/USFWS the targets of the vulnerability assessment and methodology of assessment will make
Using Assessment Results 75 Box 6.1. Informing Land Protection Priorities in Massachusetts
The Massachusetts Division of Fisheries and Wildlife recently completed a Habitat Vulnerability Assessment (HVA) to inform their planning processes (see Case Study 4 in Chapter VII). The HVA was performed by an expert panel that determined the relative vulnerability to climate change for 20 key habitat types, as well as a confidence score for each habitat evaluated and an identification of the various factors contributing to a habitat’s vulnerability ranking and confidence score. These results have been added as another factor for consideration in agency management, acquisition, and research and monitoring programs (Manomet Center for Conservation Sciences and MDFW 2010c). Potential management responses being considered include the following:
1. Promote resistance and resilience. The ability of a system or species to resist adverse climate change impacts will depend largely on its intrinsic resistance to the stressors and its resilience—its ability to recover from stress. The resilience of many species and systems has already been compromised by anthropogenic stressors and they are now in a weakened state. While there are no guarantees that increasing the resilience of these resources will safeguard them under climate change, it is certain that their current lack of resilience will render them vulnerable. Four main solutions to promoting resistance and resilience have been proposed:
• Mitigating the effects of non-climate stressors • Conserving existing biodiversity, ecological functions, and high-quality habitats • Restoring degraded habitats • Managing habitats for ecological function
2. Implement landscape-level planning. One of the main impacts of climate change will be to increase the likelihood and magnitude of shifts in the distributions of species, habitats, and ecosystems. A landscape-level planning focus will be necessary to accommodate this. Specifically, it will be important to take such a view to: 1) Identify and preserve movement corridors; 2) Improve habitat connectivity to facilitate movement of displaced organisms; and 3) Improve buffering to safeguard core, high-quality habitats.
3. Promote effective on-the-ground management of sites and habitats. Adaptation goals need to be translated into effective on-the-ground management actions that will strengthen the resistance and resilience of sites, habitats, and species under a changing climate. Specifically, site managers and biologists need to focus on two primary management goals—managing resistance and resilience, and managing change.
4. Promote and implement “climate-smart” regulation. Some of the conservation regulations that have served well in the past may not be as effective under climate change. For example, regulations that prohibit the management and manipulation of resources and habitats might not be optimal at a time when a changing climate is forcing responses in resources. In such cases it may be necessary to introduce a degree of management flexibility into these existing regulations.
76 Scanning the Conservation Horizon Using Assessment Results Selecting Conservation Setting Land Targets Protection Priorities
Climate change vulnerability information Among the most powerful strategies for can be integrated into processes aimed the long-term conservation of biodiversity at identifying species or habitats most is establishment of networks of protected in need of conservation attention, such areas that represent the full range of a as efforts to identify and prioritize SGCN region’s species and ecosystems, and for State Wildlife Action Plans. In some include multiple, robust examples of each cases, a consideration of climate change type. These principles of representation, vulnerability will cause agencies to add resiliency, and redundancy are at the core species to their SGCN lists or alter the of many comprehensive conservation level of priority of an existing SGCN. The planning and land protection efforts same may apply to habitats that agencies (Shaffer and Stein 2000; Margules and consider to be of high conservation need Pressey 2000; Scott et al. 2001). Climate or priority. change vulnerability assessments can help aid such planning efforts by augmenting While the relative vulnerability rankings knowledge of the current distribution an assessment generates may help and status of species and ecosystems managers understand which species with projections of the possible future are more and less vulnerable, it will not conditions and locations. Combining the dictate whether to focus attention on the results of species assessments may reveal most vulnerable, the least vulnerable, or landscape areas likely to have relatively something inbetween. This emphasizes the high or low species diversity or important fact that a vulnerability assessment is not habitat for species of management an endpoint, but a source of information concern. In either case, the results can be that can be incorporated into planning and used to identify priority areas for areas decision-making. Furthermore, because for protection based on the principles of vulnerability assessments should elucidate representation, resiliency, and redundancy. Land protection strategies can therefore species’ or habitat’s vulnerability, it can take into account not only existing values helpthe specific managers factors identify that optionscontribute for toreducing a and conditions, but also the likely value of that vulnerability through management and conservation actions. In some cases there specific areas under a changing climate. may be practical management options, A vulnerability assessment is not an but in other cases the factors leading to endpoint, but a source of information not feasible to address. This is an important to incorporate into planning and considerationvulnerability may in selecting be very difficultconservation or simply targets and objectives. decision-making.
Using Assessment Results 77 Informing Management Furthermore, if some of these factors Decisions are directly being managed in order to provide appropriate conditions for priority species or habitats, the degree of success in Vulnerability assessments ideally creating these conditions will come to light incorporate uncertainty about climate through their monitoring. change and about a system’s response to it. Adaptation planning ideally evaluates The potential factors to monitor that have management options across that range major effects on species or habitats will of uncertainty. As previously discussed, no doubt vary widely depending upon vulnerability assessments can help you the species and habitat type, and even to evaluate whether your existing goals, vary within the range of given species objectives, and targets are still appropriate and habitats. It will be important to in a changing climate, or to develop new ensure that the factors being measured goals, objectives, and targets (Millsap et al. provide useful information. Where key conservation goals, objectives, and targets, habitats, monitoring of these thresholds the1990). next Having step is identified to decide managementwhich actions or will isthresholds important, are especially identified if for these species factors or best achieve those aims. are themselves being addressed in Directing Monitoring Efforts management.
There will very likely be many instances Multiple management objectives and wherein the major factors affecting a multiple factors affecting species of species are not known at the time of the management concern, combined with vulnerability assessment, or there is a limited resources, necessitate that high degree of uncertainty associated monitoring programs to assess the success/ with the results. In these situations, the failure of management objectives be vulnerability assessments should help designed to yield useful information in reveal those species for which further a cost-effective manner. In some cases, research is necessary to identify key monitoring may be of the status or health of the target species or habitats, which vulnerability assessment results (Williams should help determine the effectiveness of etfactors al. 2008; and IPCCincrease 2005). confidence Assuming in successthe various management strategies. In other of this research, vulnerability assessments cases, it may be important to monitor major could be conducted again for these species factors affecting the status of species or and habitats and appropriate factors habitats. Because the process of assessing used in monitoring management results. vulnerability requires determination of the major factors affecting the status of how to manage for a species and which habitats and species, one can return to factorsUntil then, to manipulate it will be difficult and monitor to know for the vulnerability assessment to inform management purposes. decisions about the most appropriate factors to monitor. Monitoring of these factors should provide useful information about species or habitat status.
78 Scanning the Conservation Horizon Using Assessment Results Dealing with managers make adaptation planning decisions under uncertainty are adaptive Uncertainty in management and scenario planning. Adaptation Planning Adaptive Management As highlighted in Chapter V, resource managers often must make conservation adaptive management as “a systematic decisions under uncertainty, particularly approachThe U.S. Department for improving of the resource Interior defines where information about future conditions management by learning from management must be considered. Some management outcomes,” based on principles laid out by responses will be effective in meeting the National Research Council (Williams conservation goals under a range of et al. 2007; NRC 2004). The overarching potential climate futures, while others purpose of adaptive management is to enable natural resource managers and conditions (Lawler et other relevant decision- al.may 2010). need Whento be tailored future to more specific makers to deal with conditions are fairly Instead of striving for uncertainty about future certain, it makes sense to the single best outcome, conditions by supporting ask “Which actions will it may make more sense the development of produce the single best conservation projects outcome?” When there to ask “which actions based on existing will give the best chance information and then about future conditions, answeringis significant that uncertainty question of some acceptable modify their management becomes increasingly outcome.” activitiesproviding to the improve flexibility their to effectiveness as new information becomes depends on which future comes to pass. In available. It is a concept that has been suchdifficult situations because it the may answer make more sense to around for many years, and it has often ask “Which actions will give me the best chance of some acceptable outcome?” This management plans. Salafsky et al. (2001) approach is called robust decision-making, identifybeen identified a series as of a steps priority for adaptivein resource and it is essentially a bet-hedging strategy. management in conservation: Rather than maximizing the chance of the single best outcome, it seeks to maximize Start: Establish a clear and common the likelihood of an acceptable outcome. purpose While this approach may initially seem at odds with the mandate to make decisions • Step A: Design an explicit model of your based on “the best available science,” it is system not. If the best available science is telling you that there are important uncertainties • Step B: Develop a management plan that that will affect your management success, maximizes results and learning then taking a robust approach is in fact a decision based on the best available • Step C: Develop a monitoring plan to test science. Two tools that can help resource your assumptions •
Using Assessment Results 79 Step D: Implement your management and consider a broad range of options, and monitoring plans appropriate responses to the array of • future scenarios, and what management Step E: Analyze data and communicate mechanisms you can put in place that will results allow you maximum likelihood of success • Iterate: Use results to adapt and learn future scenarios. and flexibility given the array of possible Adaptive• management may be particularly Scenarios, at their simplest, are useful in cases where immediate action descriptions of some plausible future. is required to address short-term and/ They are not predictions or forecasts, and or potentially catastrophic long-term consequences, such as the collapse of Adaptive Scenario High important ecosystem services, or where Management Planning management actions are likely to have “no Optimal Hedging Low Corell 2009). Uncertainty Control regrets” near-term benefits (Ojima and Controllable Uncontrollable It is important to recognize, however, that effective adaptive management can Controllability Figure 6.1. A framework for management under different levels of uncertainty be difficult for several reasons, including (Peterson et al. 2003). conservationinsufficient long-term and management monitoring goals, politicalresources, and unclear institutional or conflicting resistance to scenario planners make no assumptions changing management practices, and/or about which scenario is most likely (if you inability to control a particular outcome knew which was most likely, you would through management (Johnson 1999). not need scenario planning). Scenarios can be quantitative or qualitative, and Scenario-Based Management Planning they may include a complex web of interconnected problems or focus on a Another framework for robust decision- simple subset of the issues. Which is more making, or for decision-making under appropriate depends on the goal of the uncertainty in general, is scenario planning. scenario planning exercise and available Just as the use of multiple climate change information. Qualitative, exploratory scenarios can help address inherent scenarios may help to set the stage for uncertainty in assessing vulnerability, the development of quantitative, targeted they also can provide a useful framework scenarios by stimulating creative thinking for informing possible adaptation and deepening managers’ and planners’ options, particularly in cases where the understanding of their system. In a levels of uncertainty about potential similar fashion, exploratory scenarios that future conditions are especially high and include a range of complexity may help to uncontrollable (Peterson et al. 2003) (see identify those elements on which it is most Figure 6.1). The goal here is to identify important to focus.
80 Scanning the Conservation Horizon Using Assessment Results Scenario planning exercises typically use
Scenarios are created for the particular scenarioaround three planning to five exercise, different and scenarios. will ideally: (1) bracket the range of plausible futures, and (2) highlight those elements of uncertainty most important to management and planning. Having developed the scenarios, managers and planners then brainstorm possible management options and look at the performance of those options across all scenarios. Are there management approaches that are effective in all scenarios? Are there management options that are highly effective in one but disastrous in others? As you go through this exercise, you can highlight areas where uncertainty about climate change or the system’s response to it is more or less important. For instance, if a particular management action is best regardless of future rainfall, decreasing uncertainty in rainfall projections would not be particularly useful. If, on the other hand, rainfall timing or intensity is the single biggest determinant of which management action is best, then you would want to focus on reducing uncertainty around those projections.
Scenario planning provides multiple Lindsay Baronoski particular decisions in uncertain Not considering climate change in conditions,benefits. It notbut onlyincreases helps thewith more making general ability of planners and managers to cope management is akin to traveling in with uncertainty. It also facilitates the unknown territory without design of monitoring programs that target key elements of uncertainty, be they a map—one is not likely to uncertainty about climatic change, system arrive at the desired destination. responses to that change, or the effect of particular management actions.
Using Assessment Results 81 Matt Greene Looking Ahead and habitats, regardless of what those factors are. This comprehensive nature of vulnerability assessments makes them all The development of vulnerability the more important as a tool for informing assessments has resulted from concern the development and implementation of about the pervasive impacts of climate management objectives. change across the landscape. With so many species and habitats likely to be affected, Regional vulnerability assessments, such it is critical that managers know the likely status of species and habitats in a changing (Case Study 7), will provide information climate. Not considering climate change usefulas that to underway different in agencies the Pacific across Northwest the in management is akin to traveling in areas. Regional collaboration across unknown territory without a map—one several states to conduct vulnerability is not likely to arrive at the desired destination or result (Lawler et al. 2010). in a time of distressed state wildlife agency budgetsassessments and maymay alsobe economically foster multi-state efficient relationships (AFWA 2009). Furthermore, vulnerability assessments is that they are An added benefit of conducting in light of the landscape-scale impact of climate change, increased collaboration just climate change. Properly executed, vulnerabilitynot specific to assessments assessing vulnerability should to species and habitat ranges move across account for the factors affecting species theamong landscape. states is likely to be beneficial as
82 Scanning the Conservation Horizon Case Studies VII. Case Studies
number of climate change The sixth study applied an integrated vulnerability assessments climate change assessment and adaptation A for species and ecosystems have framework in the Four Corners region of been completed or are currently underway the Southwest, building on the completion across the country. The following seven of a state-wide vulnerability assessment case studies provide examples of some and two adaptation-oriented workshops of the different assessment approaches for natural resource managers in New described in this guide and demonstrate Mexico. The assessment entailed evaluation the considerable variability that is possible of the level of climate exposure (based on in terms of assessment scale and scope. regionally downscaled data) in relation to
in the four states’ Wildlife Action Plans andexisting ecoregional conservation assessments. priorities This identified usersThe first with four a useful studies tool are to broader, compare “coarse enabled managers to prioritize vulnerable relativefilter” approaches, vulnerability which across can a provide range of landscapes for adaptation action. targets and at varying degrees of detail. These approaches are based primarily on static attributes of species, habitats, and most ambitious assessment described in ecosystems, and they do not involve the thisThe guide.final case The study assessment is the broadest encompasses and direct use of dynamic simulation models a very large spatial scale and a broad (although Case Studies 1 and 3 incorporate range of habitats and species. The study use of a tool for downscaling relevant incorporates elements of other approaches, climate data). Refuges, state agencies, or protected areas that are well staffed may dynamic climate projections, and climate– have the capacity to conduct these types of spaceincluding niche detailed models species-specific to identify vulnerable data, assessments primarily with existing staff. species, biodiversity “hotspots” at risk, and habitats at risk to climate-induced changes. Table 7.1 provides a general summary of habitat-based assessment and a subsequent each of these case studies. assessmentThe fifth case of study, associated which species, highlights involved a the use of a habitat response model to project the impacts of sea-level rise on coastal wetland communities, based on existing scenarios for sea-level rise. Approaches such as this often require the use of expertise that may not normally be available at one site but that may be accessible through partnerships with other agencies, organizations, or individuals.
Case Studies 83 Table 7.1: Summary of Case Studies 1. Nature- 2. EPA Endan- 3. Species 4. State- 5. Coastal 6. Integrated 7. Pacific Serve Nevada gered Species Assessment level Habitat Habitats and Framework Northwest Species Framework for the Middle Assessment Species for the Four Assessment Assessment Rio Grande for Mass. Corners Location Nevada, National New Mexico, Massachusetts, Chesapeake Southwest, Pacific and Extent statewide regional statewide Bay Region Four Corners Northwest, (two studies) region regional Status In progress Completed Completed Completed Completed Phase 1 In progress Completed Targets 263 priority Six threatened Terrestrial 20 habitats 5.1: Coastal Species and Species and animal species and endangered vertebrate wetland habitats habitats (invertebrates vertebrate species species habitats identified as and vertebrates) occupying conservation riparian habitats priorities 5.2: Marsh bird species of concern Climate Yes, down- No (used Yes, down- No (used No (used Yes, down- Yes, down- Change scaled climate published scaled climate published published scaled climate scaled climate Models? data based on projections) data based on projections) projections) data based on data based on ClimateWizard ClimateWizard, ClimateWizard multiple model and published simulations projections Other General General General General Habitat and General Climate niche, Models? characterization characterization, characterization, characterization, occupancy characteriza- habitat, and expert opinion expert opinion expert opinion model tion, expert hydrological (SLAMM) opinion models Detail Low Moderate Low Moderate Moderate Low High Work/ Low Moderate Moderate Moderate Low-Moderate Moderate High Time (application time (1 year) 5.1=1 year (2.5 years) (3-4 years) per species = 5.2=4 months 30-45 minutes) Cost $160,000 $60,000 $60,000 $70,000 5.1: $40,000 $200,000 $800,000 5.2: $25,000 Lead B. Young H. Galbraith D. Finch H. Galbraith 5.1: P. Glick C. Enquist J. Lawler
5.2: M. Wilson Citations Young et al. U.S. EPA 2009 USDA Manomet Center 5.1: Glick et Enquist and Lawler et al. (in press) Forest Service, for Conservation al. 2008a, Gori 2008 2009, 2010 Rocky Mountain Sciences and 2008b Research Station MDFW 2010a, 2010 2010b 5.2: Wilson and Watts 2009
84 Scanning the Conservation Horizon Case Studies Case Study 1. NatureServe’s Climate Change Vulnerability Index for Species in Nevada
This case study highlights how an accessible “vulnerability index” can be used to readily assess the impacts of climate change on species of concern at a state-wide scale. The state of Nevada is emerging as a leader in addressing how Mike Peterson climate change alters the way states need to manage species and habitats to maintain By 2008, it was clear to the groups that biodiversity. In 2008, a coalition of Nevada had developed and were responsible for implementing the Plan that an amendment activities to revise their state Wildlife state agencies and nonprofits initiated major changes needed to manage for Action Plan to fully address the effects was necessary to adequately reflect the of climate change. These organizations biodiversity under conditions of rapid are now well advanced in the research climate change. Updating the Plan to and review exercises that will form the account for climate change would also basis of their revisions. This case study position the state to receive federal describes how one member of the coalition, funding that might come from climate the Nevada Natural Heritage Program, change legislation. Additionally, funding is contributing to this effort by using was available through the Division of State the Climate Change Vulnerability Index Land’s Question One Conservation Bond developed by NatureServe to conduct rapid Program. Thus, the partners, including the assessments of the relative vulnerability of Nevada Department of Wildlife, The Nature Conservation Priority Species. Conservancy, Nevada Natural Heritage Program, Lahontan Audubon Society, Purpose and Great Basin Bird Observatory, successfully applied for funding to Although Nevada’s original Wildlife Action develop this amendment.
a stressor to key habitats and species, it Conservation Objective didPlan not (2006) go into identified detail about climate how change climate as warming could cause substantial ecosystem The objective of the project is to broaden change or what management actions are the applicability of Nevada’s Wildlife necessary to stem the loss of wildlife. Action Plan to understand the health of the state’s wildlife, including vulnerability to
Lead authors: Bruce Young, Jennifer Newmark, and Kristin Szabo.
Case Studies 85 climate change, and to prescribe actions to In terms of cost and time, once the conserve wildlife and key habitats before distribution of natural history information they become more rare and costly to on a species is researched and compiled— protect. The Plan develops objectives and NatureServe has already done this for many strategies for the conservation of 27 key species (available at NatureServe Explorer, habitats found in the state. http://www.natureserve.org/explorer)—it can take as little as 30–45 minutes to rank a Assessment Targets species. The cost of this assessment will be approximately $160,000. Nevada Natural Heritage is responsible for assessing the vulnerability of all 263 Assessment Approach Conservation Priority species that were The Nevada partners have divided the tasks Plan. The species include 1 mussel, 74 according to each organization’s strengths: identified in the original Wildlife Action bird, and 49 mammal species. Once this The Department of Wildlife provides assessmentsnail, 40 fish, is 7 completed, amphibian, the 20 outcome reptile, 72 will oversight and management of the project contribute to habitat vulnerability models in• coordination with the other partners. It run by a partner organization. Nevada organizes team meetings and helps Natural Heritage hopes to eventually extend solicit public comments on draft the vulnerability assessment to many more documents. The Department of Wildlife species, including plants and abundantly will also be responsible for interacting distributed species. with the U.S. Fish and Wildlife Service to obtain formal approval for revisions to the Scale and Scope Wildlife Action Plan.
The Nevada assessment is restricted to As stated above, the Nevada Natural a state-wide analysis. Developing future Heritage Program is responsible for regional assessments that examine how •assessing the vulnerability of the individual species may expand into and retreat from Conservation Priority species to climate states would be helpful to provide a more change. comprehensive picture of the complex changes in population size and location The Nature Conservancy will use taking place. Currently, though, existing a modeling approach to understand the funding mechanisms favor state-based vulnerability• of Nevada habitats to climate approaches. The time scale for the species change. vulnerability assessments is mid-century. Mid-century represents a time frame that The Lahontan Audubon Society will is before the major climate models and help facilitate workshops, work with the emissions scenarios begin to have widely •public, perform outreach activities, and divergent predictions, resulting in less help edit document drafts as they are uncertainty than for longer time horizons. written.
86 Scanning the Conservation Horizon Case Studies Climate Change Vulnerability Index
The Great Basin Bird Exposure to Documented Observatory will use climate Climate Change Response envelope• models to estimate how Precipitation bird distributions may shift as a Targeted Temperature Index Score Conservation result of climate change. Like the Efforts species assessments undertaken Sensitivity Critical Factors by Nevada Natural Heritage, the results of these models will Indirect Exposure form input into The Nature Species-speci c Conservancy habitat models. Factors Figure C1.1. Framework for NatureServe’s Climate Change Nevada Natural Heritage had a Vulnerability Index. limited amount of funding and a short time frame for completing Evidence. The Index complements standard its task of reviewing the vulnerability of Increase Likely), and one Insufficient a large number of species. The program conservation status assessments such as elected to use the Climate Change the NatureServe G- and S-rank system Vulnerability Index as a rapid and cost- that contributed to species’ designation as Conservation Priorities in the original Wildlife Action Plan. More information Theefficient Index means separates of completing a species’ this vulnerability task. about the Index as well as the Index itself into two main components: exposure can be found at http://www.natureserve. to climate change within its range and org/climatechange. inherent sensitivity to climate change (Williams et al. 2008) (see Figure C1.1). Biologists from Nevada Natural Heritage Data for these two components take the used distribution and natural history form of downscaled climate predictions information from their databases together across the range of the species within the with climate predictions downloaded assessment area (in this case, the state from the ClimateWizard to complete of Nevada) and scoring of the species assessments for all 263 Conservation against 17 factors related to its anticipated Priority species. Next, they convened a climate change sensitivity, such as dispersal panel of independent biologists familiar with Nevada wildlife to review their work factors addressing exposure and adaptive ability and habitat specificity. Additional were scored for each species. This process capacity, such as natural or anthropogenic and confirm or adjust how the factors barriers to dispersal, as well as observed is currently ongoing and will result in responses to climate change (if available) are also included. These factors are all vulnerability models that will form another sectionfinal assessments of the revised that Wildlife feed into Action habitat Plan. correlates or predictors of vulnerability to climatedocumented change. in theThe scientific outcome literature is one of sixto be possible Index categories: three Vulnerable (Extremely, Highly, and Moderately), two Not Vulnerable (Presumed Stable,
Case Studies 87 Assessment Results aquatic/wetland habitats were the factors that most commonly contributed to Like most of the world, Nevada will vulnerability to climate change. Surprisingly, anticipated land-use changes designed century climate predictions suggest to reduce greenhouse gas emissions warmingexperience of significant2.6 to 3.2 degrees warming. Celsius Mid- as a means to mitigate climate change and slight decreases or increases in (such as solar, wind, and geothermal precipitation in different parts of the projects) are another factor contributing state (Maurer et al. 2007). Results for a to vulnerability for some species due to preliminary assessment of 216 vertebrates associated habitat loss and fragmentation. and mollusks listed as Conservation Priority Good dispersal ability, broad physical species in the Nevada Wildlife Action Plan habitat requirements, migration to broad revealed that the Index sorted taxa into geographical areas or a tendency to shift widely differing levels of vulnerability to distribution in response to environmental climate change (Figure C1.2). Mollusks conditions, and demonstrated adaptation to a broad range of temperatures were the groups, whereas some mammals and factors that most commonly decreased and fish were the most highly vulnerable vulnerability. One noteworthy outcome is expand their ranges in Nevada as the climatefish may warms. increase Demonstrated their abundance adaptation or common species (i.e., NatureServe global that the Index flagged a number of currently to a limited range of precipitation regimes, conservation status rank G4 or G5) such migration to or through a few restricted as the American pika, bighorn sheep, and potentially vulnerable locations and sagebrush vole as vulnerable to or lack of regular distribution shifts in climate change. Thus, conservation status response to environmental conditions, is not a reliable proxy for vulnerability to climate change.
and dependence on specific vulnerable 120
100
80
60
40
Number of Species 20
0
Extremely Highly Moderately Presumed Increase Vulnerable Vulnerable Vulnerable Stable Likely
Figure C1.2. Vulnerability of Nevada Conservation Priority species as calculated by the Climate Change Vulnerability Index.
88 Scanning the Conservation Horizon Case Studies
Uncertainties Outcomes and Next Steps
This assessment addressed uncertainty As review of the assessments of the full in two ways. One source of uncertainty set of Conservation Priority species is concerns the differing projections by completed, The Nature Conservancy is climate models of mid-century temperature assessing the vulnerability of key habitats and precipitation regimes in Nevada. to climate change. Subsequently, the The Nevada Natural Heritage Program partners will examine the Index results for addressed this uncertainty by using each species, including the factors that most an average of an ensemble of 16 global frequently led to species being categorized circulation models as the exposure data for as vulnerable, the habitat model results, the Climate Change Vulnerability Index. The and the bird models produced by the Great results therefore are not tied to any single Basin Bird Observatory to determine the climate model. management strategies necessary to create resilient wildlife populations and mitigate A second source of uncertainty relates to potential impacts to climate change. After how a particular species is scored against receiving comments from the public, the the Index sensitivity factors. Incomplete knowledge about a species’ natural history change amendment to the Plan and submit and how it affects vulnerability for a itpartners to the U.S. will Fish finalize and theWildlife text ofService the climate for particular factor can add uncertainty to formal approval. the overall vulnerability score. The Index allows users to select more than one An amended Wildlife Action Plan will not accomplish its objective unless its this uncertainty. The Index calculates an recommendations are put into practice. overallvulnerability vulnerability value for score each using factor an to average reflect The partners responsible for preparing the of the values assigned for each factor, but climate change amendment also make up also runs a Monte Carlo simulation to the implementation team for the Plan. As explore the probability that the overall part of their work on the team, they will score could change depending on what the work to incorporate the recommendations “true” value might be for each factor scored from the Plan into ongoing private, state, with multiple values. The Index calculates and federal management activities. These actions will take many forms depending information (very high, high, moderate, or on the purview of each agency. The Nevada low)a measure depending of confidence on the percentage in species of Monte Natural Heritage Program will use the Carlo runs that yield the same overall results of the vulnerability assessment to vulnerability score as calculated with the target particularly vulnerable species for averaged data. For the Nevada species, monitoring. Comprehensive information the Monte Carlo simulations revealed that on the locations and biological conditions of vulnerable species will contribute to high or high for 61 percent, moderate for proactive management decisions that 27confidence percent, inand the low Index for 12score percent was very of the species (Young et al., in press). wildlife and development in the future. Forcould example, result in monitoring decreased dataconflicts for a between species
Case Studies 89 that has limited dispersal ability or is Nevada’s approach of having a general encountering barriers to dispersal biologist perform preliminary assessments could provide an early indication of and then inviting specialists to review the when translocations of populations results in a workshop setting proved to should be considered. be successful. The specialists provided a broader interpretation of the criteria, and In conjunction with implementation their individual expertise with particular activities, the partners are participating species went beyond information found with the Heinz Center and the Wildlife in the published literature. The general Habitat Policy Research Program to use biologist provided leadership for the performance measures to monitor the process and ensured that the criteria effectiveness of management actions. were applied consistently across taxa. For priority conservation goals, they Also, an advantage to carrying out the have begun to develop logic models that assessments within the Nevada Natural pictorially describe pathways by which Heritage Program was the easy access to factors can affect a conservation goal. The spatial information about the locations of populations of Conservation Priority can be measured to monitor progress species for calculating exposure. towardmodels eachpoint conservation to specific indicators goal. This that approach allows the partners to gauge Case Study 2. U.S. their success and alter their strategies when needed, before wasting resources on Environmental ineffective actions. Protection Agency’s Challenges and Lessons Learned Threatened and Endangered Species One challenge in developing the Index was deriving and calibrating criteria for diverse Vulnerability plants and animals. Research in the past Framework decade has led to a substantial increase in understanding about the factors that Purpose correlate with climate change vulnerability, but analyses are typically available only This case study provides an overview for selected taxonomic groups. Assembling of a framework developed to assess the an interdisciplinary team to develop, test, relative vulnerabilities of threatened and endangered animal species to climate address how the myriad natural history change and existing stressors. attributesand refine ofthe North Index American proved necessary plants and to animals confer increased or decreased vulnerability to climate change.
Lead authors: Hector Galbraith and Jeff Price.
90 Scanning the Conservation Horizon Case Studies Conservation Objective
Organisms listed as Threatened or Endangered (T&E) under the Endangered extinction due to the adverse effects of Species Act suffer a significant risk of current natural or anthropogenic stressors. Steve Maslowski/USFWS Climate change, either acting alone or by exacerbating the effects of these existing Scale and Scope stressors, may constitute an important new threat for many of these species (Peters This framework can be applied at a range 1992; Schneider and Root 2002). If future of spatial and temporal scales, depending conservation priorities, strategies, tactics, on the habitat range of the focus species as well as the climate change scenarios used. these changing circumstances, there is a needand resource to develop allocations new conservation are to reflect tools. Assessment Approach In particular, tools are needed that integrate the likely effects of both current Framework Structure and climate change stressors to identify the T&E species that may face the The framework for evaluating risks to a greatest increased risk of extinction or T&E species due to climate change and major population reductions, and the other stressors comprises four connected modules and a narrative (Figure C2.1). ecological factors that contribute to these Module 1 categorizes the comparative increasedspecific climatic, risks. physiological, and/or vulnerabilities of T&E species to existing stressors (i.e., not including climate Assessment Targets change). This “baseline” vulnerability is subsequently combined with the For this study, the T&E vulnerability categorization in Module 2 (evaluating framework was tested on six T&E species: vulnerability to climate change) into an bald eagle (now removed from the T&E estimate of overall future vulnerability list), golden-cheeked warbler, salt marsh in Module 3. Module 4 combines harvest mouse, Mount Graham red squirrel, certainty scores from Modules 1 and 2 desert tortoise, and the Lahontan cutthroat into an evaluation of the overall degree trout. These species were selected because of certainty that we can assign to the they are very different in their ecologies, framework predictions. The narrative demographics, status and distribution, that accompanies each species’ evaluation population trends, and susceptibilities to different stressors and, because of these the assigned scores in Modules 1 and 2. details the rationales and justifications for differences, provide an adequate test of the framework. The Narrative. Most categorizations in Modules 1 through 4 will be based largely on the results of literature reviews for each species being evaluated and on expert
Case Studies 91 judgment. The narrative module of the Narratives for the species evaluated in this framework reports the relevant results case study are available in the full (U.S. EPA of those reviews and opinions, and it 2009) publication.
categorization scores in the modules. Thus, Module 1—Evaluating Baseline theincludes primary the aimjustifications of the narratives for the individualis to make Vulnerability. The probable baseline (i.e., the thought process and assumptions that current) vulnerability of the study species result in the scores in Modules 1 through 4 to extinction or major population reduction transparent. is categorized by scoring those elements of its ecology, demographics, and conservation The narratives have three additional important aims: survival or extinction (irrespective of the potentialstatus that effects influence of future the likelihood climate change). of its 1. To identify main sources of uncertainty There are 11 variables included in this and those areas where additional data module: (1) current population size, (2) might reduce uncertainty population trend in the last 50 years, (3) current population trend, (4) range trend 2. To identify and describe the roles of the in last 50 years, (5) current range trend, main stressors (climate and non-climate) (6) current (non-climate) stressors, (7) in the estimate of vulnerability of the likely current stressor future trends, (8) study species individual replacement time, (9) future vulnerability to stochastic events, (10) 3. To qualitatively describe potential future vulnerability to policy/management population responses of the study species changes, and (11) future vulnerability to to the addition of climate change to the natural stressors. Each of these variables already existing stressors, and any resulting is assigned a numerical score (e.g., 1 for change in extinction risk current population size of <100 vs. 6 for current population size of >50,000),
Narrative individual scores are then combined inreflecting Module their1 into ordinal one of rankings.four baseline These vulnerability rankings: critically vulnerable Module 1 Module 2 (Vb1) for those with a score of less than Baseline Vulnerability Climate Change 18; highly vulnerable (Vb2) for a score of Vulnerability 18–25; less vulnerable (Vb3) for a score of 26–33; and least vulnerable (Vb4) for a Module 3 score greater than 33. Overall Vulnerability
Module 2—Evaluating Vulnerability Module 4 to Climate Change. In this module, the Con dence Evaluation likely vulnerability of a species to future Figure C2.1. Framework for T&E Species Vulnerability climate change is assessed and categorized Assessment. by scoring those elements of its physiology, life history, and ecology that will likely be
92 Scanning the Conservation Horizon Case Studies important determinants of its responses. more), medium (approximate probability This is based on determining ordinal between 30 and 70 percent), or low (less rankings for 10 Module 2 variables: (1) than approximately 30 percent). These physiological vulnerability to temperature, qualitative scores correspond to numeric (2) physiological vulnerability to scores of 3, 2, and 1, respectively. In precipitation change, (3) vulnerability to Module 4, the best estimate certainty climate change–induced extreme weather scores assigned to each of the variables in events, (4) dispersive capability, (5) degree Modules 1 and 3 are combined into of habitat specialization, (6) likely extent an index of the certainty associated with of habitat loss due to climate change, (7) the overall vulnerability score in Module abilities of habitats to shift at same rate 3. The total minimum score (Modules 1 as species, (8) habitat availability within and 2 combined) is 20, while the maximum new range of species, (9) dependence is 60. The numeric range between the on temporal interrelationships, and (10) two is arbitrarily and approximately dependence on other species. Again, each equally divided into three categories: of these variables is assigned a numerical High, Medium, and Low certainties and score (e.g., 1 for high sensitivity to a temperature increase vs. 4 for a species each species. a final certainty evaluation applied to increase). The numerical scores for Table C2.1. Module 3 - Overall Vulnerability Best Estimate thelikely 10 to variables benefit fromare then temperature combined into an overall evaluation of the Scoring Matrix species’ potential vulnerability Baseline (Module 1) Vulnerability Scores to climate change: critically Climate change (Module 2) Vb1 Vb2 Vb3 Vb4 vulnerable (Vc1); highly vulnerable Vulnerability scores (Vc2); less vulnerable (Vc3); and Vc1 Vo1 Vo1 Vo2 Vo3 least vulnerable (Vc4), likely to Vc2 Vo1 Vo1 Vo2 Vo3 Vc3 Vo1 Vo2 Vo3 Vo4 Vc4 Vo1 Vo2 Vo3 Vo4 Modulebenefit from 3—Evaluating climate change. Overall Vulnerability. In this module, Vc5 Vo2 Vo3 Vo4 Vo4 the “best estimate” scores from Modules 1 and 2 are combined into a Important Framework Attributes matrix to produce an overall best estimate evaluation and score of the species’ Process Transparency. The intended vulnerability to climate change and end result of the framework is to produce important existing stressors (Table C2.1). evaluations of the relative vulnerabilities of T&E species to climate change and Module 4—Certainty Evaluation. The other stressors. However, it is important approximate level of certainty with which that the process and reasoning through each “best estimate” score in Modules which the evaluation was arrived at be well 1, 2, and 3 is categorized separately in documented and transparent. This will be essential in modifying species evaluations (approximate probability of 70 percent or if new data are gathered that cast doubt on the modules. These are codified as high
Case Studies 93 previous assessments. Ensuring Sources of Information and Expert process transparency and documenting Opinion. Some of the scores determined important assumptions is as important in this framework may be based on a component of the framework as quantitative and empirical data (e.g., producing predictive scores. abundance estimates based on actual census data) published in peer-reviewed Framework Precision and Accuracy. The results of a predictive framework will many less well-studied species, it is likely be speculative. Thus, in the absence of thatscientific many or of “gray” the framework literature. scores However, will forbe a posteriori knowledge, this framework based not on actual empirical data, but will provides approximations of species’ ranked comprise rankings based on expert opinion. vulnerabilities. It is not intended that its results be considered completely accurate Some species may benefit from climate or precise estimations of a species’ absolute change. Not all species may be adversely vulnerability—the results should be affected by climate change; it is possible regarded as indications of the comparative vulnerabilities of T&E species. regimes (for example, due to their habitats beingsome mayexpanded, benefit or from to their new competitors climatic or Treatment of Certainty/Uncertainty. predators being adversely affected). It is See below. this reality. essential that an effective framework reflect
Table C2.2. Summary of Species Evaluation Results Species Module 1 Module 2 Climate Module 3 Best Module 3 Module 4 Baseline Scores Change Scores Estimate Scores Alternate Scores Certainty Score Golden-cheeked Vb2 (highly vulnerable Vc1 (critically Vo1 (critically Vo2 (highly) High warbler vulnerable) vulnerable) Bald eagle Vb3 (less vulnerable) Vc3 (less vulnerable) Vo3 (less vulnerable) Vo2, Vo4 High
(highly, least) Salt marsh harvest Vb2 (highly vulnerable Vc2 (highly vulnerable) Vo1 (critically Vo1, Vo2 (critically, Medium mouse vulnerable) highly) Mount Graham red Vb2 (highly vulnerable) Vc2 (highly vulnerable) Vo1 (critically Vo1, Vo2 High squirrel vulnerable)
(critically, highly) Desert tortoise Vb3 (less vulnerable) Vc2 (highly vulnerable Vo2 (highly Vo1, Vo3 Medium vulnerable)
(critically, less) Lahontan cutthroat Vb2 (highly vulnerable) Vc2 (highly vulnerable) Vo1 (critically Vo1, Vo2 (critically, Medium trout vulnerable) highly)
94 Scanning the Conservation Horizon Case Studies Assessment Results of the certainty/uncertainty associated with it. In this framework the degree of The results of running the evaluative certainty is assessed in two ways. First, framework on the six test T&E species is when scoring each module variable, “best shown in Table C2.2. The golden-cheeked estimate” and alternate (possible but warbler, salt marsh harvest mouse, Mount less likely) scores are assigned by those Graham red squirrel, and Lahontan applying the framework, based on the cutthroat trout were categorized as information gleaned in the literature. “critically vulnerable.” The desert tortoise These are intended to capture the range was ranked “highly vulnerable,” and the of responses that may occur, rather than bald eagle (no longer listed as threatened focusing on a single “point estimate” of or endangered, except for the Southwest responses. Second, each individual variable population) was scored “less vulnerable.” score is assigned a ranking certainty Species that are most vulnerable tend evaluation (i.e., high, medium, or low level to be: restricted in their distributions, of certainty). This three-point ranking is small in population size, undergoing population reductions, habitat specialists, for the IPCC Third Assessment Report based on the five-category scale developed and found in habitats that are likely to be (Moss and Schneider 2000) (see Chapter most adversely affected by future climate V). These rankings are then combined change. Conversely, species like the bald into an assessment of the degree of eagle, which are widely distributed, are certainty that should be associated with are considered to be stable or increasing, vulnerability. For the species tested, the the final assessment of the species’ overall scoredflexible least in their vulnerable. habitat preferences Thus, the and greatest uncertainties are associated with predictions of the model are consistent our relatively poor knowledge about the with what might be expected based on potential for direct, physiological effects the ecologies and demographics of the on animal species; relationships between test species. changes in temperature and precipitation regimes and the physiologies and Uncertainties behaviors of animals are apparently poorly understood. Uncertainty is inevitable when attempting to anticipate the effects of future stressors Outcomes and Next Steps on organisms. Such uncertainty may have This framework was developed for the U.S. variability of likely future climates, the physiologicalmany sources, sensitivity including ofthe the specifics species, or of Research and Development. The full Environmental Protection Agency’s Office uncertainty about its demographics, report is currently awaiting publication population dynamics, or habitat ecology, and general release. For information on or about the likely responses of habitats, when this will occur or for a copy of the or critical habitat components, to full report, readers should contact Susan climate change. Any prediction regarding Herrod-Julius at: [email protected]. future vulnerability would be of limited practical value without an evaluation
Case Studies 95 Case Study 3. climate change projections. We describe an application of the scoring tool to terrestrial Species Vulnerability Assessment for the the Middle Rio Grande of New Mexico, and providespecies ina synopsisa specific of geographical the results of region, this Middle Rio Grande, regional assessment. New Mexico Background Summary Land managers need adaptation This case study describes a method for and mitigation strategies to manage scoring terrestrial species that have species within the context of ecosystem potential to be vulnerable to climate responses to a changing climate and change. The assessment tool seeks to varying responses of individual species. synthesize complex information related to The U.S. Forest Service Rocky Mountain projected climate changes into a predictive Research Station (RMRS) in Albuquerque tool for species conservation. The tool was support tools for managers to anticipate designed to aid managers in prioritizing is creating scientifically based decision- species management actions in response to climate-related changes and respond strategically to managing those effects. Vulnerability of species and populations to changes induced by global warming will be assessed using a scoring system being designed by RMRS. This system assigns 1970 - 2006 2041 - 2060 2071 - 2090 scores from synthesized information related to the probability of climate-related population declines due to a number of factors, including: natural disturbances
requirements (e.g., link with seasonal food resources,(e.g., flooding, breeding wildfire); ponds); breeding nonbreeding requirements (e.g., habitat changes, Base data sources: Daly et al. 1994, stopover sites); dispersal potential (e.g., Dominguez et el. 2008 connectivity of habitats, mobility); and exacerbating factors (e.g., rarity, proximity Figure C3.1. Current and future moisture stress in Upper to human populations). Scores for an and Middle Rio Grande. 1970—2006 = trend in moisture individual species are then combined to deficit. Preliminary future forecast: 2041–2060 = departure create an overall prediction of vulnerability (difference) in deficit relative to 1951–2006 baseline; to climate change. 2071–2090 = departure in deficit relative to 1951–2006 baseline (Bosque Working Group 2008).
Lead authors: Deborah M. Finch, Megan M. Friggens, and Karen E. Bagne.
96 Scanning the Conservation Horizon Case Studies The Bosque or riparian forest along the sources. In addition, new information can Middle Rio Grande has high value for be incorporated into the scoring process wildlife but because of competing land as climate modeling becomes more and water uses is also vulnerable to sophisticated and predictions more precise. degradation. Global climate predictions include higher temperatures in the species, those with the highest scores, is Southwest, more variable rainfall, and more oneIdentification step toward of theimplementing most vulnerable an effective drought periods, which are conditions management program. that will exacerbate the present issues (Christensen et al. 2007). Moisture stress In this case study, we summarize results along the Upper and Middle Rio Grande is projected to increase dramatically RMRS 2010) delivered April 2010 to our through this century (Figure C3.1). In sponsors,from a final U.S. report Fish (USDAand Wildlife Forest Service Service addition, human populations in the Region 2 (Bosque Improvement Initiative) region are expected to grow considerably, and the U.S. Forest Service, Washington putting more pressure on natural systems competing for resources. Management application of the tool to species in a actions can often be taken to mitigate Office. The final report describes an impacts, but is most effective when Middle Rio Grande. specific geographical region known as the than reactionary. Land managers for the Conservation Objective scientifically based and anticipatory rather from knowledge related to projected effects Global climate change has the potential ofMiddle climate Rio change Grande on Bosque species would in their benefit area. to affect habitats and species worldwide Depending on actions and magnitude of within a relatively short period of time climate change effects, this strategy will and, in fact, appears to already be altering help conserve biodiversity. ecosystems (reviewed by McCarty 2001; Peñuelas and Filella 2001; Root et al. Purpose 2003). In addition to current conservation challenges such as habitat loss, toxins, and This case study documents a new method exploitation that have long been part of for assessing the relative risk to persistence species management programs, current of individual species under projected climate change is relatively new and its changes in temperature, precipitation, impact is expected to grow. Already a and related climate phenomena (Bagne and Finch 2008). The RMRS assessment at risk to changes in climate in New Mexico. consists of a scoring system focused on Withnumber increasing of species droughts have been projected, identified as simple predictive criteria for terrestrial populations of those species sensitive to drought conditions such as white-tailed designed to be applied by managers. Relativevertebrate vulnerability species and is was assessed specifically through and Goat Peak pika are likely to decline the use of scores generated for individual (Enquistptarmigan, and southwestern Gori 2008). Particularwillow flycatcher, habitat types, such as alpine tundra, are expected system that allows the user to incorporate to decline along with the species dependent dataspecies. and The information assessment from is a flexiblevariety of on them (Walther et al. 2005).
Case Studies 97 Assessment Targets Field Guide to the Plants and Animals of the were initially identified using the This assessment focused on the terrestrial Middle Rio Grande Bosque (Cartron et al. vertebrate species occupying riparian 2008). Archival Middle Rio Grande data habitats (known locally as “the Bosque,” on birds, amphibians, reptiles, and bats a Spanish word meaning “forest”) along (from D. M. Finch, RMRS) were also used the Middle Rio Grande in New Mexico. to identify species for scoring. We removed This area is bounded by Elephant Butte species that were not resident within the Reservoir to the south and Cochiti Dam Middle Rio Grande Bosque for at least part to the north. We created a future climate of the year, as well as those species that scenario for the Middle Rio Grande Bosque occur primarily in upland habitats. Rare as well as upland areas surrounding the species were included only if they were Bosque. Vertebrate species for the region known to breed within the Bosque. Species that used the riparian corridor solely for migration, or otherwise had an intermittent or transient presence within the Bosque, were not included in the assessment list. Scale and Scope
Vulnerability of species and populations to changes induced by global warming was assessed using a scoring system similar to that developed for identifying avian populations at risk by Partners in Flight (Panjabi et al. 2005). A detailed account of the scoring for each species resulted. Scores are adjustable as new information arises. The basic process was as follows:
Compile information on vertebrate species of the Middle Rio Grande Bosque • Compile information on projected climate change effects for the Middle Rio Grande• Basin
In consultation with experts, create scores for individual species for each variable• Figure C3.2. Major physiographic and hydrologic Create composite scores and prioritize features of the Middle Rio Grande Basin. species by vulnerability •
98 Scanning the Conservation Horizon Case Studies The assessment for this case study The assessment comprises a series of evaluated the full array of terrestrial questions that focus around variables vertebrate species inhabiting riparian woodlands along the Middle Rio Grande in impacts of climate change on the ability New Mexico. The geographical boundaries ofor individualtraits believed species to reflect to survive the potential and of the Middle Rio Grande are demarcated reproduce. Each question is accompanied as the stretch of the river from Cochiti Dam by a series of potential responses that, in (at its northernmost edge) downstream turn, are associated with a simple score 160 miles to San Marcial and Elephant of 1 (vulnerable), 0 (neutral or unknown), Butte Reservoir, and the Bosque portion of it extends south to San Acacia (Figure C3.2). climate change effects on species were The Middle Rio Grande Valley includes four or −1 (resilience). Variables related to New Mexico counties (Sandoval, Bernalillo, factors: habitat, physiology, phenology, Valencia, and Socorro) and six Indian andidentified biotic frominteractions. four broad categories or pueblos (Cochiti, Santo Domingo, San Felipe, Santa Ana, Sandia, and Isleta). In addition, we considered whether traits exhibited three primary functions of good Assessment Approach scoring variables for assessment designs (Beissinger et al. 2000): (1) repeatability, The system assigns scores from (2) relation to quantitative values, and (3) synthesized information related to independence from other scoring variables. probability of population declines as Our adherence to these criteria, however, related to anticipated changes that was limited because quantitative data directly or indirectly result from climate currently available for species’ response change. Possible scoring variables are: to climate change are rare and inherently exposure to natural disturbance (e.g., imprecise for future projections. A separate
(e.g., link with seasonal food resources, quantity and quality of data used to score a temperatureflooding, wildfire); range, breeding breeding requirements ponds); speciesscore for is uncertainly also included. that reflects the nonbreeding requirements (e.g., habitat changes, stopover sites); dispersal potential Two types of scores are provided for (e.g., geographic barriers, mobility); and different purposes: categorical scores exacerbating factors (e.g., specialist species, and overall scores. Overall scores can physiological limitations, competition with be used to rank species and identify the invasive species). Scores in each category most vulnerable or resilient species. are assigned to individual species on an Alternatively, overall scores can also be ordinal scale from lowest to highest risk. used to categorize species into broad For example, when looking at natural vulnerability levels. Because not all species disturbance a species may have a high score attributes translate to equal risk from climate change, composite scores allow are predicted to increase. A species may for prioritization of species vulnerabilities haveif it negatively a low score responds if it responds to wildfires, positively which to for complex information. Categorical disturbances that are expected to increase or if its habitat is not threatened by these areas or traits related to vulnerability. sources of disturbance. scores can be used to identify specific
Case Studies 99 Scores for individual criteria can identify Results vulnerable and resilient characteristics that could aid in single-species management The southwestern United States is expected to experience relatively large temperature decisions. Identified areas of vulnerability can be used to target the most effective region include an increase in the severity or resilience and their relative influence management actions (i.e., creation of andincreases duration and of specific drought predictions periods, more for the heat corridors, land acquisition, captive waves, greater variation in precipitation, breeding). In addition, scores can be adjusted and categories added as effects and increased evapotranspiration and of climate change manifest or threats are salinizationincreased wildfires (Easterling and 2000;insect Fieldoutbreaks, et al.
of species for management will have to greatest consequence will be the impact identified by new research. Prioritization additionally consider issues such as current of2007; these Garfin changes and onLenart southwestern 2007). Perhaps water of vulnerability, economic feasibility, and regulations. Managers can also incorporate expected to experience a change in seasonal their own knowledge and local issues into resources. Specifically, the Southwest is considering species prioritization. (Seagerflood regimes, et al. 2007). reduced snowpack, and an Middle Rio Grande Bosque Climate overall reduction in river and stream flows and Habitat Projections We predict there will be less open water, shorter duration for ephemeral ponds, and We gathered information on projections a decline in wetland habitats in the Middle for climate and related phenomena as well Rio Grande Bosque. These changes will as disturbances and vegetation types to lead to a general loss of riparian vegetation use for our assessment of the Middle Rio and a narrowing of the riparian corridor. Grande species. To maintain relevance to In addition, invasive tamarisk species are management planning, we used projections likely to increase to the detriment of native for a period 20 to 50 years in the future. We used ClimateWizard to estimate changes projections for 2030 showed decreases in precipitation and temperatures for the incottonwood Great Basin species. conifer Specific woodlands vegetation (from region. We used vegetation projections 10 percent to 1 percent), semi-desert created by Rehnfeldt et al. (2006) to grasslands (38 percent to 25 percent), estimate changes in area and distribution and a complete loss of Plains grasslands of major vegetation types (Brown et (estimated to comprise 52 percent of the current habitat). Chihuahuan desert scrub predictions regarding changes to the was predicted to increase from 0 percent to al. 1998). We also outlined specific riparian and upland habitat of the region 74 percent. using additional information from primary literature sources. Finally, we considered Amphibians. Of the nine species of the effect of extreme weather conditions and disturbances, which although more found to be vulnerable to climate change. Theamphibians most vulnerable assessed, species five species was the are more critical to wildlife populations than western chorus frog; three species had averagedifficult changes.to accurately project, may be
100 Scanning the Conservation Horizon Case Studies Rick Lewis neutral scores; and only one, the American Fourteen species had neutral scores bullfrog, appears to be somewhat resilient to climate change effects. Species with warming trends. The southwestern water-dependant larval or adult life stages and five species may benefit from future were most vulnerable to future climate endangered), the western yellow-billed projections. The three species of spadefoot cuckoowillow flycatcher(a candidate (federally for federal listed listing), as toads and the Great Plains toad tended and the common yellowthroat depend toward lower scores because we predicted on riparian habitat and were among the that the scrub and grassland habitats used most vulnerable to potential population by these species would increase in the declines under future climate projections. Middle Rio Grande region. The primary habitat of these species is expected to decline, they are sensitive to Reptiles. Nineteen of 29 reptiles are heat, and they rely on climate-driven cues vulnerable to climate change in the Middle and/or resource pulses, which are likely to Rio Grande Valley. Five species had high change under future scenarios. The three vulnerability scores. The Great Plains most resilient species, spotted towhee, skink was the most vulnerable species, while the common kingsnake was the least are habitat generalists with a good capacity vulnerable. In general, species that have a tohouse respond finch, to and resource brown-headed variation. cowbird, some level of vulnerability to the future Most of the bird species we assessed that expectedspecific reliance changes. on riparian habitat show forage aerially on insects obtained positive
Birds. Of the 40 species of birds assessed, (cliff swallow, barn swallow, ash-throated 25 species (51 percent) had scores overall scores, reflecting vulnerability and bank swallow). An exception to this trendflycatcher, was thenorthern eastern rough-winged bluebird. The swallow, three wasreflecting ranked an as overall the most vulnerability vulnerable to and climate species of woodpeckers had positive scores. thechange. bank The swallow southwestern was the leastwillow vulnerable. flycatcher Of the four species of raptors, three had
Case Studies 101 Uncertainties
All species assessments are prone to errors relating to uncertainties regarding species biology. We found considerable variation in the data available for each species under review. To account for this, we include an uncertainty score. For each of the categories, Habitat, Physiology, Phenology, and Biotic Interaction, we score a species based on the quality and quantity Bruce Stein of data that was available for completing the assessment question. These scores are positive overall scores; western screech then added at the end of the assessment for owl was the most vulnerable, followed by an overall uncertainty score. This score is Cooper’s hawk and great horned owl. By useful not only for estimating the potential contrast, the score for American kestrel for errors in our assumptions regarding was negative.
areas that are in need of further monitoring Mammals. Thirty-six mammals were orspecies research vulnerability efforts. but clearly identifies assessed for their potential vulnerability to climate change in the study region. The Assessments of species vulnerability New Mexico meadow jumping mouse to climate change also have sources appears most vulnerable, while the desert shrew was the least vulnerable to future predictions based on future climate expected changes. Sixteen percent of the scenarios.of uncertainty First, specific uncertainties to the aim exist of mammal species were vulnerable to climate regarding climate projections produced change effects, 44 percent appear to be by climate models. To reduce the impact only slightly impacted, and 5 percent may of individual error attributed to variations in a single model, we used temperature general, species that appear most at risk ranges and precipitation estimates that givenbenefit future from climatefuture projected change predictions changes. In were produced in ensemble models, which are animals with a high reliance on average values from several individual riparian areas (the New Mexico meadow models. Vegetation projections and other jumping mouse and beaver), require dense sources of data regarding future climate conditions are also prone to errors related (the woodrat), or are at risk of mismatch to methodological procedures. We used betweenvegetation critical or specific resources vegetation and breeding features most of these tools to gain a perspective (the hoary bat and black bear). Species not
climate changes tended to be opportunistic projections,on the trends we of relied change less rather on subtle than define breedersexpected (jackrabbit,to be overly desertinfluenced shrew), by future with changesdefinitive (slight future loss scenarios. or shift Forin certain vegetation a wide diversity of habitat associations habitat types), than on projections of including habitats that are expected to severe change (total loss of habitat). We increase in the future.
102 Scanning the Conservation Horizon Case Studies also considered predicted loss of vegetation tool that will be added to an appropriate type to be a robust result, whereas Website such as the U.S. Forest Service estimates of habitat shifts or invasion Climate Change Resource Center Website are more tenuous given the greater (http://www.fs.fed.us/ccrc/). unpredictability of potential plant invasion and establishment and the time lag under Case Study 4. which these transitions will occur. Vulnerability of Finally, there is uncertainty regarding the Massachusetts Fish realized response of species to climate change. We have attempted to consider a and Wildlife Habitats comprehensive suite of traits that allow to Climate Change us to gauge species capacity to tolerate greater variation in resources, higher Purpose and temperatures, and habitat changes. We have also attempted to account for species Conservation Objective interactions within the framework of our tool. Though these efforts have improved This case study describes the use of expert the applicability of our tool, it is likely that elicitation to assess the vulnerability of there remain unpredictable consequences habitats at a state-wide scale. Funded by of climate change for species. a grant from the Wildlife Conservation Society, Manomet Center for Conservation Outcomes and Next Steps Sciences began working in early 2008 with the Massachusetts Division of Fisheries and We will take these results and identify Wildlife and other partners, including The management (adaptation) strategies and Nature Conservancy, to make the state’s actions for terrestrial species in the Middle existing Wildlife Action Plan “climate- Rio Grande Basin. We have also expanded smart.” We are presenting the results of the application of the vulnerability assessment scoring tool to other locations in the American Southwest, including scoring of endangered species on the Coronado National Forest, southeastern Arizona, and on the military bases Fort Huachuca and Barry Goldwater Range. Manuscripts that describe the scoring tool in detail and a Rocky Mountain Research Station General Technical Report that provides the full results of the Middle Rio Grande vulnerability assessment are in review. We are also beginning the process of developing a Web-based version of the J&K Hollingsworth/USFWS
Lead authors: Hector Galbraith and John O’Leary.
Case Studies 103 It is important to note that, when we began Table C4.1. Habitat Types Evaluated this study, our assumption was that climate Forested habitats change was a new and separate issue Spruce-fir forest apart from all of the current stressors that Northern hardwood forest affect the resources we are responsible for Southern/central hardwood forest Pitch pine-scrub oak forest Wildlife Action Plan. Once we realized that climateand that change we had impacts already were identified not a inseparate the Freshwater aquatic habitats issue but would interact with existing Cold water Rivers and Streams stressors and could be thought of as an Large cold water lakes exacerbating factor to current stressors, Smaller cold water lakes and ponds it made it easier to understand how the Warm water ponds, lakes and rivers assessment process could be used to look Cold water kettle ponds at the combined impacts from current Connecticut and Merrimack mainstems stressors under climate change conditions. Also, one of the most important lessons Freshwater wetland habitats learned from this study is that staff buy-in Emergent marsh is critical for the assessment as well as the Shrub swamp implementation phase. Spruce-fir boreal swamp Atlantic white cedar swamp Scale and Scope Riparian forest This assessment focused on state- Hardwood swamp Vernal pools scenarios of a doubling and tripling of CO2 Coastal habitats specificconcentrations species inand the habitats atmosphere. under In the terms Intertidal mud/sandflats of time and cost, the project took about 12 Saltmarsh months to complete and cost approximately Brackish marsh $70,000, including travel and in-kind costs. Assessment Targets this project in a series of reports. This
Twenty Massachusetts habitats were project by describing how biodiversity selected for evaluation (Table C4.1). These conservationfirst report provides is currently background carried outon the by cover most of the habitats listed in the the Massachusetts Division of Fisheries Wildlife Action Plan, although they differ and Wildlife, the history, objectives, and from the Action Plan habitats in two methods of the Wildlife Action Plan, and respects. First, some habitats listed in the how the climate in Massachusetts has Plan are unlikely to be vulnerable to climate been changing and how it is expected to change (caves and mines, rocky ridgelines change over the remainder of this century. and talus slopes, rocky coastlines) and In subsequent reports, we address habitat were not considered in our analyses. and species vulnerabilities, likely ecological Second, some important habitat types shifts under climate change, and potential are subsumed within the overall habitat management/conservation options.
104 Scanning the Conservation Horizon Case Studies categories listed in the Action Plan. For The main purpose of this expert panel was example, there are many types of upland to provide answers to the vulnerability forest in Massachusetts, each of which may questions raised above. differ in their responses to climate change. For this reason we divided the Action The entire expert panel met twice at the Plan upland forest category into several beginning of the evaluative process and distinctly different habitat types (e.g., were provided with the following materials: central hardwoods, etc.) and evaluated each A PowerPoint presentation on how separatelyspruce–fir forest,(for example, northern see hardwoods, the evaluation the climate is projected to change in for northern hardwoods forest at the end of Massachusetts• over the present century. this case study). These habitat subdivisions This presentation was based on the most were based on information contained in recent and detailed climate modeling Swain and Kearsley (2000). studies that had been performed in the Northeast, particularly those of Hayhoe Assessment Approach et al. (2006), and was intended to be a “primer” for non–climate change scientists The primary questions addressed in this on the details of likely future climate phase of our adaptation work are: change (e.g., temperature, type, amount and timing of precipitation, extreme events, etc.). and wildlife habitats rank in terms of their likely• How comparative do the Wildlife vulnerabilities Action Plan to fish climate A list of the important habitat variables change? that should be considered when evaluating climate• change impacts, based on How will the representation of these existing literature on the intrinsic and habitats in Massachusetts be altered by a extrinsic factors that determine the likely changing• climate? vulnerabilities of species and habitats (e.g., Parmesan and Galbraith 2004; Parmesan and Yohe 2003; Root et al. 2003; Schneider assigned to the above predictions? and Root 2002). • What degree of confidence can be Which vertebrate Species in Greatest An appraisal of how, in general, climate Need of Conservation are likely to be most change is likely to affect habitats and vulnerable• to climate change? biomes.• This was based on previous work on the relationships between habitat To answer these questions we formed an distribution and extent and climate change expert panel of Massachusetts Division of and how habitats around the world are Fisheries and Wildlife, Nature Conservancy, currently responding to climate change and Manomet Center ecologists and wildlife (e.g., Parmesan and Galbraith 2004; biologists. The panel included much of the Parmesan and Yohe 2003; Root et al. 2003; professional expertise in Massachusetts on Schneider and Root 2002; IPCC 2007c). the status, distribution, conservation, and threats to fish, wildlife, and their habitats.
Case Studies 105 formed (freshwater aquatic habitats, forested habitats, freshwater wetlands). For each habitat type, a preliminary or straw- man vulnerability analysis was prepared in advance of the subgroup meetings and deliberations. This background information
but only to generate and guide thought andwas discussionnot intended among as a definitive the expert-panel analysis members.
Based on above materials and their expert judgment, participants in each of the subgroups were asked in face-to-face discussions to evaluate the comparative vulnerabilities of the habitats for which they have expertise under the two Paul F. Wagner emissions scenarios, asked to score them on the vulnerability scale, identify likely A habitat vulnerability scoring system future ecological trajectories, assign developed in collaboration with Division of Fisheries• and Wildlife staff. This provides a non-climate stressors that could interact framework for evaluating the comparative withconfidence and exacerbate scores, and the identify effects otherof climate vulnerabilities of Massachusetts habitats. change. Immediately after this subgroup It is based on prior work evaluating meeting, the straw-men analyses were the vulnerabilities of species to climate change and extends across the spectrum of expected responses, from habitats that may circulatedrevised to reflectto the subgroups the subgroup for discussions.further be at risk of being entirely eliminated from These modified analyses were then the state (scoring 7), to habitats likely to be relatively unaffected by climate change habitatcomment analyses and finalization. were compiled At the into conclusion a (scoring 4), to habitats that may extend of the subgroup process, the finalized their distributions greatly within the state entire expert panel so that all could have in response to climate change (scoring 1). anunified opportunity report and for circulatedcomment, roundirrespective the of habitat type. These comments were A three-point scoring system for
be• ascribed to vulnerability scores. This incorporated into this finalized analysis. assessing the levels of confidence that can Assessment Results from one developed for the IPCC process confidence scoring system was modified This section summarizes the assessments (Moss and Schneider 2000). described above by presenting the vulnerability scores, the levels of After the meetings of the entire expert panel, three habitat subgroups were ecosystem trajectories under climate change,confidence and associated potential adaptationwith them, options.likely
106 Scanning the Conservation Horizon Case Studies Drawing from the habitat impacts, evaluations for the 20 habitat types assessment participants were able to areThe presented vulnerability in Figure scores C4.1 and andconfidence Table identify some of the associated Species C4.2. For nine of the 20 habitat types the in Greatest Need of Conservation (SGNC) vulnerability scores for a tripling of CO2 that are likely to be most at risk under the exceed that for the doubling, although by various scenarios for CO2 concentration (Tables C4.3 and C4.4). the more extreme climate changes expected underrelatively the smallformer. increments. However, forThis 10 reflects of the Uncertainties habitats the two emissions scenarios resulted in identical vulnerability scores This assessment relied on expert elicitation and for one (shrub swamp) the tripling to quantify levels of uncertainty. The scenario resulted in a lower vulnerability score. These data indicate that a doubling C4.2 range from High (n = 15), to Medium confidence scores for the habitats in Table of CO2 (n = 18), to Low (n = 6). Having developed on the habitats and that extending the exposure is sufficient to a tripling to trigger has relatively major effects small that: (1) a score of High indicates that the additional impacts. habitatand used is thesemore scoreslikely than we are not confident to conform to the allocated vulnerability score. Thus,
spruce–fir forest scores High vulnerability 7
e 6 r o
c 5 S
y
t 4 il i
b 3 a r
e 2 n l u
V 1
0 h t t t t t t s s s s k h s s h p p p p s l s s s s s a s s e c r d e e o e e e e m e r r m m m m k k k a a r r r r r o a a a r a a a a a a a o o o o o l r l l m e p f f f
m m
w w w w po n r
e r l t ,
r t
da l s s s s r l i d s e s
h a e e l t r M a t l T - an f s o d b , t n i i a e a S t a oa k e r s ood f r o o u k v e r e c gen t r e i w w o a c b r w w r
r e K u h eda r V u p o a an d r v e d w d d i c d S r r i r
l l B t p r d c R e m B o e o r t u s S t r E c i c ha r c ha r a a
i
e h l t t r e H n w ne - a
i r a w g r
ll e e p w m c
a
i h r t t h nne c La r en t d r a m l n c o c t o S i o a / l C W P t N C n A he r
ou t Habitat S Figure C4.1. Habitat vulnerability to climate change. The lefthand bar in each pair represents a doubling of CO2, while the righthand bar is a tripling.
Case Studies 107 Table C4.2. Vulnerability and Confidence Scores (in parentheses) Habitat Lower Emissions Scenario Higher Emissions Scenario Forested Habitats Spruce-fir forest 6 (High) 7 (High) Northern hardwood forest 5 (Medium) 6 (Medium) Central/southern hardwood forest 1 (Medium) 1 (Medium) Pitch pine-scrub oak forest 4 (Medium) 4 (Medium) Freshwater aquatic habitats Cold water Rivers and Streams 5 (High) 6 (High) Large cold water lakes 5 (Medium) 6 (Medium) Smaller cold water lakes and ponds 7 (High) 7 (High) Cold water kettle ponds 5 (Low) 5 (Low) Warm water ponds, lakes and rivers 2 (Medium) 2 (Medium) Connecticut and Merrimack main- 5 (Medium) 6 (Medium) stems Wetland Habitats Emergent marsh 5 (High) 6 (High) Shrub swamp 4/5 (Medium) 2 (Medium) Spruce-fir boreal swamp 6 (High) 7 (High) Atlantic white cedar swamp 4 (Medium) 5 (Medium) Riparian forest 5 (Low) 5 (Low) Hardwood swamp 4 (Medium) 5 (Medium) Vernal pools 4 (Low) 5 (Low) Coastal Habitats Intertidal mud/sandflats 6 (High) 6 (High) Saltmarsh 1 (High) 1 (High) Brackish marsh 6 (High) 6 (High)
actual vulnerability score of 3 or 5, but indicates that it is very unlikely that the not 2 or 6. We are much less certain about “actual”(6 or 7) andvulnerability High confidence. score for To this us, habitat this would deviate from 6 or 7; and (2) a habitat and these should be read as implying considerablehabitats that haveuncertainty Low confidence in our scorings. scores is less certain. However, we consider it unlikelythat has athat confidence the “actual” score vulnerability of Medium of Outcomes and Next Steps such a habitat would deviate much from the allocated score. For example, we consider This assessment shows that different it possible that a habitat that scores 4 on ecological systems are more or less the vulnerability score and had a Medium vulnerable to climate change and, uncertainty score could in fact have an consequently, that we can expect to see
108 Scanning the Conservation Horizon Case Studies major changes in their distributions across Different organisms have different intrinsic the Massachusetts landscape. What forms rates of response to climate change. For will these changes take? Until relatively example, a northeastern warbler such recently our dominant model of change as the American redstart can potentially was for habitats or biomes to slowly shift its breeding range northward by replace each other as their optimum several hundred kilometers in only a few climatic conditions shifted. Thus, we days. However, the majority of the plants might expect to see the highly vulnerable that make up the breeding habitat of this species are far less able to respond being replaced by northern hardwood as rapidly—a similar shift could take forestspruce–fir as it forestsmoved atupslope upper toelevations track its decades or centuries. Rather than entire optimum climatic conditions. This model ecosystems or communities shifting their of entire communities shifting has been an distributions across the landscape, we are important step in our evaluations of what much more likely to see them dissociating may occur to habitats under climate and separating, depending on their intrinsic change (e.g., VEMAP 1995). However, it is simplistic and may not represent what may potentially novel combinations, upslope actually occur. orresponse further rates, north. and This reconfiguring, dissociation intoand
Table C4.3. Habitats and associated vertebrate SGNC most at risk from a doubling of atmospheric CO2 concentration. Spruce-fir Forest Spruce-fir Smaller Cold Brackish Marsh Intertidal Mud and Boreal Swamp Water Ponds Sand Flat Blue-spotted Northern Sharp-shinned hawk Diamondback terrapin Peregrine falcon salamander leopard frog Sharp-shinned Blackpoll warbler American eel American bittern Piping plover hawk White-throated American White sucker Least bittern Ruddy turnstone sparrow woodcock Moose Moose Green heron Northern harrier Sanderling Bobcat Water shrew King rail Red knot
Common tern Snowy egret
Short-eared owl American oystercatcher
American black duck Short-billed dowitcher
Snowy egret Whimbrel
Black-crowned night-heron
Sora Saltmarsh sharp-tailed sparrow Seaside sparrow
Eastern meadowlark
Case Studies 109 Table C4.4. Habitats and additional vertebrate SGNC most at risk from a tripling of atmospheric
CO2 concentration. Northern Hardwood Coldwater Rivers and Large Coldwater Mainstem Rivers Emergent Marsh Forest Stream Lakes Jefferson salamander Spring salamander White sucker American shad Northern leopard frog Blue-spotted salamander Longnose sucker Common loon Shortnose sturgeon Spotted turtle Bog turtle Slimy sculpin Bald eagle Atlantic sturgeon Bog turtle Ruffed grouse Blacknose dace Alewife Blanding’s turtle Broad-winged hawk Brook trout American shad Pied-billed grebe Canada warbler Burbot American eel American bittern Rock shrew Atlantic salmon Atlantic salmon Least bittern Indiana Myotis Louisiana waterthrush Fallfish King rail Eastern small-footed bat Green heron Sora Southern bog lemming Bald eagle Black-crowned night-heron Green heron Snowy egret Common moorhen Sedge wren Willow flycatcher Moose
are able to tolerate cooler temperatures model of how ecological communities may and move upslope quickly. Thus, the bereconfiguring affected by isclimate the current change. dominant northern hardwood forest will permeate
How will this be expressed in spread uphill as the climate continues to warmthe spruce–fir. and as lower-elevation This permeation plants zone spread will It is unlikely that over the next century uphill and cold-adapted species die out. weMassachusetts will see the fishnorthern and wildlife hardwood habitats? forest community, with all its associated plant The process of uphill permeation by and animal species, simply move uphill northern hardwood forest species and to occupy a vacuum left by a retreating will result in the shorter term (the next plant and animal species. What is perhaps 50uphill to 100 retreat years) by spruce–firin a spectrum forest of forestspecies morespruce–fir likely forest is that and in theall itsshorter associated term (the next few decades) we will begin to see the elimination of the most climatically foresttypes replacingoriginally the occurred, original it spruce–fir.will likely In belowest replaced elevation by a communityareas where resembling the spruce–fir forest, while the forest may retain much higher-elevation northern hardwoods. ofvulnerable its overall components structure and of thecomposition. spruce–fir At somewhat higher elevations, the new At the same time that this is happening, forest type will likely comprise a mixture of the two community types. At the highest colonized by lower-elevation species that the spruce–fir forest may be increasingly elevation types of spruce and fir trees may
110 Scanning the Conservation Horizon Case Studies persist, but it is likely that the understory the major elements of large-scale change, will now comprise many lower-elevation we must be wary of relying entirely on species. Given enough time (perhaps by the deterministic models (e.g., northern hardwood forest and all its associated may be gone. This does not necessarily meanend of that this itcentury) has been the entirely spruce–fir replaced forest from higher elevations), think more by the northern hardwood community: probabilistically,species will oust spruce–firand be prepared forests for some species that are today characteristic unforeseen surprises. others that are characteristic of northern Example: hardwoodsof spruce–fir may forests not bemay able persist, to move while Northern Hardwoods Forest Vulnerability far enough uphill. Thus, while we do Evaluation anticipate community movement, it is NTWHCS category: Appalachian northern less deterministic than some of our early hardwood forest models might predict and resembles a State ranking: S5 Vulnerability score: 5 and 6 (lower- and complete replacement. higher-emissions scenarios, respectively) long-term community reshuffle, rather than The description above implies a long, Rationale Confidence evaluation: High However, it is possible that the transitions With the distributional range of this couldslow process happen of much reshuffling more quickly of species. if habitat extending from Quebec in the stochastic events intrude. For example, north, to high-elevation areas of Virginia the warming temperatures could greatly and West Virginia, Massachusetts is close to the center of this community type’s geographical distribution. In alreadybenefit somehappening of the with invertebrate mountain pests pine Massachusetts, where it is the predominant beetlethat afflict in the northeastern West (Carroll forests. et al. 2003)This is hardwood forest, it is generally restricted and the northward spread of hemlock to an altitudinal range of about 1,000 wooly adelgid in the East (NECIA 2006). to 3,000 feet, being more adapted to More frequent or increased-intensity colder temperatures and shorter growing attacks by such pests could result in a seasons than southern/central hardwood greater incidence of tree die-off, much more standing dead timber, and larger Mature northern hardwood forest tends toforest be dominated (but less so by than sugar spruce–fir maple, yellow forest). birch, and American beech, mixed with affectedand more system, intense causing fires. If a severe much moreenough, rapid white pine, and eastern hemlock at lower transitionthese circumstances to the new could habitat “flip” type. the elevations, with red spruce and balsam
The considerations described above apply to all climate-induced transitions that forestfir becoming (Swain important and Kearsley at the 2000). highest Black we may see in Massachusetts habitats cherry,elevations white where ash, itred grades maple, into white spruce–fir birch, over the next century. What this means and gray birch often dominate early- for conservationists, planners, or land successional northern hardwood forest managers is that while we can identify following disturbance.
Case Studies 111 Within the broad matrix of northern middle-elevation Massachusetts could hardwood forest a number of variants transition to a southern/central hardwood occur, depending on local conditions. These community. Based on an elevation lapse include rich mesic forests (dominated rate of 1.8 degrees Fahrenheit for every by sugar maple, basswood, and a unique 330 feet (http://www.spc.noaa.gov/exper/ assemblage of herbaceous plants), hemlock soundings/help/lapse.html), even the groves (on cool north-facing slopes or in low-emissions scenario could contract the ravines), and transition forests (which lower edge of this community upward by include oaks, hickories, and other species at least 1,000 feet. If this were the case, more typical of southern/central hardwood then its range would be restricted to higher forest). Northern hardwood forest is not elevations (>2,000 feet) in the Berkshires. Under the high-emissions scenario this suppression over the past three centuries upward movement potential could be close maya fire-adapted have extended community the range and of human this habitat fire to 2,000 feet. This would mean that this in New England at the expense of oak habitat type might occur only at the highest
pers. comm.). This forest type is vulnerable Based on this, a vulnerability score of 5 has toforest, attack which by insects, is fire tolerant including (J. gypsy Scanlon, beenelevations, assigned where under it has the replaced low-emissions spruce–fir. moth and hemlock wooly adelgid, and scenario and a score of 6 under the high- beech scale disease. Disturbance from evaluation of Medium. (at least temporary) dominance of white emissions scenario, with a confidence pineblowdown, over other logging, species. or fire In canareas lead closer to the to human habitation or powerline cuts, because uncertainties exist regarding how nonnative plant species, including Japanese thisThe communityconfidence scoretype mightis only be Medium affected by barberry, Japanese knotweed, etc., can form climate-related and non-related factors. dense growths. Northern hardwood forest is vulnerable
Being mainly a higher-elevation and northern community, it may be expected ato consequence fire. In contrast, of increasing southern/central temperatures, that this habitat will contract its range droughts,hardwood and forests soil aredrying more is firemore tolerant. frequent If in Massachusetts as the climate warms. This contraction may be latitudinal and the transformation of areas currently elevational. At the highest elevations dominatedor hotter fires, by the this former could acceleratehabitat type to areas dominated by the latter. At lower elevations it is likely that at leastit is likely some to of replace the species spruce–fir considered forest. Other stressors that could be exacerbated characteristic of northern hardwoods and that are most temperature sensitive (e.g., on northern hardwoods include insect sugar maple, hemlock) will be replaced pestsby change (wooly and adelgids inflict adverse are already impacts by elements of the southern/central eliminating large tracts of hemlocks in hardwood forest (e.g., white oak, hickories, Massachusetts, and emerald ash borer and etc.). Thus, what is currently northern Asian longhorned beetle are spreading hardwood forest over much of low- and rapidly north toward and into the state).
112 Scanning the Conservation Horizon Case Studies Lower Elevations (<2,000 feet) Higher Elevations (>2,000 feet)
Southern/Central Hardwood Forest Northern Hardwood Forest
Loss of Sensitive Species Colonization by Southern (e.g., Sugar Maple, Hemlock) Species (e.g., Oaks, Hickory)
Increased Fire Frequency/Intensity
Northern Hardwood Habitat Loss: Climate Change: Forest - Urbanization - Warming - Non-native Species - Drought
Increased Pests and Pathogens
Figure C4.2. Conceptual model of how climate change and other stressors might affect Northern Hardwood Forest in Massachusetts.
If, as seems likely, warming temperatures colonization by nonnative plant species, as facilitate overwinter survival of these pests the state’s population and peoples’ housing and allow them to spread further north expectations continue to grow. These or upwards in elevation in Massachusetts, the northern hardwood forest could be future status of this habitat. A conceptual adversely affected. Under the higher- modelfactors of could how be climate major and influences non-climate on the emissions scenario, an increased frequency stressors might affect northern hardwoods in Massachusetts is shown in Figure C4.2. and frequencies of insect/pathogen Thus, while we conservatively score the attack,and severity could ofpotentially fire, and greatereliminate intensities this vulnerability of this habitat type as 5 and habitat type from the state. Moreover, this 6 under the lower- and higher-emissions community type is currently under stress scenarios, it is possible that this might from human rural development and from underestimate its vulnerability.
Case Studies 113 Case Study 5. information about the vulnerability of the Chesapeake Bay region’s coastal habitats Vulnerability to to sea-level rise. Sea-Level Rise in the Conservation Objective Chesapeake Bay This assessment is intended to enhance This case provides an example of climate and support efforts to address the impacts change vulnerability assessments for both of climate change as part of the ongoing habitats and species in the Chesapeake regional strategy to restore and protect the ecological integrity of the Chesapeake Bay studies, one building off the other, to and surrounding coastal areas. identifyBay region. how Specifically, sea-level rise it highlights is likely to two affect coastal wetland habitats and associated Assessment Targets marsh bird species in the region.
National Wildlife Federation wetland habitats based on categories Coastal Habitat Study establishedThe study focused for the specificallyU.S. Fish and on Wildlife coastal Service’s National Wetlands Inventory. In 2007, the National Wildlife Federation, Habitat types include: swamp, tidal swamp, working with Warren Pinnacle Consulting, inland fresh marsh, tidal fresh marsh, Inc., initiated a study to identify the potential impacts of sea-level rise on (brackish) marsh, saltmarsh, estuarine coastal wetland habitats in the Chesapeake transitional marsh, irregularly flooded Bay region, including Delaware Bay and the ocean beaches of southern New Jersey, Scalebeach, tidal and flat, Scope and ocean beach. Delaware, Maryland, and Virginia. This was a regionally focused study, Purpose covering coastal habitats from Delaware Bay to the north, the Chesapeake Bay, and Local, state, and federal agencies, non- the ocean beaches from southeastern New governmental organizations, and other Jersey down to the Virginia–North Carolina stakeholders have already begun to develop border. In terms of temporal scale, the climate change adaptation strategies for the study looked at a range of sea-level rise scenarios from the 2001 IPCC assessment, with the expected changes to their habitats ranging from 0.31 meters to 0.69 meters region to help fish, wildlife, and people cope and communities, including sea-level rise, in eustatic sea-level rise by 2100 (building over 25-year time steps from year 2000 with unforeseen impacts. The purpose of levels). The study also modeled a rise as well as build in the flexibility to deal this assessment is to provide decision- of up to 2 meters by 2100 to accommodate makers involved in this process with for recent studies that suggest a
Lead authors: Patty Glick and Michael Wilson.
114 Scanning the Conservation Horizon Case Studies possible during this century (Vermeer improved through high-vertical-resolution andsignificantly Rahmstorf greater 2009; sea-level Overpeck rise and is Weiss Lightin many Detection places hasand been Ranging significantly (LiDAR) 2010; Rahmstorf et al. 2007). This study data; historic sea-level trends and the salt cost approximately $40,000 and took just boundary (the elevation below which lands under 1 year to complete. are periodically inundated by salt water) are calculated using National Oceanic and Assessment Approach Atmospheric Administration tidal gauge data; and factors contributing to relative The assessment was conducted by applying sea-level change can be determined Version 5.0 of the SLAMM model, which was designed to simulate the dominant opinion, and other sources. processes involved in wetland conversions through the scientific literature, expert multitude of different habitat types under variousand shoreline scenarios modifications of sea-level among rise. There a are a number of modeling tools available to assess the impacts of sea-level rise, ranging from simple “bathtub” models that show general inundation of coastal lands based on relative land elevation (e.g., DGESL 2003) to more detailed models that can simulate dynamic ecological processes at multiple spatial scales (e.g., Barataria- Terrebonne Estuarine Landscape Spatial Simulation Model, or BTELSS) (McLeod et al. 2010; Reyes et al. 2000). The SLAMM ago and has been applied in a number of studies.model was (Craft first et developed al. 2009; Galbraith two decades et al. 2002; Lee et al. 1992; Park et al. 1989). It provides an accessible, middle-of-the- road tool that allows for fairly detailed, within the constraints of relatively limited USFWS datascientifically availability, sound budgets, regional and assessments time. The SLAMM model addresses the relative Data inputs for SLAMM are generally sensitivity of habitat types to sea-level readily available in most areas of the United rise based on known ecological traits such States. For example, habitat composition as the tolerance for salinity of associated data are based on the National Wetlands plant species. Elements of exposure to Inventory; coastal elevation data are sea-level rise are based on land elevation available through the U.S. Geological as well as the scenarios of sea-level rise Survey’s National Elevation Dataset, which modeled. Adaptive capacity is addressed
Case Studies 115 in terms of both intrinsic and extrinsic subsidence, sedimentation, and marsh factors. One of the most important features primary production. In the Chesapeake of SLAMM is its ability to capture important Bay, for example, many areas have been localized differences in relative sea level. naturally subsiding (declining in elevation), As mentioned previously in this guide, which increases the amount of relative estimates of global sea-level rise due to sea-level rise affecting the region, making climate change are based on what is called those areas more vulnerable. In contrast, eustatic sea-level rise, which refers to the some coastal habitats such as marshes and changes in ocean volume due to thermal beaches may at least to some extent be expansion and the melting of glaciers able to accommodate moderate changes in and ice sheets. At the localized level, the sea level by increasing in elevation due to amount of relative sea-level rise can vary the buildup of sediments. Rates of change due to factors (both natural and human- among these variables can be included in the model. In addition, SLAMM can in vertical land elevation, such as land incorporate areas where habitats may not influenced) that determine changes be able to move inland as sea level rises due to the existence of coastal armoring such as seawalls and dikes (factors that reduce the adaptive capacity of these habitats). Finally, the model output can be displayed as percentage changes in habitat area as well as shown on maps, both of which are useful for decision-makers. Assessment Results
Model results vary considerably across
changes to coastal wetlands and other habitatsthe region, occur but inoverall the eastern the most and significant southern regions of the Chesapeake Bay, most of Delaware Bay, and along the coastal barrier islands and beaches. For example, under
Greg Breese/USFWS the scenario of a 0.69-meter sea-level rise, which is the IPCC’s (2001a) A1B Max scenario for the year 2100, coastal marshes would be inundated with salt water, converting 83 percent of brackish marsh to saltmarsh or open water (see Table C5.1). Overall, the area of tidal marshes (including tidal freshwater marsh, irregularly
saltmarsh) declines by 36 percent under thisflooded scenario. marsh, Ocean transitional and estuarine saltmarsh, beaches and
116 Scanning the Conservation Horizon Case Studies Table C5.1. Percentage changes in habitat area for entire study site (Glick et al. 2008a). also fare poorly, declining by 69 percent vulnerable is because, in addition to facing and 58 percent, respectively. In addition, eustatic sea-level rise, land subsidence more than half of the region’s important is greater than for many other parts of tidal swamp is at risk, declining by 57 the Chesapeake Bay due to groundwater percent by 2100. Many coastal plant and withdrawal for agriculture (U.S. FWS 2005). animal species have adapted to a certain In addition, marshes in much of the Eastern Shore appear to have relatively lower rates composition, so these shifts will make of natural accretion (Kearney et al. 1998). habitatslevel of salinity, more favorable tidal influence, for some and species habitat It is important to note that high-quality and less so for others. LiDAR data were used for this study site.
Figure C5.1 shows map-based model Uncertainties results for the Blackwater National Wildlife Refuge, Maryland, and surrounding areas It is important for decision-makers to under the 0.69-meter scenario (IPCC A1B recognize that, as with all predictive Max) (just one of multiple scenarios run). models, SLAMM has its limitations and is subject to uncertainty. One element of extent of coastal habitats occur at this site, uncertainty is associated with projections includingSignificant a changesloss of 39 in percent the composition of dry land, and for future sea-level rise. The SLAMM model 92 percent of inland freshwater marsh, and itself does not project changes in eustatic sea level—it relies instead on sea-level rise marsh, much of which converts to open scenarios developed by others, such as the water.94 percent One of theirregularly reasons flooded this area (brackish) is so IPCC. As discussed above, there is currently
Case Studies 117 Figure C5.1. Sea level rise and marsh conversion projections for Blackwater National Wildlife Refuge, Maryland (Glick et al. 2008a).
118 Scanning the Conservation Horizon Case Studies a wide range of possible scenarios for sea- assumptions and areas of uncertainty are level rise, depending on the assumptions made transparent when communicating and factors used. Rather than choosing any the study results. Users can then use their one scenario, we applied multiple scenarios own judgment about their relevance for in our analysis to inform potential informing decisions regarding coastal management responses. restoration and management actions. Ultimately, the response strategies will Apart from any uncertainties in the vary for different areas, and more sea-level rise projections themselves, in large part on the quality of the data bydetailed, identifying site-specific factors studiesthat have may not be been inputs.confidence For example,in SLAMM areas projections where LiDAR depends effectivelywarranted capturedto supplement by the these model findings or elevation data are unavailable have a remain uncertain. wider “zone of uncertainty” with respect to the delineation of vulnerable lands at or Outcomes and Next Steps address qualitatively in the results for this The results of this analysis were published particularbelow a specific study. elevation,In addition, a factor given that weit as two documents, including a detailed is a relatively simple model, SLAMM does technical report (Glick et al. 2008a) and a not capture some of the more complex summary (Glick et al. 2008b). In addition, systematic changes that could occur the National Wildlife Federation has made over time, which may include dynamic all data and model results available to changes in marsh accretion or localized interested parties, including local, state, geomorphology. For example, the SLAMM and federal agencies, academic institutions, model used for this assessment (Version and non-governmental organizations, to 5.0) includes a simplifying assumption support additional analyses. that rates of accretion (and subsidence) will remain linear into the future. This Center for Conservation Biology assumption, in particular, has been a Marsh Bird Study source of criticism of the SLAMM model (e.g., Cahoon and Guntenspergen 2010; Purpose Kirwan and Guntenspergen 2009). It is important to note that newer versions of Based on the results of the National SLAMM (e.g., Version 6.0), incorporate Wildlife Federation’s sea-level rise dynamic feedbacks for marsh accretion study described above, scientists at the based on elevation, distance to river or tidal Center for Conservation Biology at the channels, and salinity (Clough et al. 2010). College of William and Mary and Virginia Commonwealth University have conducted However, the use of the simplifying assessment of the potential impacts of assumption for rates of relative sea-level sea-level rise on the population size of changes in this or similar studies does saltmarsh breeding birds within the not mean that its results are not useful. Chesapeake Bay (Wilson and Watts 2009). As with any vulnerability assessment, The Chesapeake Bay region supports 30 what is important is that the major model percent of the total saltmarsh cover along
Case Studies 119 the Atlantic Coast and is an important often interspersed with shrubs such as breeding area for several marsh bird marsh elder or saltbush. Clapper rails, species of concern. Based on assessments Virginia rails, marsh wrens, and seaside by the National Wildlife Federation, the sparrows, on the other hand, use both U.S. Climate Change Science Program (U.S. low and high marsh but reach their CCSP 2009b), and others, it is clear that the highest densities in the low marsh, which saltmarsh ecosystem of the region is highly is inundated with each tidal cycle and vulnerable to the impacts of sea-level rise. dominated by smooth cordgrass and black needlerush. Conservation Objective Scale and Scope The objective of this assessment was to examine the potential impact of sea-level The GIS-based study used the National rise on the capacity of the Chesapeake Wildlife Federation study’s habitat change Bay region to support saltmarsh bird projections for several of the sea-level rise populations during the breeding season. scenarios, including 0.39 meters by 2100, 1 Because marsh bird abundance and meter by 2100, and 2 meters by 2100 (Glick et al. 2008a). For their purposes, some of characteristics of marshes such as salinity, the habitat types were aggregated into vegetation,distribution patch are influenced size, and elevation, by the physical the high saltmarsh, low saltmarsh, and impact of sea-level rise on the population transitional saltmarsh to correspond with size of each species is more complex than the marsh types preferred by various simply calculating net changes in marsh marsh bird species. This study cost area alone. Given that SLAMM is able to approximately $25,000 and took about 4 identify changes in the composition of months to complete. coastal wetland habitats, as well as their extent, the National Wildlife Federation Assessment Approach study provides an excellent platform for this assessment. For the purposes of this study, each GIS raster cell from the National Wildlife Assessment Targets Federation SLAMM analysis was assigned a value for the dominant habitat cover This assessment focused on several marsh category. We then converted the rasters birds of concern, including the willet, for emergent tidal marshes that were black rail, saltmarsh sparrow, clapper rail, coded as high salt marsh, low salt marsh, Virginia rail, marsh wren, and seaside and transitional salt marsh into vector- formatted polygons using ArcMap 9.1 habitat requirements based on the (Environmental Systems Research Institute compositionsparrow. Each of of coastal these speciesmarsh types. has specific The 2005) and grouped any like-valued cells willet, black rail, and saltmarsh sparrow, for example, rely exclusively on high marsh, marsh patch. This grouping allowed which is only inundated during extreme neighboring≤30 meters from cells eachto share other the into identity a single of high-tide events and dominated by plants a larger marsh complex while cells >30 such as saltmeadow hay, saltgrass, and meters away from an aggregation would be
120 Scanning the Conservation Horizon Case Studies grouped into a different marsh complex. We between the years 2000 and 2100. These placed the resulting aggregations into one reductions likely represent conservative estimates since global rates of sea-level (2) 1–5 hectares, (3) 5–10 hectares, (4) rise are predicted to accelerate. Population 10–50of five marshhectares, area and classes: (5) >50 (1) hectares. <1 hectare, reductions of 69 to 80 percent are possible if sea levels increase by 1 meter or greater. Bird population sizes for each map iteration The differential impacts of sea-level rise values of bird density (birds/ha), based on on high versus low marsh are projected researchwere identified conducted by extrapolating within the Chesapeake known to result in dramatic changes in the bird Bay. Available habitat was delimited community on a regional scale. Marsh for each species by its (1) geographic birds such as the black rail and saltmarsh distribution over the study area, (2) habitat sparrow that rely exclusively on high associations, and (3) patch-level incidence marsh habitat will be at very high risk of rates for each area class. These associations extirpation from the region. Habitat area were then incorporated into an equation for these species is projected to decline that calculated changes in species’ to about 50 percent of current area under populations associated with projected the 0.39-meter scenario. Habitat for both changes in marsh habitat area due to sea- the black rail and saltmarsh sparrow was level rise. reduced 89 percent from the 1-meter scenario and 99 percent from the 2-meter To address potential uncertainty, scenario for each species, respectively. In addition, populations of clapper rails, each population estimate by substituting Virginia rails, willets, seaside sparrows, confidence intervals were calculated for and marsh wrens were projected to decline interval values of each bird density for by about 40 percent in the 0.39-meter patchthe upper area and class lower values 95 in percent their modeling confidence scenario, about 80 percent in the 1-meter equation. As these factors change over time scenario, and about 75 percent in the due to sea-level rise, so do the population 2-meter scenario. sizes and distribution of associated marsh bird species. Uncertainties (Addressed above.) Assessment Results Outcomes and Next Steps The total area of saltmarsh was reduced between initial conditions and the year The results of this study illustrate the need 2100 for all sea-level rise scenarios. These changes were dominated by reductions that reduce the adverse impacts of sea-level of high marsh. Sea-level rise was also to address specific management actions responsible for a shift in the rank order loss of marsh area is likely, particularly of abundance between low marsh and underrise on the marsh higher birds. projections Because forsignificant eustatic high marsh. In turn, model results suggest sea-level rise, there will be an increasing that the current rates of sea-level rise will need to protect remaining marsh patches to reduce the size of breeding populations compensate for these losses. of widespread species 35 to 40 percent
Case Studies 121 Case Study 6. An being observed and documented in the region, including recent widespread forest Integrated Climate dieback and population changes in native Change Assessment species (Inouye et al. 2000; Breshears et al. 2005). Natural resource professionals need Framework in the to better understand the regional impacts Four Corners Region of climate change so that they can take informed adaptive action in managing the Summary landscapes that provide ecosystem services to human communities and habitat for a This case study describes the use of a diversity of species. To address the need for regional-scale vulnerability assessment information and guidance on responding to the potential consequences of climate of potential management responses. In change, The Nature Conservancy initiated theframework western that United includes States, identification temperatures the Southwest Climate Change Initiative have exceeded global averages by 70 (SWCCI), a collaborative effort between The percent relative to the 20th century Nature Conservancy and partners (state, (Saunders et al. 2008). The region federal, non-governmental organizations, faces a 90 percent chance of continued universities) in the Four Corners states. More information about this study can in rivers and streams, and higher rates of be found at: http://nmconservation.org/ evaporationdeclining snowpack, from reservoirs, earlier peak leading flows to increased competition for already over- projects/new_mexico_climate_change/. allocated water resources (IPCC 2007b). Purpose and Warming trends are projected to continue, Conservation Objective with climate in the semiarid Southwest resembling “Dust Bowl” conditions by In 2006, the New Mexico Chapter of The mid-century (Seagar et al. 2007). Ecological consequences of these effects are already climate change ecology and adaptation program.Nature Conservancy Building on first the launchedcompletion its of own a state-wide climate change assessment and two adaptation-oriented workshops for natural resource managers in New Mexico (Enquist and Gori 2008; Enquist et al. 2008), The Nature Conservancy then launched the SWCCI. The Initiative
apply an integrated assessment approach thatspecifically examines seeks regional to further climate develop impacts, and prioritizes adaptation actions based on
B. Klein potential vulnerability, and identifies specific climate adaptation strategies in Lead authors: Patrick McCarthy and Carolyn Enquist.
122 Scanning the Conservation Horizon Case Studies priority landscapes in the region using assessment of climate exposure using an adaptation planning framework. The conservation objective of the project is climate metrics include departures and to provide a straightforward approach trendsthe ClimateWizard in temperature, analysis precipitation, tool. Specific and for integrating climate change into moisture stress (i.e., evaporative demand). existing conservation and management We mapped and analyzed these metrics planning and decision-making processes. development of this suite of nested-scale approaches,Ultimately, with the continuedproject seeks refinement to cultivate and dialogue and collaboration between conservation practitioners, managers, and policy-makers from multiple jurisdictions, so that they can work toward reducing the vulnerability and increasing the resilience of current and future conservation priorities to ongoing climate change. Scale and Scope
The assessment covers an area that comprises four states of the southwestern United States: Arizona, Colorado, New Mexico, and Utah. The New Mexico state- Michael D. Peterson level climate and vulnerability analyses took 2.5 years to complete, with an across the four states both retrospectively approximate cost of $75,000, including and prospectively. For historical analyses, ClimateWizard uses publicly available external partners. The expanded SWCCI PRISM climate data set at a cell resolution projectsignificant conducted in-kind thecontributions baseline regional from of 4 square kilometers (Daly 1994). We assessment in 1 year; the adaptation used an ensemble of 16 IPCC AR4 global planning workshops in each of the circulation models statistically downscaled four states were implemented over the to 12 kilometers for the future analyses course of 1.5 years (Phase 1). Together, (Maurer et al. 2007). Annual and seasonal trends were evaluated for the time periods approximately $200,000. analyzed (1950 to 2006, 1970 to 2006, the first phase of the SWCCI project cost 2020 to 2039, and 2069 to 2099). Statistical Assessment Approach summaries were generated for each climate and Targets metric and time period for the region and for each state, ecoregion, and watershed. Regional Assessment To identify potential vulnerability, we next evaluated climate change exposure in The SWCCI integrated framework begins relation to existing conservation priorities, with a spatially explicit rapid regional or targets (e.g., geographies, habitats,
Case Studies 123 and goal. The third step involves four states’ Wildlife Action Plans and building a conceptual ecological model and ecoregionaland sensitive assessments. species), identified We used in species the assessing known and potential climate occurrence data provided by NatureServe change impacts. As part of this model, we for our species-focused analyses, consider other social, political, economic, NatureServe vegetation analysis data for and ecological constraints on the system, habitats, and spatial data of The Nature how they are interconnected, and how Conservancy’s network of conservation they will likely be affected by the selected areas for priority geographies. climate change scenario. In the fourth step, we identify management intervention Landscape Adaptation Planning points for the target, or the components of the system that management can affect. The second part of the SWCCI integrated We then identify management actions, framework aims to identify a suite of or adaptation strategies, that can be place-based adaptation strategies for taken at those points to address climate potentially vulnerable species, ecosystems, change impacts. The fifth step consists or natural processes in priority landscapes. of evaluating the identified adaptation strategies based on their effectiveness developed by a Wildlife Conservation (e.g., ability to affect target), robustness to Society–NationalWe specifically use Center an approach for Ecological recently variations in the climate change scenario Analysis & Synthesis working group and/or management objective, and the comprising managers and scientists from ability to monitor success. We also address academia, non-governmental organizations, the issue of uncertainty associated with a and federal agencies. This framework strategy, asking if the effect on the target will be “win-win” (where there could be managers to work in a transparent, participatoryallows local scientific process expertsto identify and climate approach), or “proactive/anticipatory” change threats and impacts and translate (mostmultiple risky). benefits), “no regrets” (less risky this information into a portfolio of strategies that are applicable to the Results landscape of interest. The proposed strategies can then be evaluated in the Regional Assessment social, political, regulatory, and economic contexts that motivate and constrain Temperature increased across the management goals and policies. Subsequent Southwest from 1951 to 2006, rising an iterations allow the incorporation of newly average of 0.5 degrees Celsius (about 1 available information. degree Fahrenheit) every 30 years. The temperature increase was consistent across As the first step in this process, participants select a target and related 100 percent of the watersheds, habitats, management goals. In the second step, ecoregions,scales, rising and significantly states that across we evaluated. 70 to climate change Precipitation also increased across the scenario (e.g., warmer–drier with increased region in the historic period, increasing by climatewe define extremes) a plausible to apply to the target about 10 percent every 30 years. However,
124 Scanning the Conservation Horizon Case Studies Maja Smith this trend was less consistent across scales, places may be the ones that may have rising in only 36 to 75 percent of the units more resilience to predicted future we evaluated. Several areas were actually temperature changes—at least in the shorter term. These include the Upper Climate change models predict that Pecos (New Mexico) and Upper Arkansas temperaturedrier, but none will of continuethese were to significant.rise in future, (Colorado) watersheds. between 1 to 3 degrees Celsius by 2020 to 2039 and 3 to 5 degrees Celsius by 2080 to Landscape Adaptation Planning 2099. The temperature rise pattern is fairly consistent across scales. Predicted changes Using the results of the regional assessment in precipitation are less consistent across in a convened meeting of stakeholders,
in each of the Four Corners states for percenttime and to space, +11 percent ranging in from 2080 −6 to percent 2099. implementationwe identified case of study the collaborative landscapes to +6 percent in 2020 to 2039 and −13 Wildlife Conservation Society (WCS) and Our preliminary analyses suggest that National Center for Ecological Analysis and many sensitive-species-rich watersheds Synthesis (NCEAS) adaptation planning are among those that have been (1951 to process. We subsequently have engaged 2006) and will continue to be exposed to representatives from local agencies and extreme temperature changes. In particular non-governmental organizations to plan we found that the Lower Colorado-Lake and implement the process in medium- Mead (Arizona) and Upper Colorado- sized workshops in each landscape for Dirty Devil (Utah) watersheds may be the local resource managers. The goal of the most vulnerable places in the Southwest workshops is to translate available climate given these criteria and large percentage change science, so that managers can of freshwater and endemic species. overcome uncertainty paralysis and begin Conversely, there are sensitive-species-rich to identify strategies for helping species watersheds that have experienced less and ecosystems adapt to climate change. exposure during the past 55 years. These
Specific objectives of the workshops are
Case Studies 125 to (1) provide background information on Uncertainties climate change and its effects in the focal landscapes and (2) identify opportunities Assessment Climate Data for learning, collaboration, and application of the adaptation planning process for At a spatial resolution of 4 square natural resource management in the kilometers, the PRISM model estimates focal landscapes. As of December 2009, monthly temperature and precipitation we have held two workshops in the using a combination of climate station data, Jemez Mountains of New Mexico and the a digital elevation model, and expert-based Gunnison Basin of Colorado. Additional knowledge of complex climatic processes, workshops are being planned for the Four such as rain shadows and temperature Forest Restoration Initiative focal area inversions. As a statistically interpolated in the vicinity of Flagstaff, Arizona (April raster data set, there are inherent 2010), and for the Bear River Basin of uncertainties associated with data that are Utah (May 2010). further away from points (or data cells) with actual station data. These “inbetween” areas should be interpreted with some caution, particularly those values associated with a single data cell. Problems with station data can also exist, as some of these may have shorter periods of record or may have been physically moved during the course of data collection. Furthermore, geographical areas with complex topography may be particularly subject to data anomalies (e.g., the mountainous area of Colorado). Uncertainty associated with downscaled future climate models was also addressed.
Assessment Targets
We viewed conservation priorities as “surrogates of sensitivity” because we presumed all units of native biodiversity, especially drought-sensitive species, are likely to be sensitive on some level to rapid and abrupt climate change, despite adaptive ability (e.g., Sala et al. 2000). This may be especially true for species, ecosystems, or places that have formalized
Maja Smith conservation status, given that they were
identified at least in part because of
126 Scanning the Conservation Horizon Case Studies and one more extreme—for two ecological etc.) that may render them susceptible to human-inducedspecific attributes threats. (e.g., rarity, In a subsequent endemism, phase of the project, using results of the thatprocess–based many of the conservation conservation features, strategies fire alreadyand in-stream being plannedflows. Participants or implemented found in the Jemez Mountains can be used to tofirst species phase of as concern a guide, and we willhabitat investigate types. prepare for climate change. But, even under Althoughspecific climate-related our approaches sensitivities do not specifically related the more conservative of the two climate measure a conservation priority’s adaptive change scenarios we explored, the scale, capacity, we generated hypotheses of which sequencing, priority, and cost of these conservation priorities are most and least strategies will likely need to be adjusted vulnerable to ongoing climate change were if management objectives are to be met. developed. These will be used to further by the overall group included: system- Examples of priority strategies identified Outcomesdevelop and refine and future Next analyses. Steps climate change; improvement of riparian ecosystemwide management health by planning fencing forout fire elk and We are currently in the process of writing cattle or by reducing the landscape’s up the results of our regional assessment in a report format written for natural resource management; widening the prescribed professionals from across the Southwest. elk herd; landscape-scale ecological fire Our target distribution date is June 1, weather conditions under which prescribed 2010. Our target completion date is June burningfire window can be(i.e., implemented); expanding the and suite applying of 2010, when we will post it to the project’s forest thinning prescriptions to promote Website (see above). We also are in the snowpack retention and maximum process of writing a project team charter also listed numerous actions that could and further developing our assessment beprecipitation taken to carry infiltration. out these Participants strategies, methodsto identify and specific related next funding steps opportunities.for refining opportunities for implementation. Jemez Mountains Climate Change and they identified both barriers to and Adaptation Workshop Gunnison Basin Climate Change Adaptation Workshop workshops to be organized by the SWCCI, This was the second workshop organized heldThis wason April the first21 to in 22, a series 2009, of in four Los Alamos, by the SWCCI, held on December 2 to 3, New Mexico. More than 50 representatives 2009, in Gunnison, Colorado. Fifty-seven of state and federal agencies, tribal representatives of 20 state and federal governments, and non-governmental agencies, local governments, academic organizations participated. Over the institutions, and non-governmental course of 2 days, managers, scientists, and organizations participated. Using two conservation practitioners worked together climate change scenarios, managers to identify adaptation strategies under two climate change scenarios—one moderate framework for three “conservation identified strategies using the adaptation
Case Studies 127 features.” Similar to the Jemez Mountains of climate change, impacts to species, workshop, many strategies resemble ecosystems, and ecological processes, existing ones. Examples of priority strategic actions that emerged from the workshop Participants also acknowledged that for the three conservation features include: effectiveand to refine communication the identified among strategies. local stakeholders and policy-makers is critical Gunnison sage-grouse: to building trust and to engaging people in the development of realistic management •- Retain water in most vulnerable sage- objectives as we face the possibility of grouse habitats via restoration of seeps and undesired future conditions.
agricultural practices Overall workshop planners were pleased springs and implementing more efficient with the outcomes of the two workshops, - Improve and restore nesting and noting that the planning framework gives wintering sage-grouse habitats managers the opportunity to document
Gunnison headwaters: options in what is a relatively transparent process.logic and Additionally, assumptions, participants justify specific •- Manage upland vegetation for needs. Perhaps most importantly, the maintenance workshopsidentified future provided information a forum andto promote research groundwater recharge and base flow landscape-level collaboration and - Construct and/or restore wetland continued dialogue via the formation of complexes informal climate change learning networks.
Alpine wetlands: workshops describing emergent and •- Build snow fences to augment water divergentA final, synthetic strategies, report recommendations of the four for inputs moving the process forward, and lessons learned is targeted for completion by - Increase buffer zones around alpine the autumn of 2010. Pending additional wetlands funding, we hope to conduct follow-up workshops focused on implementation, Participants of both workshops recognized that more work is needed to develop management strategies using an adaptive strategies to reduce the impacts predicted managementtesting, and monitoring framework. of identified under extreme climate change scenarios. The ecological changes that could occur under these scenarios require more intensive and extensive management intervention or perhaps even wholesale changes in management goals. Participants expressed an interest in continuing to
work together to refine our understanding
128 Scanning the Conservation Horizon Case Studies Case Study 7. Pacific Northwest Climate Change Vulnerability Assessment Summary
This case study describes the early stages of a broad, regional-scale vulnerability assessment for species and habitats
that is applying a range of assessment Veni approaches. This collaborative project is creating a digital database of climate 2009). Precipitation is also projected to change sensitivities of species and systems change, with general increases projected
which are inherently most sensitive to intense seasonal precipitation cycle— in the Pacific Northwest and will identify climate change. Species and systems autumnsfor the Pacific and winters Northwest, may, and in fact, with become a more sensitivities are based on physiology, wetter and summers may become drier. habitat requirements, life history, dispersal Regional climate models indicate that ability, population growth rates, location, overall extreme precipitation in western ecological climate effects, and disturbance Washington will increase and the snowpack regime effects. Sensitivity is being in the Cascades will decrease (Mote and Salathé 2009). and pertinent data sets. This database will identified by experts, scientific literature, provide natural resource managers with Purpose of the Vulnerability critical information that can be combined Assessment with the management tools already in their toolbox to address climate change and better prepare for the future. Assessment was developed in response toThe the Pacific considerable Northwest challenge Vulnerability that climate Background—Climate change poses to natural resource managers. Change in the Pacific Leading this voluntary approach, the Northwest University of Washington is partnering with key collaborators, such as scientists, natural resource managers, and increased by about 0.8 degrees Celsius and modelsPacific Northwest project warming temperatures of 2.0 degrees have Northwest region to conduct a climate– Celsius by the 2040s and 3.3 degrees ecologicalconservation vulnerability planners in assessment. the Pacific Celsius by the 2080s (Mote and Salathé
Lead authors: Michael Case and Josh Lawler.
Case Studies 129 David J. Mills Conservation Objective Scale and Scope
The assessment will provide managers The assessment covers an area that extends and planners with information about beyond the borders of Washington, Oregon, which species and systems will be most and Idaho (see Figure C7.1). Current vulnerable to climate change and in project funding to populate and analyze the what ways they will be vulnerable. This database will continue until 2011. However, information can be used to prioritize access and use of the database will continue management actions, design adaptation into the future and will eventually cover a strategies, and apportion scarce resources. broader spatial scale.
Assessment Targets The resource needs for this type of
The assessment targets both species and will likely require a total of $800,000 ecological systems (habitats). One of the andassessment 3 to 4 years are significant to complete), (the but project it is goals of this project is to develop a digital being conducted at a regional scale, database of inherent climate change considers both selected species and sensitivities for species and systems of broad habitat categories, and combines the resources of state and federal and to provide resource managers agencies, academic institutions, and andconcern decision-makers throughout the with Pacific important Northwest non-governmental organizations. information about how species and systems will likely respond to climate change. The Assessment Approach project database and related modeling assessments will also allow researchers This project is a collaboration among researchers, managers, and planners at the regarding the potential impacts of climate University of Washington, U.S. Geological changeto address on naturalimportant resources. scientific questions Survey, The Nature Conservancy, the National Park Service, the U.S. Forest
130 Scanning the Conservation Horizon Case Studies Service, the Washington Department of Fish this part of the project is to statistically and Wildlife, the University of Idaho, the downscale climate change projections National Wildlife Federation, the Oregon to produce projected changes in climate Department of Fish and Wildlife, and Idaho at approximately 1-square-kilometer Fish and Game. resolution for at least six different future climate projections. This relatively high- resolution data will provide information Vulnerability Assessment includes two The Pacific Northwest Climate Change that will highlight and detail the sensitivity todistinct climate components; change of species the first and is ahabitats database in the study region. For this project,
North inherent susceptibility to climate change. Cascades sensitivity is defined as a measure of the Canadian Some species and systems are inherently Rocky Mountains Okanagan more susceptible to climate change Paci c Northwest than others. The estimated sensitivity of Coast individual species is currently assessed on the following characteristics: ability of the Willamette Valley Puget Trough species to disperse and whether dispersal Georgia Basin Middle Rockies- West Blue Mountains barriers exist; dependency on disturbance Cascades Utah-Wyoming Rocky Mountains
Columbia East Cascades Plateau physiology and ecology (i.e., sensitivity Modoc Plateau regimes such as fire or flood regimes; to temperature, precipitation, salinity, Klamath Mountains pH, CO2); dependency on and persistence California of climatically sensitive habitats (such North Coast as alpine areas, shallow wetlands, and perennial streams); whether the species is a generalist or specialist, and whether Figure C7.1. Study area. its existence is tied to other species. The sensitivity of ecosystems and communities that is more applicable to regional planning will be based on hydrological sensitivities, and natural resource management. We component species sensitivities, proximity will then use these downscaled climate to the coast, and the effects of disturbance projections in conjunction with soils data regimes. The database will be used in as inputs to a dynamic global vegetation conjunction with a sensitivity index to model to project potential changes in produce a ranking of more than 400 the vegetation of the region. Vegetation species and systems with respect to their sensitivity to climatic change. composition of different basic plant functionaltypes will betypes defined such by as theirbroad-leafed relative The second component of the assessment deciduous trees and grasses. These involves modeling the potential effects of vegetation data will then be used to model climate change on species and habitats of the current and potential future ranges for 12 bird and mammal species in the the Pacific Northwest. The initial step of
Case Studies 131 characteristics into the database experts are asked to assign uncertainties along with their answers and these uncertainties are reported in the database output. Additionally, the project assigns levels of uncertainty to the simulated climatic and bioclimatic changes and the simulated vegetation changes. Assessment Results
While still in the early stages of populating the database, preliminary analysis has examined the overall sensitivity (calculated by the database index) of more than 20 species present on the Olympic Peninsula (see Figure C7.2). The results of this analysis highlight the potential vulnerability of species that rely on sensitive habitats and have stronger physiological sensitivities to climate change (Halofsky et al., in press). Species that occupy vulnerable habitats, such as the Richard Dalby Olympic torrent salamander (headwater streams), Cascades frog and Van Dyke’s salamander (aquatic habitats), Dogstar use a hierarchical modeling approach to projectPacific Northwest. future species Specifically, distributions. we will First, we will use continental-scale models (e.g., Olympicskipper butterfly marmot, (meadows),mountain goat, Makah Clark’s Lawler et al. 2009) to predict future species copper butterfly (wetlands), and the distributions at a 50-square-kilometer (high-elevation habitats), were generally resolution across the study region. These rankednutcracker, as highly and gray-crowned sensitive to climate rosy finch predictions will provide estimates of change. Similarly, specialist species in the future range boundaries of the each terms of habitat and diet, such as Clark’s species. Then, within the projected future nutcracker, northern spotted owl, gray- ranges, we will project future distributions at a 1-square-kilometer resolution using regional distribution models that take squirrel,crowned wererosy finch,ranked Van as Dyke’smoderately salamander, to both changes in climate and changes in highlyAmerican sensitive marten, to andclimate northern change. flying More vegetation (habitat) into account. generalist species, such as the barred owl, black bear, Roosevelt elk, snowshoe hare, Relevant uncertainties related to the data and mountain beaver, were not ranked as and analyses are highlighted throughout highly for sensitivity to climate change. the project and database. For example, when entering species and system
132 Scanning the Conservation Horizon Case Studies 100 90 Sensitivity Score 80 70 60 50 40 30 20 10 0
Figure C7.2. Climate change sensitivity scores for selected species on the Olympic Peninsula.
Uncertainties Temporal variations in uncertainties (e.g., the uncertainty in AOGCM future There are many different types of climate• simulations tends to increase the uncertainty associated with forecasts farther out in time the predictions are from of future climate changes (Giorgi 2005) the present) and the impacts that these changes will have on species and ecosystems. These Uncertainties created by variability uncertainties are often not explicitly among AOGCM simulations (e.g., one described in climate change studies, making AOGCM• may simulate dry winter conditions for a region while another AOGCM determine how much they should rely on simulates wetter winter conditions for the particularit difficult forresults. conservation In response, planners we are to in the same region) process of identifying and describing the uncertainties associated with the different Uncertainties inherent in the vegetation results produced by our analyses. Examples and species distribution models of uncertainties include: • To compensate for the high uncertainty Uncertainties associated with model associated with future climate projections, simulations of particular variables (e.g., this project will produce six maps (one for there• tends to be more certainty in AOGCM each of the six climate change simulations) temperature simulations than in AOGCM of projected future ranges for each of precipitation simulations) the modeled species. By overlaying the projected distributions, we will be
Case Studies 133 able to highlight model agreement and magnitude and spatial character of future disagreement for each species and to climate change (i.e., exposure). The identify areas where species ranges are modeled changes in species distributions simulated to change the most. reveal how climate change may affect the future distributions of key species in Outcomes and Next Steps the region.
Both the database and modeling The preliminary results presented above components of this study make an illustrate how this sensitivity assessment important contribution to understanding process can be useful in identifying species the impacts of future climate change on and groups of species that will likely be most sensitive to climate change. As Northwest. The database assesses the demonstrated at the Olympic Peninsula the species and habitats of the Pacific expert meeting where some of the data conservation targets to climatic changes. were collected, this process can lead to Thesensitivity analyses of theof the Pacific projected Northwest’s climate useful discussions about how individual change and simulated vegetation change and groups of species may be affected provide an indication of the potential by climate change. While the results of this initial analysis are still under development, project leads are in the process of incorporating recommendations and concerns. For example, in addition to the default index equation, database
individualizedusers will also beoutput. able to weight specific characteristics and thus influence their Ultimately, this study will integrate these results and produce an assessment of the climate change vulnerability of species and
these data, we will describe and assign asystems level of in uncertainty the Pacific toNorthwest. the projected Using changes and, as in the other parts of the study, we will work with conservation scientists to produce documentation, analyses, and visual displays of data that will be useful in developing management iStockphoto and planning responses to climate change impacts. This project is committed to providing relevant and useful information and thus will continue to evolve to meet the needs of managers and decision-makers in the region.
134 Scanning the Conservation Horizon Glossary may be due to natural internal processes Glossary or external forcings, or to persistent anthropogenic changes in the composition Adaptive capacity. The ability of a system of the atmosphere or in land use to adjust to climate change (including (IPCC 2007b). climate variability and extremes) to moderate potential damages, to take Climate change adaptation. Climate advantage of opportunities, or to cope with change adaptation for natural systems the consequences (IPCC 2001b). is a management strategy that involves identifying, preparing for, and responding Adaptive management. A systematic to expected climate changes in order to approach for improving resource promote ecological resilience, maintain management by learning from management ecological function, and provide the outcomes. In principle, its purpose is to necessary elements to support biodiversity enable natural resource managers and and sustainable ecosystem services (Glick other relevant decision-makers deal et al. 2009). with uncertainty about future conditions by supporting the development of Climate model. A numerical conservation projects based on information representation of the climate system based on the physical, chemical, and biological to modify their management activities properties of its components, their available and then providing the flexibility to improve their effectiveness as new interactions and feedback processes, and information becomes available (Williams accounting for all or some of its known et al. 2007). properties (IPCC 2007a). The climate system can be represented by models of Carbon dioxide equivalent (CO2-eq). varying complexity. The concentration of carbon dioxide that would cause the same amount of radiative Confidence. forcing as a given mixture of carbon dioxide be used to characterize uncertainty that and other greenhouse gases. In the IPCC is based on expert A level judgment of “confidence” as to the can reports, CO2-eq includes the six greenhouse correctness of a model, an analysis, or a statement. Standard terminology by the 2007b). gases specified in the Kyoto Protocol (IPCC least 9 out of 10 chance of being correct, Climate. “Average weather” patterns or IPCC puts “very high confidence” at “at trends for a particular region over a period out of 10 chance” (IPCC 2005). of many years. (NCAR 2004). and “very low confidence” at “less than 1 Downscaling. A method that derives Climate change. Climate change refers local- to regional-scale (10 to 100 to a change in the state of the climate that kilometers) information from larger-scale models or data analyses. In statistical and/or the variability of its properties, downscaling, a statistical relationship is can be identified by changes in the mean and that persists for an extended period, derived between observed local climate typically decades or longer. Climate change variables and predictors at the scale of
Glossary 135 global climate model output. Dynamical Ecological thresholds. An ecological downscaling, or regional climate modeling, threshold is the point at which there is an explicitly simulates the process-based abrupt change in an ecosystem quality, physical dynamics of the regional climate property, or phenomenon, or where system using a high-resolution, limited-area small changes in an environmental driver climate model (IPCC 2007b). produce large responses in the ecosystem (Groffman et al. 2006).
Exposure. The nature and degree to which
variations (IPCC 2001b). a system is exposed to significant climate Forcing mechanisms. Anything that changes the energy balance of the earth’s system, leading to a net change in the earth’s average temperature. Examples include regular variations in the earth’s orbit, changes in ocean circulation, volcanic eruptions, and changes in the composition of the earth’s atmosphere (IPCC 2007a).
Global climate model. Global climate models are large, three-dimensional coupled models that incorporate the latest understanding of the physical processes at work in the atmosphere, oceans, and earth’s surface. They range from lower-level General Circulation Models (GCMs) to coupled Atmosphere–Ocean General Circulation Models (AOGCMs) (IPCC 2007b).
Greenhouse gas. Greenhouse gases are the gaseous constituents of the atmosphere, both natural and anthropogenic, that
wavelengths within the spectrum of thermalabsorb and infrared emit radiationradiation atemitted specific by the earth’s surface, the atmosphere itself, and by clouds. The primary greenhouse gases include water vapor (H2O), carbon dioxide (CO2), nitrous oxide (N2O), methane (CH4), George Ritchey ozone (O3), although the Kyoto Protocol also addresses several entirely human- made greenhouse gases, including sulfur
136 Scanning the Conservation Horizon Glossary Sensitivity. Sensitivity is the degree to which a system is affected, either adversely (IPCChexafluoride 2007a). (SF6), hydrofluorocarbons (HFCs), and perfluorocarbons (PFCs) (U.S. CCSP 2008b). Phenology. Recurring plant and animal or beneficially, by climate-related stimuli Uncertainty. An expression of the degree to which a value (e.g., the future state of the life cycle events, such as flowering, climate system) is unknown. Uncertainty leafing, or migration. Refugia. Physical environments that are can result from lack of information or less affected by climate change than other from disagreement about what is known areas (e.g., due to local currents, geographic or even knowable. It may have many types location, etc.) and are thus a “refuge” from climate change for organisms (U.S. CCSP 2008b). terminology,of sources, from or uncertain quantifiable projections errors in of the humandata to ambiguouslybehavior. Uncertainty defined concepts can therefore or Resilience. The amount of change or be represented by quantitative measures, disturbance that can be absorbed by a for example, a range of values calculated system [e.g., an organism, population, by various models, or by qualitative community, or ecosystem] before the judgment of a team of experts (see Moss of processes and structures (i.e., the andstatements, Schneider for 2000; example, Manning reflecting et al. the 2010; system is redefined by a different set ecosystem recovers from the change or IPCC 2007b). disturbance without a major phase shift) (U.S. CCSP 2008b). Vulnerability. The degree to which a system is susceptible to, or unable to cope Resistance. The ability of an organism, with, adverse effects of climate change, population, community, or ecosystem including climate variability and extremes. to withstand a change or disturbance It is a function of the sensitivity of a particular system to climate changes, its function. From a management perspective, exposure to those changes, and its capacity without significant loss of structure or resistance includes both (1) the concept of to adapt to those changes (IPCC 2007a). taking advantage of/boosting the inherent (biological) degree to which species are Vulnerability assessment. A key tool able to resist change and (2) manipulation for carrying out adaptation planning, of the physical environment to counteract/ and informing the development and resist physical/biological change (U.S. implementation of climate-smart resource CCSP 2008b). management practices.
Risk assessment. A risk assessment is Weather. Weather is the mix of events the process of identifying the magnitude that happen each day in our atmosphere, or consequences of an adverse event or including temperature, rainfall, and impact occurring as well as the probability humidity. that it will occur (Jones 2001).
Glossary 137 Acronyms
20C2M 20th Century Climate Coupled Model AFWA Association of Fish and Wildlife Agencies AOGCM Atmosphere-Ocean General Circulation Model AR4 Fourth Assessment Report of the IPCC AR5 Fifth Assessment Report of the IPCC BTELSS Barataria-Terrabonne Estuarine Landscape Spatial Simulation Model CCSP U.S. Climate Change Science Program CMIP Coupled Model Intercomparison Project
CO2 Carbon Dioxide DGESL Department of Geosciences Environmental Studies Laboratory DGVM Dynamic Global Vegetation Model EPA U.S. Environmental Protection Agency ESA Endangered Species Act ET Evapotranspiration FWS U.S. Fish and Wildlife Service GAP Gap Analysis Program GARP Genetic Algorithm for Rule-set Production GCM General Circulation Model HVA Habitat Vulnerability Assessment IPCC Intergovernmental Panel on Climate Change ISAB Independent Scientific Advisory Board IUCN International Union for Conservation of Nature LiDAR Light Detection and Ranging MAGICC Model for the Assessment of Greenhouse-gas Induced Climate Change MAPSS Mapped Atmospheric-Plant Soil System MDFW Massachusetts Division of Fisheries and Wildlife NARCCAP North American Regional Climate Change Assessment Program NECIA Northeast Climate Impacts Assessment NRC National Research Council PDSI Palmer Drought Severity Index ppm Parts Per Million RCP Representative Concentration Pathway RHESSys The Regional Hydro-Ecologic Simulation System RMRS Rocky Mountain Research Station SGCN Species of Greatest Conservation Need SGNC Species of Greatest Need for Conservation SLAMM Sea Level Affecting Marshes Model SRES Special Report on Emissions Scenarios SWCCI Southwest Climate Change Initiative T&E Threatened and Endangered TNC The Nature Conservancy USDA U.S. Department of Agriculture USGCRP U.S. Global Change Research Program UV Ultraviolet VEMAP The Vegetation/Ecosystem Modeling and Analysis Project VIC Variable Infiltration Capacity
138 Scanning the Conservation Horizon Resources Climate Change Vulnerability Assessment Resources
Publications items/2674.php.adaptation/nairobi_work_programme/ Vulnerability Assessment knowledge_resources_and_publications/ Frameworks and Guidance U.S. Environmental Protection Agency (EPA). 2009. A Framework for Categorizing Intergovernmental Panel on Climate the Relative Vulnerability of Threatened Change (IPCC). 2005. Guidance Notes for and Endangered Species to Climate Change. Lead Authors of the IPCC Fourth Assessment EPA/600/R-09/011. National Center for Report on Addressing Uncertainties. Environmental Assessment, Washington, Intergovernmental Panel on Climate D.C. http://www.epa.gov/ncea. Change, Geneva. http://www.ipcc.ch/pdf/ supporting-material/uncertainty-guidance- U.S. Global Change Research Program note.pdf. (USGCRP). 2009. Global Climate Change Impacts in the United States. A report of the National Research Council (NRC). 2009. USGCRP. USGCRP, Washington, D.C. http:// Informing Decisions in a Changing www.globalchange.gov/usimpacts/. Climate. Panel on Strategies and Methods for Climate-Related Decision Support, Williams, S.E., L.P. Shoo, J.L. Isaac, A.A. Committee on the Human Dimensions of Hoffmann, and G. Langham. 2008. Towards Global Change. Division of Behavioral and an integrated framework for assessing the Social Sciences and Education. National vulnerability of species to climate change. Academies Press, Washington, D.C. http:// PloS Biology 6: 2621–2626. http://www. plosbiology.org/article/info:doi/10.1371/ id=12626&page=R1. www.nap.edu/openbook.php?record_ journal.pbio.0060325.
Olmos, S. 2001. Vulnerability and General Climate Change Adaptation to Climate Change: Concepts, Issues, Assessment Methods. Climate Change Adaptation Principles Knowledge Network, International Institute and Planning for Sustainable Development, Winnipeg. Association of Fish and Wildlife Agencies (AFWA). 2009. Voluntary Guidance for Conventionhttp://www.cckn.net/pdf/va_foundation_ on Climate Change (UNFCCC). States to Incorporate Climate Change 2008.final.pdf. Compendium United Nations on Methods Framework and Tools into State Wildlife Action Plans and Other to Evaluate Impacts of, Vulnerability and Management Plans. AFWA, Washington, Adaptation to, Climate Change. UNFCC, Bonn, Germany. http://unfccc.int/ D.C. http://www.fishwildlife.org/pdfs/ ClimateChangeGuidance%20Document_ Final_reduced%20size.pdf.
Resources 139 Glick, P., A. Staudt, and B. Stein. http://cses.washington.edu/cig/outreach/ 2009. A New Era for Conservation: Review of Climate Change Adaptation etal-2009.pdf. Literature. National Wildlife Federation, seminarfiles/2009seminars/Mawdsley- Washington, D.C. http://www.nwf.org/ National Research Council (NRC). 2010. News-and-Magazines/Media-Center/ Adapting to the Impacts of Climate Change. Reports/Archive/2009/~/media/ America’s Climate Choices: Panel on PDFs/Global Warming/Reports/ Adapting to the Impacts of Climate Change. CimateChangeAdaptationLiterature National Academies Press, Washington, Review.ashx. D.C. http://americasclimatechoices.org/ paneladaptation.shtml. Groves, C., M. Anderson, C. Enquist, E. Girvetz, T. Sandwith, L. Schwarz, and U.S. Climate Change Science Program R. Shaw. 2010. Climate Change and (U.S. CCSP). 2008. Preliminary Review of Conservation: A Primer for Assessing Adaptation Options for Climate-sensitive Impacts and Advancing Ecosystem-based Ecosystems and Resources. A report by Adaptation in The Nature Conservancy. the U.S. Climate Change Science Program The Nature Conservancy, Arlington, VA. and the Subcommittee on Global Change http://conserveonline.org/workspaces/ Research. S.H. Julius and J.M. West (eds.); climateadaptation/documents/a-primer- J.S. Baron et al. (authors). U.S. EPA, for-assessing-impacts/view.html. Washington, D.C. http://downloads. climatescience.gov/sap/sap4-4/sap4-4- Heller, N.E., and E.S. Zavaleta. 2009. Biodiversity management in the face of climate change: A review of 22 West,final-report-all.pdf. J.M., S.H. Julius, P. Kareiva, C. Enquist, years of recommendations. Biological J.J. Lawler, B. Petersen, A.E. Johnson, and Conservation 142: 14–32. http://www. M.R. Shaw. 2009. U.S. natural resources and lagunadesantarosa.org/knowledgebase/ climate change: Concepts and approaches for management adaptation. Environmental Zavaleta%202009.pdf. Management 44: 1001–1021. http://www. sites/default/files/Heller%20and%20 ncbi.nlm.nih.gov/pmc/articles/PMC2791483/ Lawler, J.J. 2009. Climate change adaptation strategies for resource management and conservation planning. Annals of the New Williams,pdf/267_2009_Article_9345.pdf. B.K., R.C. Szaro, and C.D. Shapiro. York Academy of Sciences 1162: 79–98. 2007. Adaptive Management: The U.S. http://training.fws.gov/branchsites/lkm/ Department of the Interior Technical Guide. Adaptive Management Working Group, U.S. Department of the Interior, Washington, Mawdsley,climate_change/june_09/cc-adaptreview.pdf. J.R., R. O’Malley, and D.S. D.C. http://www.doi.gov/initiatives/ Ojima. 2009. A review of climate- AdaptiveManagement/TechGuide.pdf. change adaptation strategies for wildlife management and biodiversity conservation. Conservation Biology 23: 1080–1089.
140 Scanning the Conservation Horizon Resources Web-Based Tools NatureServe Climate Change Vulnerability Index: http://www.natureserve.org/ Bioclimatic metrics: WorldClim, prodServices/climatechange/ http://www.worldclim.org/. ClimateChange.jsp. Climate Adaptation Knowledge Niche and occupancy ecological Environment (CAKE): response models: http://www.cakex.org/ GARP, http://www.nhm.ku.edu/ desktopgarp; Maxent, http://www. ClimateWizard: cs.princeton.edu/~schapire/maxent/; http://www.climatewizard.org. Regression Trees and Random Forests, http://rattle.togaware.com/rattle- Compendium of conceptual ecological download.html; Bioclim, http://software. response models: informer.com/getfree-bioclim-download- software/. http://www.fileheap.com/software/ NOAA Coastal Climate Adaptation: Droughtconceptual_data_model.html. indices and hydrologic models: http://collaborate.csc.noaa.gov/ http://drought.unl.edu/whatis/indices. climateadaptation/default.aspx http://www.hydro.washington.edu/ USDA Forest Service Climate Change Lettenmaier/Models/VIC/.htm; Variable Infiltration Capacity model, Resource Center: http://www.fs.fed.us/ccrc/. Ecological models: The Regional Hydro- Ecologic Simulation System (RHESsys), setup/downloads/downloads.html; PnET, http://www.pnet.sr.unh.edu/download.http://fiesta.bren.ucsb.edu/~rhessys/ html; the CENTURY model, http://nrel. colostate.edu/projects/century5/.
Expert opinion ecological response models: Bayesian Analysis Toolkit, http://www.mppmu.mpg.de/bat/; Treeage Pro, http://www.treeage.com/products/ index.html; Delphi Decision Aid, http:// armstrong.wharton.upenn.edu/delphi2/.
GAP Program models: http://www.nbii. gov/portal/server.pt/community/maps_ Landscapeand_data/1850/species_modeling/7000. model: PATCH, http://www.epa.gov/wed/pages/ Rick Kessler news/03June/schumaker.htm.
Resources 141 Literature Cited
Allen, C.D., A.K. Macalady, H. Chenchouni, D. Bachelet, N. McDowell, M. Vennetier, T. Kitzberger, A. Rigling, D.D. Breshears, E.H. Hogg, P. Gonzalez, R. Fensham, Z. Zhang, J. Castro, N. Demidova, J.H. Lim, G. Allard, S.W. Running, A. Semerci, and N. Cobb. 2010. A global overview of drought and heat- induced tree mortality reveals emerging climate change risks for forests. Forest Ecology and Management 259: 660–684.
Allen, R.G., L.S. Pereira, D. Raes, and M. Smith. 1998. Crop Evapotranspiration: Guidelines for Computing Crop Water Requirements. FAO Irrigation and Drainage Paper No. 56. FAO, Rome.
Anderson, M.G. and C.E. Ferree. 2010. Conserving the Stage: Climate Change and the Geophysical Underpinnings of Species Diversity. PLoS ONE 5: e11554. doi:10.1371/journal.pone.0011554
Association of Fish and Wildlife Agencies (AFWA). 2009. Voluntary Guidance for States to Incorporate Climate Change George Ritchey into State Wildlife Action Plans and Other Aber, J.D., S.V. Ollinger, C.A. Federer, P.B. Management Plans. Association of Fish Reich, M.L. Goulden, D.W. Kicklighter, and Wildlife Agencies, Climate Change J.M. Melillo, and R.G. Lathrop Jr. 1995. Wildlife Action Plan Working Group, Predicting the effects of climate change Washington, D.C. on water yield and forest production in the Northeastern U.S. Climate Research 5: Aubry, K.B., K.S. McKelvey, and J.P. Copeland. 207–222. 2007. Distribution and broadscale habitat relations of the wolverine in the AchutaRao, K.M., and K.R. Sperber. 2002. contiguous United States. Journal of Wildlife Simulation of the El Niño Southern Management 71: 2147–2158. Oscillation: Results from the Coupled Model Intercomparison Project. Climate Bachelet, D., J.M. Lenihan, C. Daly, R.P. Dynamics 19: 191–209. doi:10.1007/ Neilson, D.S. Ojima, and W.J. Parton. 2001. s00382-001-0221-9. MC1: A Dynamic Vegetation Model for
142 Scanning the Conservation Horizon Literature Cited Estimating the Distribution of Vegetation Beier, P., and B. Brost. 2010. Use of and Associated Ecosystem Fluxes of land facets to plan for climate change: Carbon, Nutrients, and Water. Technical Conserving the arenas, not the actors. documentation. Version 1.0. U.S. Forest Conservation Biology 24: 701–710.
Corvallis, OR. Beissinger, S.R.J., M. Reed, J.M. Wunderle Jr., Service Pacific Northwest Research Station, S.K. Robinson, and D.M. Finch. 2000. Report Bagne, K.E., and D.M. Finch. 2008. of the AOU Conservation Committee on the Developing a tool to assess wildlife species Partners in Flight species prioritization vulnerability to climate change. In: New plan. Auk 117: 549–561. Mexico Forestry and Climate Change Workshop, Albuquerque, NM; November Belote, R.T., J.F. Weltzin, and R.J. Norby. 20, 2008. Forest Guild, Santa Fe, NM. http:// 2003. Response of an understory plant community to elevated CO2 depends on Finch.pdf (accessed August 6, 2010). differential responses of dominant invasive www.forestguild.org/climate_change/ species and is mediated by soil water Bailey, R. 2009. Ecosystem Geography: availability. New Phytologist 161: 827–835. From Ecoregions to Sites. Second edition. Springer, New York. Berg, M.P., E.T. Kiers, G. Driessen, M. van der Heijden, B.W. Kooi, F. Kuenen, M. Ballesteros-Barrera, C., E. Martinez-Meyer, Liefting, H.A. Verhoef, and J. Ellers. 2010. and H. Gadsden. 2007. Effects of land-cover Adapt or disperse: Understanding species transformation and climate change on the persistence in a changing world. Global distribution of two microendemic lizards, Change Biology 16: 587–598. Genus Uma, of northern Mexico. Journal of Herpetology 44: 733–740. Berliner, L.M., R.A. Levine, and D.J. Shea. 2000. Bayesian climate change assessment. Bamber, J.L., R.E.M. Riva, B.L.A. Vermeersen, Journal of Climate 13: 3805–3820. and A.M. LeBrocq. 2009. Reassessment of the potential sea-level rise from a collapse Bernardo, J., R.J. Ossola, J. Spotila, and K.A. of the West Antarctic Ice Sheet. Science 15: Crandall. 2007. Interspecies physiological 901-903. variation as a tool for cross-species assessments of global warming-induced Baum, A.K., K.J. Haynes, F.P. Dillemuth, and endangerment: Validation of an intrinsic J.T. Cronin. 2004. The matrix enhances the determinant of macroecological and effectiveness of corridors and stepping phylogeographic structure. Biology Letters stones. Ecology 85: 2671–2676. 3: 695–699.
Bearhop, S., W. Fielder, R.W. Furness, S.C. Biro, P.A., J.R. Post, and D.J. Booth. Votier, S. Waldron, J. Newton, G.J. Bowen, 2007. Mechanisms for climate-induced P. Berthold, and K. Farnsworth. 2005. Assortative mating as a mechanism for experiments. Proceedings of the National rapid evolution of a migratory divide. Academymortality of of Sciences fish populations of the United in whole-lake States of Science 310: 502–504. America 104: 9715–9719.
Literature Cited 143 Blaustein, A.R., S.C. Walls, B.A. Bancroft, J.J. Boyd, J. 2007. The endpoint problem. Lawler, C.L. Searle, and S.S. Gervasi. 2010. Resources 165: 25–28. Direct and indirect effects of climate change on amphibian populations. Diversity 2: Bradley, M.P., and E.R. Smith. 2004. 281–313. Using science to assess environmental vulnerabilities. Environmental Monitoring Bohle, H.G., T.E. Downing, and M.J. and Assessment 94: 1–7. Watts. 1994. Climate change and social vulnerability: Towards a sociology and Bradshaw, W.E., and C.M. Holzapfel. 2006. geography of food insecurity. Global Evolutionary response to rapid climate Environmental Change 4: 37–48. change. Science 312: 1477–1478.
Bosch, J., L.M. Carrascal, L. Durán, S. Walker, Breshears, D.D., N.S. Cobb, P.M. Rich, K.P. and M.C. Fisher. 2007. Climate change and Price, C.D. Allen, R.G. Balice, W.H. Romme, outbreaks of amphibian chytridiomycosis J.H. Kastens, M.L Floyd, J. Belnap, J.J. in a montane area of Central Spain: Is there Anderson, O.B. Myers, and C.W. Meyer. a link? Proceedings of the Royal Society B: 2005. Regional vegetation die-off in Biological Sciences 274: 253–260. response to global-change-type drought. Proceedings of the National Academy of Bosque Working Group. 2008. Our bosque Sciences of the United States of America 102: forests in a changing climate. In: New 15144–15148. Mexico Forestry and Climate Change Workshop, Albuquerque, NM; November 20, 2008. Forest Guild, Santa Fe, NM. http:// pollution in eliciting the “bleaching” responseBrown, B.E. of 2000.symbiotic The cnidarians.significance of Bosque.pdf (accessed August 9, 2010). International Journal of Environment and www.forestguild.org/climate_change/ Pollution 13: 392–415. Both, C., S. Bouwhuis, C.M. Lessells, and M.E. Visser. 2006. Climate change and Brown, D.E., F. Reichenbacher, and E. population declines in a long-distance Franson. 1998. A Classification of North migratory bird. Nature 441: 81. American Biotic Communities. University of Utah Press, Salt Lake City. Botkin, D.B., H. Saxe, M.B. Araújo, R. Betts, R.H.W. Bradshaw, T. Cedhagen, P. Chesson, Brown, J.H. 1995. Macroecology. University T.P. Dawson, J.R. Etterson, D.P. Faith, S. of Chicago Press, Chicago. Ferrier, A. Guisan, A. Skjoldborg Hansen, D.W. Hilbert, C. Loehle, C. Margules, M. Burkett, V.R., D.A. Wilcox, R. Stottlemyer, New, M.J. Sobel, and D.R.B. Stockwell. 2007. W. Barrow, D. Fagre, J. Baron, J. Price, Forecasting the effects of global warming J.L. Nielsen, C.D. Allen, D.L. Peterson, G. on biodiversity. BioScience 57: 227–236. Ruggerone, and T. Doyle. 2005. Nonlinear dynamics in ecosystem response to Box, G.E.P., and N.R. Draper. 1987. Empirical climatic change: Case studies and policy Model-building and Response Surfaces. implications. Ecological Complexity 2: Wiley Series in Probability and Statistics, 357–394. New York.
144 Scanning the Conservation Horizon Literature Cited Cahoon, D.R., and G.R. Guntenspergen. Christensen, J.H., B. Hewitson, A. Busuioc, 2010. Climate change, sea-level rise, A. Chen, X. Gao, I. Held, R. Jones, R.K. and coastal wetlands. National Wetlands Kolli, W.-T. Kwon, R. Laprise, V. Magaña Newsletter 32: 8–12. Rueda, L. Mearns, C.G. Menéndez, J. Räisänen, A. Rinke, A. Sarr, and P. Whetton. Caldeira, K., and M.E. Wickett. 2003. 2007. Regional climate projections. p. Oceanography: Anthropogenic carbon and 847–940. In: Climate Change 2007: The ocean pH. Nature 425: 365. Physical Science Basis. S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, Carilli, J.E., R.D. Norris, B.A. Black, S.M. K.B. Averyt, M. Tignor, and H.L. Miller Walsh, and M. McField. 2009. Local (eds.) Contribution of Working Group I stressors reduce coral resilience to to the Fourth Assessment Report of the bleaching. PLoS One 4: e6324. doi:10.1371/ Intergovernmental Panel on Climate journal.pone.0006324. Change. Cambridge University Press, Cambridge, UK. Carroll, A.L., S.W. Taylor, J. Rénière, and L. Safranyik. 2003. Effects of climate change Christensen, L., C.L. Tague, and J.S. Baron. on range expansion by the mountain pine 2008. Spatial patterns of simulated beetle in British Columbia. p. 223–232. transpiration response to climate In: Mountain Pine Beetle Symposium: variability in a snow dominated mountain Challenges and Solutions, Kelowna, British ecosystem. Hydrological Processes 22: Columbia; October 30–31, 2003. Natural 3576–3588. Resources Canada, Canadian Forest Service, Church, J.A., N.J. White, R. Coleman, BC-X-399, Victoria, BC. K. Lambeck, and J.X. Mitrovica. 2004. Pacific Forestry Centre, Information Report Estimates of the regional distribution of Cartron, J.-L., D.C. Lightfoot, J.E. Mygatt, sea level rise over the 1950-2000 period. S.L. Brantley, and T.K. Lowrey. 2008. A Journal of Climate 17: 2609-2625. Field Guide to the Plants and Animals of the Middle Rio Grande Bosque. University of New Mexico Press, Albuquerque.
Chapman, W.L., and J.E. Walsh. 2007. Simulations of Arctic temperature and pressure by global coupled models. Journal of Climate 20: 609–632.
Chiba, S. 1998. A mathematical model for long-term patterns of evolution: Effects of environmental stability and instability on macroevolutionary patterns and mass extinctions. Paleobiology 24: 336–348.
Bill Houghton
Literature Cited 145 Clough, J.S., R.A. Park, and R. Fuller. Covey, C., K.M. AchutaRao, U. Cubasch, 2010. SLAMM 6 beta technical P. Jones, S.J. Lambert, M.E. Mann, T.J. documentation. Warren Pinnacle Phillips, and K.E. Taylor. 2003. An overview Consulting, Inc., Warren, VT. of results from the Coupled Model Intercomparison Project. Global and Cole, K.L. 2010. Vegetation response to Planetary Change 37: 103–133. early Holocene warming as an analog for current and future changes. Conservation Craft, C., J. Clough, J. Ehman, S. Joye, R. Park, Biology 24: 29–37. S. Pennings, H. Guo, and M. Machmuller. 2009. Forecasting the effects of accelerated sea-level rise on tidal marsh ecosystem services. Frontiers in Ecology and the Environment 7: 73–78.
Cramer, W., A. Bondeau, F.I. Woodward, C. Prentice, R.A. Betts, V. Brovkin, P.M. Cox, V. Fisher, J.A. Foley, A.D. Friend, C. Kucharik, M.R. Lomas, N. Ramankutty, S. Sitch, B. Smith, A. White, and C. Young-Molling. 2001. Global response of terrestrial ecosystem structure and function to CO2 and climate change: Results from six dynamic global vegetation models. Global Change Biology 7: 357–373.
Curry, R., B. Dickson, and I. Yashayaev. 2003. Diadem A change in the freshwater balance of the Comer, P., D. Faber-Langendoen, R. Evans, Atlantic Ocean over the past four decades. S. Gawler, C. Josse, G. Kittel, S. Menard, M. Nature 426: 826–829. Pyne, M. Reid, K. Schultz, K. Snow, and J. Teague. 2003. Ecological Systems of the Daly, C., R.P. Nielson, and D.L. Phillips. United States: A Working Classification 1994. A statistical-topographic model for of U.S. Terrestrial Systems. NatureServe, mapping climatological precipitation over Arlington, VA. mountainous terrain. Journal of Applied Meteorology 33: 140–158. Copeland, J.P., K.S. McKelvey, K.B. Aubry, A. Landa, J. Persson, R.M. Inman, J. Krebs, E. Department of Geosciences Environmental Lofroth, H. Golden, J.R. Squires, A. Magoun, Studies Laboratory (DGESL). 2003. Climate M.K. Schwartz, J. Wilmot, C.L. Copeland, change and sea level: Maps of susceptible R.E. Yates, I. Kojola, and R. May. 2010. The areas. University of Arizona, Tucson. http:// bioclimatic envelope of the wolverine (Gulo www.geo.arizona.edu/dgesl/research/ gulo): Do climatic constraints limit its geographic distribution? Canadian Journal of Zoology 88: 233–246. (accessedother/climate_change_and_sea_level/ March 10, 2010). sea_level_rise/sea_level_rise_technical.htm
146 Scanning the Conservation Horizon Literature Cited Donner, S.D., T.R. Knutson, and M. Conservation Implications of Emerging Oppenheimer. 2006. Model-based Moisture Stress Due to Recent Climate assessment of the role of human-induced Changes in the Southwestern United States. climate change in the 2005 Caribbean coral Technical report. The Nature Conservancy, bleaching event. Proceedings of the National Arlington, VA and Wildlife Conservation Academy of Sciences of the United States of Society, Bronx, NY. America 104: 5483–5488. Enquist, C., and D. Gori. 2008. Implications Downing, T.E., and A. Patwardhan. 2003. of Recent Climate Change on Conservation Vulnerability Assessment for Climate Priorities in New Mexico. Technical report. Adaptation. APF Technical Paper 3. The Nature Conservancy and Wildlife United Nations Development Programme, Conservation Society, Santa Fe, NM. New York. Environmental Systems Research Institute. Dunham, J., B. Rieman, and G. Chandler. 2005. ArcMap 9.1. ESRI, Redlands, CA. environmental variables on the distribution Fahrig, L. 2007. Non-optimal animal of2003. bull Influences trout within of streamstemperature at the and southern movement in human-altered landscapes. margin of its range. North American Journal Functional Ecology 21: 1003–1015. of Fisheries Management 23: 894–904. Fay, P.A., and M.J. Schultz. 2009. Easterling, D.R. 2000. Climate extremes: Germination, survival, and growth of grass Observations, modeling, and impacts. and forb seedlings: Effects of soil moisture Science 289: 2068–2074. variability. Acta Oecologica 35: 679–684.
Easterling, W.E., B.H. Hurd, and J.B. Smith. Field, C.B., L.D. Mortsch, M. Brklacich, D.L. 2004. Coping with Global Climate Change: Forbes, P. Kovacs, J.A. Patz, S.W. Running, The Role of Adaptation in the United States. and M.J. Scott. 2007. North America. p. Pew Center on Global Climate Change, 617–652. In: Climate Change 2007: Impacts, Arlington, VA. Adaptation and Vulnerability. M. Parry, O.F. Canziani, J.P. Palutikof, P.J. van der Linden, Elmqvist, T., C. Folke, M. Nyström, G. and C.E. Hanson (eds.) Contribution of Peterson, J. Bengtsson, B. Walker, and Working Group II to the Fourth Assessment J. Norberg. 2003. Response diversity, Report of the Intergovernmental Panel ecosystem change, and resilience. Frontiers on Climate Change. Cambridge University in Ecology and the Environment 1: 488–494. Press, Cambridge, UK.
Emanuel, K. 2005. Increasing Flannigan, M.D., B.J. Stocks, and B.M. destructiveness of tropical cyclones over Wotton. 2000. Climate change and forest the past 30 years. Nature 436: 686–688. Science of the Total Environment 262: 221–229. Enquist, C., E. Girvetz, and D. Gori. 2008. A fires. Climate Change Vulnerability Assessment Foden, W., G. Mace, J.-C. Vié, A. Angulo, for Biodiversity in New Mexico, Part 2: S. Butchart, L. DeVantier, H. Dublin, A. Gutsche, S. Stuart, and E. Turak. 2008.
Literature Cited 147 Species susceptibility to climate change Galbraith, H., R. Jones, R. Park, J. Clough, S. impacts. p. 77–87. In: Wildlife in a Changing Herrod-Julius, B. Harrington, and G. Page. World: An Analysis of the 2008 IUCN Red List 2002. Global climate change and sea-level of Threatened Species. J.-C. Vié, C. Hilton- rise: Potential losses of intertidal habitat Taylor, and S.N. Stuart (eds.) IUCN, Gland, for shorebirds. Waterbirds 25: 173–183. Switzerland. Gallopín, G. 2007. Linkages between Folke, C., S. Carpenter, B. Walker, M. Sheffer, vulnerability, resilience, and adaptive T. Elmqvist, L. Gunderson, and C.S. Holling. capacity. In: Formal Approaches to 2004. Regime shifts, resilience, and Vulnerability workshop, Potsdam Institute biodiversity in ecosystem management. for Climate Impact Research, Potsdam; Annual Review of Ecology and Systematics September 13–14, 2007. 35: 557–581.
Frederiksen, M., M.P. Harris, F. Daunt, change effects on Southwest water P. Rothery, and S. Wanless. 2004. Scale- resources.Garfin, G., and Southwest M. Lenart. Hydrology 2007. Climate 6: 16–17, 34. dependent climate signals drive breeding phenology of three seabird species. Global Garzón, M.B., R. Blazek, M. Neteler, R. Change Biology 10: 1214–1221. Sánchez de Dios, H.S. Ollero, and C. Furlanello. 2006. Predicting habitat Fuhlendorf, S.D., and D.M. Engle. 2001. suitability with machine learning models: Restoring heterogeneity on rangelands: The potential area of Pinus sylvestris L. in Ecosystem management based on the Iberian Peninsula. Ecological Modelling evolutionary grazing patterns. BioScience 197: 383–393. 51: 625–632. Gawler, S.C., M.G. Anderson, A.P. Olivero, Füssel, H.-M. 2007. Vulnerability: A and M. Clark. 2008. The Northeast Habitat generally applicable conceptual framework Classification and Mapping Project Final for climate change research. Global Report. A report to the Virginia Department Environmental Change 17: 155–167. of Game and Inland Fisheries on behalf of the Northeast Association of Fish and Füssel, H.-M., and R.J.T. Klein. 2005. Climate Wildlife Agencies for the National Fish change vulnerability assessments: An and Wildlife Foundation (NFWF Project evolution of conceptual thinking. Climatic 2006-0181-003). NatureServe, Boston. Change 75: 301–329. Giorgi, F. 2005. Climate change prediction. Galatowitsch, S., L. Frelich, and L. Phillips- Climatic Change 73: 239–265. Mao. 2009. Regional climate change adaptation strategies for biodiversity Glick, P., J. Clough, and B. Nunley. 2008a. conservation in a midcontinental region Sea-level Rise and Coastal Habitats in the of North America. Biological Conservation Chesapeake Bay Region: Technical Report. 142: 2012–2022. National Wildlife Federation, Reston, VA.
148 Scanning the Conservation Horizon Literature Cited Glick, P., A. Staudt, and B. Nunley. 2008b. Sea-level Rise and Coastal Habitats of the Chesapeake Bay: A Summary. National Wildlife Federation, Reston, VA.
Glick, P., A, Staudt, and B. Stein. 2009. A New Era for Conservation: Review of Climate Change Adaptation Literature. National Wildlife Federation, Reston, VA.
Gottard, K., and S. Nylin. 1995. Adaptive plasticity and plasticity as an adaptation: A selective review of plasticity in animal morphology and life history. Oikos 74: 3–17.
Grimsditch, G.D. and R.V. Salm. 2006. Coral Reef Resilience and Resistance to Bleaching. The International Union for the Conservation of Nature and Natural Resources and The Nature Conservancy, Gland, Switzerland.
Groffman, P.M., J.S. Baron, T. Blett, A.J. Gold, I. Goodman, L.H. Gunderson, B.M. Levinson, M.A. Palmer, H.W. Paerl, G.D. Peterson, N.L. Poff, D.W. Refeski, J.F. Reynolds,
M.G. Turner, K.C. Weathers, and J. Wiens. Ingrid Taylar 2006. Ecological thresholds: The key to Groves, C., M. Anderson, C. Enquist, E. successful environmental management Girvetz, T. Sandwith, L. Schwarz, and or an important concept with no practical R. Shaw. 2010. Climate Change and application. Ecosystems 9: 1–13. Conservation: A Primer for Assessing Impacts and Advancing Ecosystem-based Grossman, D.H., D. Faber-Langendoen, Adaptation in The Nature Conservancy. The A.S. Weakley, M. Anderson, P. Bourgeon, Nature Conservancy, Arlington, VA. R. Crawford, K. Goodin, S. Landaal, K. Metzler, K. Patterson, M. Pyne, M. Ried, Gunderson, L.H. 2000. Ecological resilience and L. Sneddon. 1998. International – in theory and application. Annual Review Classification of Ecological Communities: of Ecological Systems 31: 425-439. Terrestrial Vegetation of the United States. Volume I. The National Vegetation Hagy, J.D., W.R. Boynton, C.W. Keefe, and K.V. Classification System: Development, Status, Wood. 2004. Hypoxia in Chesapeake Bay, and Applications. The Nature Conservancy, 1950–2001: Long-term change in relation Arlington, VA. Estuaries 27: 634–658. to nutrient loading and river flow.
Literature Cited 149 Halofsky, J.E., et al. In press. Climate change Chem, to stream and soil chemistry in an and wildlife habitat management at alpine watershed. Ecological Modelling 200: Olympic National Forest and Park. General 493–510. Technical Report. USDA Forest Service. Harvell, C.D., K. Kim, J.M. Burkholder, Hamon, W.R. 1961. Estimating potential R.R. Colwell, P.R. Epstein, D.J. Grimes, E.E. evapotranspiration. Journal of the Hoffmann, E.K. Lipp, A.D.M.E. Osterhaus, Hydraulics Division 87: 107–120. R.M. Overstreet, J.W. Porter, G.W. Smith, and G.R. Vasta. 2007. Emerging marine Handmer, J.W., S. Dovers, and T.E. Downing. diseases—Climate links and anthropogenic 1999. Societal vulnerability to climate factors. Science 285: 1505–1510. change and variability. Mitigation and Adaptation Strategies for Global Change Hayhoe, K., C. Wake, B. Anderson, X.-L. 4: 267–281. Liang, E. Maurer, J. Zhu, J. Bradbury, A. DeGaetano, A. Stoner, and D. Wuebbles. Hannah, L., G. Midgley, S. Andelman, M. 2008. Regional climate change projections Araújo, G. Hughes, E. Martinez-Meyer, R. for the Northeast USA. Mitigation and Pearson, and P. Williams. 2007. Protected Adaptation Strategies for Global Change 13: area needs in a changing climate. Frontiers 425–436. doi:10.1007/s11027-007-9133-2. in Ecology and the Environment 5: 131–138. Hayhoe, K., C. Wake, T. Huntington, L. Hansen, L.J. and J.R. Hoffman. 2011. Climate Savvy: Adapting Conservation and Resource Anderson, J. Bradbury, A. DeGaetano, T. Management to a Changing World. Island Troy,Luo, M. and Schwartz, D. Wolfe. J. 2006. Sheffield, Past E.and Wood, future B. Press, Washington, D.C. changes in climate and hydrological indicators in the U.S. Northeast. Climate Hansen, L., and J. Biringer. 2003. Overview. Dynamics 28: 381–407. Building resistance and resilience to climate change. p. 9–14. In: Buying Time: A Heemskerk, M., K. Wilson, and T. Pavao- User’s Manual for Building Resistance and Zuckerman. 2003. Conceptual models as Resilience to Climate Change in Natural tools for communication across disciplines. Systems. L.J. Hansen, J.L. Biringer, and J.R. Conservation Ecology 7(3): Article 8. Hoffman (eds.) World Wildlife Fund, Berlin. http://www.consecol.org/vol7/iss3/art8 (accessed August 9, 2010). Harrison, P.A., P.M. Berry, N. Butt, and M. New. 2006. Modeling climate change Heller, N.E., and E.S. Zavaleta. 2009. impacts on species’ distributions at Biodiversity management in the face of the European scale: Implications for climate change: A review of 22 years of conservation policy. Environmental Science recommendations. Biological Conservation & Policy 9: 116–128. 142: 14–32.
Hartman, M.D., J.S. Baron, and D.S. Ojima. Hoegh-Guldberg, O., P.J. Mumby, A.J. Hooten, 2007. Application of a coupled ecosystem- chemical equilibrium model, DayCent- Harvell, P.F. Sale, A.J. Edwards, K. Caldeira, N.R.S. Knowlton, Steneck, P.C.M. Greenfield, Eakin, R. E. Iglesias-Prieto, Gomez, C.D.
150 Scanning the Conservation Horizon Literature Cited N. Muthiga, R.H. Bradbury, A. Dubi, and M.E. Intergovernmental Panel on Climate Hatziolos. 2007. Coral reefs under rapid Change (IPCC). 2001b. Climate Change 2001: Impacts, Adaptation and Vulnerability. Science 318: 1737-1742. Contribution of Working Group II to climate change and ocean acidification. the Third Assessment Report of the Hughes, T.P., A.H. Baird, D.R. Bellwood, M. Intergovernmental Panel on Climate Card, S.R. Connolly, C. Folke, R. Grosberg, O. Change. Cambridge University Press, Hoegh-Guldbert, J.B.C. Jackson, J. Kleypas, Cambridge, UK. J.M. Lough, P. Marshall, M. Nyström, S.R. Intergovernmental Panel on Climate Roughgarden. 2003. Climate change, human Change (IPCC). 2005. Guidance Notes for impacts,Palumbi, andJ.M. thePandolfi, resilience B. Rosen, of coral and reefs. J. Lead Authors of the IPCC Fourth Assessment Science 301: 929-933. Report on Addressing Uncertainties. Intergovernmental Panel on Climate Hunter, M.L. 2007. Climate change and Change, Geneva, Switzerland. moving species: Furthering the debate on assisted colonization. Conservation Biology Intergovernmental Panel on Climate 21: 1356–1358. Change (IPCC). 2007a. Climate Change 2007: Synthesis Report. Contribution of Working Groups I, II, and III to the (ISAB). 2007. Climate Change Impacts on Fourth Assessment Report of the ColumbiaIndependent River Scientific Basin Fish Advisory and Wildlife. Board Intergovernmental Panel on Climate Northwest Power and Conservation Change. Core Writing Team, R.K. Pachauri, Council, Portland, OR. and A. Reisinger (eds.) Intergovernmental Panel on Climate Change, Geneva, Inouye, D.W. 2008. Effects of climate change Switzerland.
Ecology Intergovernmental Panel on Climate Change on89: phenology, 253–262. frost damage, and floral (IPCC). 2007b. Climate Change 2007: The abundance of montane wildflowers. Physical Science Basis. Contribution of Inouye, D.W., B. Bar, K.B. Armitage, and B.D. Working Group I to the Fourth Assessment Inouye. 2000. Climate change is affecting Report of the Intergovernmental Panel altitudinal migrants and hibernating on Climate Change. S. Solomon, D. Qin, M. species. Proceedings of the National Manning, Z. Chen, M. Marquis, K.B. Averyt, Academy of Sciences of the United States of M. Tignor, and H.L. Miller (eds.) Cambridge America 97: 1630–1633. University Press, Cambridge, UK.
Intergovernmental Panel on Climate Intergovernmental Panel on Climate Change (IPCC). 2001a. Climate Change Change (IPCC). 2007c. Climate Change 2001: The Scientific Basis. Contribution of 2007: Impacts, Adaptation and Vulnerability. Working Group I to the Third Assessment Contribution of Working Group II to the Report of the Intergovernmental Panel Fourth Assessment Report of the on Climate Change. Cambridge University Intergovernmental Panel on Climate Press, Cambridge, UK. Change. M.L. Parry, O.F. Canziani, J.P.
Literature Cited 151 Palutikof, P.J. van der Linden, and C.E. Hanson (eds.) Cambridge University Press, and application. Quarterly Review of Biology Cambridge, UK. 81:Jax, 237–258.K. 2006. Ecological units: Definitions
Intergovernmental Panel on Climate Johnson, B.L. 1999. Introduction to the Change (IPCC). 2007d. Climate Change special feature: Adaptive management— 2007: Mitigation. Contribution of Working Group III to the Fourth Assessment Report Conservation Ecology 3: 10. of the Intergovernmental Panel on Climate Scientifically sound, socially challenged? Change. B. Metz, O.R. Davidson, P.R. Bosch, Johnson, W.C., B.V. Millett, T. Gimanov, R.A. R. Dave, and L.A. Meyer (eds.) Cambridge Voldseth, G.R. Gunterspergen, and D.E. University Press, Cambridge, UK. Naugle. 2005. Vulnerability of northern prairie wetlands to climate change. Integrated Resource Planning Committee. BioScience 55: 863–872. 1993. Public Participation Guidelines for Land and Resource Management Planning. Jones, R.N. 2001. An environmental risk assessment/management framework BC, Canada. for climate change impact assessments. The Land Use Coordination Office, Victoria, Natural Hazards 23: 197–230. Isaac, J.L. 2009. Effects of climate change on life history: Implications for extinction risk Kareiva, P., C. Enquist, A. Johnson, S.H. in mammals. Endangered Species Research Julius, J. Lawler, B. Peterson, L. Pitelva, R. 7: 115–123. Shaw, and J.M. West. 2008. Synthesis and conclusions. p. 9.1–9.67. In: Preliminary Janetos, A., L. Hansen, D. Inouye, B.P. Kelly, Review of Adaptation Options for Climate- L. Meyerson, B. Peterson, and R. Shaw. sensitive Ecosystems and Resources. S.H. 2008. Biodiversity. p. 151–181. In: The Julius and J.M. West (eds.) A report by Effects of Climate Change on Agriculture, the U.S. Climate Change Science Program Land Resources, Water Resources, and and the Subcommittee on Global Change Biodiversity in the United States. Synthesis Research, Washington, D.C. and Assessment Product 4.3. Report by the U.S. Climate Change Science Program Kearney, M.S., R.E. Grace, and J.C. Stevenson. and the Subcommittee on Global Change 1998. Marsh loss in Nanticoke Estuary, Research. U.S. Department of Agriculture, Chesapeake Bay. Geographical Review 78: Washington, D.C. 205–220.
Janzen, F.J. 1994. Climate change and Kellermann, V., B. van Heerwaarden, temperature-dependent sex determination C.M. Sgrò, and A.A. Hoffmann. 2009. in reptiles. Proceedings of the National Fundamental evolutionary limits in Academy of Sciences of the United States of ecological traits drive Drosophila species America 91: 7487–7490. distributions. Science 325: 1244–1246.
152 Scanning the Conservation Horizon Literature Cited Kelly, A.E., and M.L. Goulden. 2008. Rapid shifts in plant distribution with recent climate change. Proceedings of the National Academy of Sciences of the United States of America 105: 11823–11826.
Kelly, P.M., and W.N. Adger. 2000. Theory and practice in assessing vulnerability to climate change and facilitating adaptation. Climatic Change 47: 325–352.
Kennedy, V.S., R.R. Twilley, J.A. Kleypas, J.H. Cowan Jr., and S.R. Hare. 2002. Coastal and Marine Ecosystems and Global Climate Change: Potential Effects on U.S. Resources. Pew Center on Global Climate Change, Arlington, VA.
Kinnison, M.T., and N.G. Hairston Jr. 2007. Eco-evolutionary conservation biology: Contemporary evolution and the dynamics of persistence. Functional Ecology 21: 444–454.
Kirwan, M.L., and G.R. Guntenspergen. 2009. Accelerated sea-level rise—A response to Craft et al. Frontiers in Ecology and the Environment 7: 126–127.
Klein, R.J.T., and R.J. Nicholls. 1999. David Bowman/USFWS Assessment of coastal vulnerability to climate change. Ambio 28: 182–187. Lavorel, S., S. McIntire, J. Landsberg, and T.D.A. Forbes. 1997. Plant functional Kuffner, I., and A. Tihansky. 2008. Coral Reef Builders Vulnerable to Ocean Acidification. classifications: From general groups to U.S. Geological Survey, Reston, VA. disturbance. Trends in Ecology & Evolution 12:specific 474–478. groups based on response to Kunkel, K.E., X. Liang, J. Zhu, and Y. Lin. 2006. Can CGCMs simulate the Lawler, J.J. 2009. Climate change adaptation twentieth-century “warming hole” in the strategies for resource management and central United States? Journal of Climate conservation planning. Annals of the New 19:4137–4153. York Academy of Sciences 1162: 79–98.
Literature Cited 153 Lawler, J.J., S.L. Shafer, D. White, P. Kareiva, Luck, G.W., G.C. Daily, and P.R. Erlich. E.P. Maurer, A.B. Blaustein, and P.J. Bartlein. 2003. Population diversity and ecosystem 2009. Projected climate-induced faunal services. Trends in Ecology & Evolution 16: change in the Western Hemisphere. Ecology 331–336. 90: 588–597. Luedeling, E., M. Zhang, and E.H. Girvetz. Lawler, J.J., T.H. Tear, C. Pyke, R. Shaw, P. 2009. Climatic changes lead to declining Gonzalez, P. Kareiva, L. Hansen, L. Hannah, winter chill for fruit and nut trees in K. Klausmeyer, A. Aldous, C. Bienz, and S. California during 1950–2099. PLoS Pearsall. 2010. Resource management in a One 4: e6166. changing and uncertain climate. Frontiers in Ecology and the Environment 8: 35–43. Lunetta, K.L., L.B. Hayward, J. Segal, doi:10.1890/070146. and P. Van Eerdewegh. 2004. Screening large-scale association study data: Lawler, J.J., D. White, R.P. Neilson, and A.R. Exploiting interactions using random Blaustein. 2006. Predicting climate-induced forests. BMC Genetics 5: Article 32. range shifts: Model differences and model doi:10.1186/1471-2156-5-32. reliability. Global Change Biology 12: 1568–1584. MacMynowski, D.P., and T.L. Root. 2007. Climate and the complexity of migratory Lee, J.K., R.A. Park, and P.W. Mausel. phenology sexes, migratory distance, and 1992. Application of geoprocessing and arrival distributions. International Journal simulation modeling to estimate impacts of Biometeorology 51: 361–373. of sea-level rise on the northeast coast of Florida. Photogrammetric Engineering and Manning, M.R., J. Edmonds, S. Emori, A. Remote Sensing 58: 1579–1586. Grubler, K. Hibbard, F. Joos, K. Kainuma, R.F. Keeling, T. Kram, A.C. Manning, M. Meinshausen, R. Moss, N. Nakicenovic, K. of climate change on the abundance, Riahi, S.K. Rose, S. Smith, R. Swart, and D.P. distribution,Leech, D.I., and and H.Q.P. phenology Crick. 2007. of woodland Influence van Vuuren. 2010. Misrepresentation of bird species in temperate regions. Ibis 149: the IPCC CO2 emission scenarios. Nature 128–145. Geoscience 3: 376–377.
Logan, J.A., and J.A. Powell. 2001. Ghost Manomet Center for Conservation Sciences forests, global warming, and the mountain and Massachusetts Division of Fisheries pine beetle. American Entomologist 47: and Wildlife (MDFW). 2010a. Climate 160–173. Change and Massachusetts Fish and Wildlife: Volume 1 Introduction and Background. Logan, J.A., J. Régnière, and J.A. Powell. Massachusetts Division of Fisheries and 2003. Assessing the impacts of global Wildlife, Westborough. warming on forest pest dynamics. Frontiers in Ecology and the Environment 1: 130–137. Manomet Center for Conservation Sciences and Massachusetts Division of Fisheries and Wildlife (MDFW). 2010b. Climate
154 Scanning the Conservation Horizon Literature Cited Change and Massachusetts Fish and Wildlife: Mawdsley, J.R., R. O’Malley, and D.S. Volume 2 Habitat and Species Vulnerability. Ojima. 2009. A review of climate- Massachusetts Division of Fisheries and change adaptation strategies for wildlife Wildlife, Westborough. management and biodiversity conservation. Conservation Biology 23: 1080–1089. Manomet Center for Conservation Sciences and Massachusetts Division of Fisheries McCarty, J.P. 2001. Ecological consequences and Wildlife (MDFW). 2010c. Climate of recent climate change. Conservation Change and Massachusetts Fish and Biology 15: 320–331. Wildlife: Volume 3 Habitat Management. Massachusetts Division of Fisheries and McLeod, E., B. Poulter, J. Hinkel, E. Wildlife, Westborough. Reyes, and R. Salm. 2010. Sea-level rise impact models and environmental Margules, C.R., and R.L. Pressey. 2000. conservation: A review of models and their Systematic conservation planning. Nature applications. Ocean & Coastal Management, 405: 243–253. doi:10.1016/j.ocecoaman.2010.06.009.
Marsh, H., A. Dennis, H. Hines, A. Kutt, McKenzie, D., Z. Gedalof, D.L. Peterson, and K. McDonald, E. Weber, S. Williams, and J. Winter. 2007. Optimizing allocation and conservation. Conservation Biology of management resources for wildlife. 18:P. Mote. 1–13. 2004. Climate change, wildfire, Conservation Biology 21: 387–399. McNulty, S.G. 2002. Hurricane impacts on Martin, T.E. 2001. Abiotic vs. biotic U.S. carbon sequestration. Environmental Pollution 116: S17–S24. species, with implications for climate change.influences Ecology on habitat 82: 175–188. selection of coexisting McRae, B.H., N.H. Schumaker, R.B. McKane, R.T. Busing, A.M. Solomon, and C.A. Burdick. Martin, T.E. 2007. Climate correlates of 20 2008. A multi-model framework for years of trophic changes in a high-elevation simulating wildlife population response riparian system. Ecology 88: 367–380. to land-use and climate change. Ecological Modelling 219: 77–91. Martinez-Meyer, E. 2005. Climate change and biodiversity: Some considerations Meinshausen, M., N. Meinshausen, W. Hare, in forecasting shifts in species’ potential S.C.B. Raper, K. Frieler, R. Knutti, D.J. Frame, distributions. Biodiversity Informatics 2: and M.R. Allen. 2006. Multi-gas emissions 42–55. pathways to meet climate targets. Climatic Change 75: 151–194. Maurer, E.P., L. Brekke, T. Pruitt, and P.B. Duffy. 2007. Fine-resolution climate Metzger, M.J., R. Leemans, and D. Schröter. projections enhance regional climate 2005. A multidisciplinary multi-scale change impact studies. Eos 88(47): 504. framework for assessing vulnerabilities doi:10.1029/2007EO470006. to global change. International Journal of Applied Earth Observation and Geoinformation 7: 253–267.
Literature Cited 155 David Cartier
Millennium Ecosystem Assessment. next generation of scenarios for climate 2005. Ecosystems and Human Well-being: change research and assessment. Nature Synthesis. Island Press, Washington, D.C. 463: 747–756.
Millsap, B.A., J.A. Gore, D.E. Runde, and S.I. Moss, R., M. Babiker, S. Brinkman, E. Calvo, Cerulean. 1990. Setting priorities for the T. Carter, J. Edmonds, I. Elgizouli, S. Emori, L. Erda, K. Hibbard, R. Jones, M. Kainuma, Florida. Wildlife Monographs 111: 1–57. J. Kelleher, J. F. Lamarque, M. Manning, B. conservation of fish and wildlife species in Matthews, J. Meehl, L. Meyer, J. Mitchell, N. Milly, P.C.D., J. Betancourt, M. Falkenmark, Nakicenovic, B. O’Neill, R. Pichs, K. Riahi, S. R.M. Hirsch, Z.W. Kundzewicz, D.P. Rose, P. Runci, R. Stouffer, D. van Vuuren, J. Lettenmaier, and R.J. Stouffer. 2008. Weyant, T. Wilbanks, J. P. van Ypersele, and Stationarity is dead: Whither water M. Zurek. 2008. Towards New Scenarios management? Science 319: 573–574. for Analysis of Emissions, Climate Change, Impacts, and Response Strategies. Technical Mitrovika, J.X., N. Gomez, and P.U. Clark. Summary. Intergovernmental Panel on Climate Change, Geneva. Antarctic collapse. 2009. Science 6: 753. 2009. The sea-level fingerprint of West Moss, R.H., and S.H. Schneider. 2000. Morris, J.T., P.V. Sundareshwar, C.T. Nietch, B. Uncertainties in the IPCC TAR: Kjerfve, and D.R. Cahoon. 2002. Responses Recommendations to lead authors for more of coastal wetlands to rising sea level. consistent assessment and reporting. p. Ecology 83: 2869–2877. 33–51. In: Guidance Papers on the Cross Cutting Issues of the Third Assessment Moss, R.H., J.A. Edmonds, K.A. Hibbard, M.R. Report of the IPCC. R. Pechauri, T. Tanguchi, Manning, S.K. Rose, D.P. van Vuuren, T.R. and K. Tanaka (eds.) World Meteorological Carter, S. Emori, K. Kainuma, T. Kram, G.A. Organization, Geneva, Switzerland. Meehl, J.F.B. Mitchell, N. Nakicenovic, K. Riahi, S.J. Smith, R.J. Stouffer, A.M. Thomson, Mote, P.W., and E.P. Salathé Jr. 2009. Future J.P. Weyant, and T.J. Wilbanks. 2010. The In: The Washington Climate Change Impacts climate in the Pacific Northwest. p. 21–43.
156 Scanning the Conservation Horizon Literature Cited Assessment: Evaluating Washington’s Future National Research Council (NRC). 2009. in a Changing Climate. Climate Impacts Informing Decisions in a Changing Group, University of Washington, Seattle. Climate. Panel on Strategies and Methods for Climate-Related Decision Support, Murray, D.L., E.W. Cox, W.B. Ballard, H.A. Committee on the Human Dimensions of Whitlaw, M.S. Lenarz, T.W. Custer, T. Barnett, Global Change. Division of Behavioral and and T.K. Fuller. 2006. Pathogens, nutritional Social Sciences and Education. National Academies Press, Washington, D.C. a declining moose population. Wildlife Monographsdeficiency, and 166: climate 1–30. influences on National Research Council (NRC). 2010. Adapting to the Impacts of Climate Change. Naeem, S. 1998. Species redundancy and America’s Climate Choices: Panel on ecosystem reliability. Conservation Biology Adapting to the Impacts of Climate Change. 12: 39–45. National Academies Press, Washington, D.C.
Najjar, R.G., H.A. Walker, P.J. Anderson, E.J. Nemani, R.R., H. Hashimoto, P. Votava, F. Barron, R.J. Bord, J.R. Gibson, V.S. Kennedy, Melton, M. White, and W. Wang. 2009. C.G. Knight, J.P. Megonigal, R.E. O’Connor, Monitoring and forecasting ecosystem C.D. Polsky, N.P. Psuty, B.A. Richards, L.G. dynamics using the Terrestrial Observation Sorenson, E.M. Steele, and R.S. Swanson. and Prediction System (TOPS). Remote 2000. The potential impacts of climate Sensing of Environment change on the Mid-Atlantic coastal region. Climate Research 14: 219–233. New, M., and M. Hulme. 113:2000. 1497−1509. Representing uncertainty in climate change scenarios: Nakicenovic, N., J. Alcamo, G. Davis, B. A Monte-Carlo approach. Integrated Assessment 1: 203–213. A. Grübler, T.Y. Jong, T. Kram, E. Lebre La Rovere,deVries, L. J. Michaelis,Fenhann, S. S. Gaffin, Mori, T.K. Morita, Gregory, Nitschke, C.R., and J.L. Innes. 2008. W. Pepper, H. Pitcher, L. Price, K. Riahi, Integrating climate change into forest A. Roehrl, H.-H. Rogner, A. Sankovski, M. management in South-Central British Schlesinger, P. Shukla, S. Smith, R. Swart, S. Columbia: An assessment of landscape van Rooijen, N. Victor, and Z. Dadi. 2000. vulnerability and development of a climate- Intergovernmental Panel on Climate Change smart framework. Forest Ecology and Special Report on Emissions Scenarios. Management 256: 313–327. Cambridge University Press, Cambridge, UK. Northeast Climate Impacts Assessment National Center for Atmospheric Research (NECIA). 2006. Climate Change in the U.S. (NCAR). 2004. Weather and climate basics. Northeast. A Report of the Northeast Climate www.eo.ucar.edu/basics/index.html Impacts Assessment. Union of Concerned (accessed March 10, 2010). Scientists, Cambridge, MA.
National Research Council (NRC). 2004. Noss, R.F. 1990. Indicators for monitoring Adaptive Management for Water Resources biodiversity: A hierarchical approach. Planning. National Academies Press, Conservation Biology 4: 355–364. Washington, D.C.
Literature Cited 157 Nussey, D.H., E. Postma, P. Grenapp, and p. 271–311. In: Global Change: Effects on M. Evisser. 2005. Selection on heritable Coniferous Forests and Grasslands. A.I. phenotypic plasticity in a wild bird Breymeyer, D.O. Hall, J.M. Melillo, and G.I. population. Science 310: 304–306. Agren (eds.) SCOPE Volume 56. John Wiley & Sons, Chichester, UK. Nylin, S., and K. Gotthard. 1998. Plasticity in life-history traits. Annual Review of Olmos, S. 2001. Vulnerability and Entomology 43: 63–83. Adaptation to Climate Change: Concepts, Issues, Assessment Methods. Climate Change Nyström, M., C. Folke, and F. Moberg. 2000. Knowledge Network, International Institute Coral reef disturbance and resilience in a for Sustainable Development, Winnipeg. human-dominated environment. Trends in Ecology & Evolution 15: 413–417. Opdam, P., and D. Wascher. 2004. Climate change meets habitat fragmentation: Nyström, M., N. Graham, J. Lokrantz, Linking landscape and biogeographical and A. Norström. 2008. Capturing the scale levels in research and conservation. cornerstones of coral reef resilience: Biological Conservation 117: 285–297. Linking history to practice. Coral Reefs 27: 795–809. Orr, J.C., V.J. Fabry, Om Aumont, L. Bopp, S.C. Doney, R.A. Feely, A. Gnanadesikan, O’Connor, R.J., M.T. Jones, D. White, C. N. Gruber, A. Ishida, F. Joos, R.M. Key, K. Honsaker, T. Loveland, B. Jones, and Lindsay, E. Maier-Remer, R. Matear, P. E. Preston. 1996. Spatial partitioning Monfray, A. Mouchet, R.G. Najjar, G.-K. of environmental correlates of avian Plattner, K.B. Rodgers, C.L. Sabine, J.L. biodiversity in the conterminous United Sarmiento, R. Schlitzer, R.D. Slater, I.J. States. Biodiversity Letters 3: 97–110. Totterdell, M.-F. Weirig, Y. Yamanoka, and A. Yool. 2005. Anthropogenic ocean Odum, E.P. 1953. Fundamentals of Ecology. W.B. Saunders, Philadelphia. impact on calcifying organisms. Nature 437:acidification 681–686. over the 21st century and its Ojima, D., and R. Corell. 2009. Adaptation to land use and global change thresholds. IOP Overpeck, J.T. and J.L. Weiss. 2009. Conference Series: Earth and Environmental Projections of future sea level becoming Science 6: 342012. doi:10.1088/1755-1307 more dire. Proceedings of the National /6/4/342012. Academy of Sciences 106: 21461-21462.
Ojima, D.S., W.J. Parton, M.B. Coughenour, Panjabi, A.O., E.H. Dunn, P.J. Blancher, W.C. J.M.O. Scurlock, T.B. Kirchner, T.G.F. Kittel, Hunter, B. Altman, J. Bart, C.J. Beardmore, D.O. Hall, D.S. Schimel, E. Garcia Moya, T.G. I.T. Berlanga, G.S. Butcher, S.K. Davis, D.W. Gilmanov, T.R. Seastedt, A. Kamnalrut, J.I. Demarest, R. Dettners, W. Easton, H. Gomez Kinyamario, S.P. Long, J.-C. Menaut, O.E. de Silva Garza, E.E. Iñigo-Elias, D.N. Pashley, Sala, R.J. Scholes, and J.A. van Veen. 1996. C.J. Ralph, T.D. Rich, K.V. Rosenberg, C.M. Impact of climate and atmospheric carbon Rustay, J.M. Ruth, J.S. Wendt, and T.C. Will. dioxide changes on grasslands of the world. 2005. The Partners in Flight Handbook on
158 Scanning the Conservation Horizon Literature Cited Species Assessment. Version 2005. Partners Poleward shifts in geographical ranges of in Flight Technical Series No. 3. http://www. rmbo.org/pubs/downloads/Handbook2005. warming. Nature 399: 579–583. pdf (accessed August 10, 2010). butterfly species associated with regional Parmesan, C., and G. Yohe. 2003. A globally Park, J. 2008. Assessing climate change under uncertainty: A Monte Carlo approach. impacts across natural systems. Nature Masters project submitted in partial 421:coherent 37–42. fingerprint of climate change of Environmental Management degree in Pearson, R.G., and T.P. Dawson. 2003. thefulfillment Nicholas of School the requirements of the Environment for the Master and Predicting the impacts of climate change on Earth Sciences of Duke University. Duke the distribution of species: Are bioclimate University, Durham, NC. http://dukespace. envelope models useful? Global Ecology and lib.duke.edu/dspace/handle/10161/525 Biogeography 12: 361–371. (accessed August 10, 2010). Peñuelas, G.E., and I. Filella. 2001. Park, R.A., M.S. Trehan, P.W. Mausel, and R.C. Responses to a warming world. Science Howe. 1989. The effects of sea-level rise on 294: 793–795. U.S. coastal wetlands. p. 1-1–1-55. In: The Potential Effects of Global Climate Change Petchey, O.L., and K.J. Gaston. 2009. Effects on the United States: Appendix B—Sea- on ecosystem resilience of biodiversity, level Rise. J.P. Smith and D.A Tirpak (eds.) extinctions, and the structure of regional EPA-230-05-89-052. U.S. Environmental species pools. Theoretical Ecology 2: Protection Agency, Washington, D.C. 177–187.
Parmesan, C. 2005. Biotic response: Peters, D.P.C., R.A. Pielke Sr., B.T. Range and abundance changes. p. 41–55. Bestelmeyer, C.D. Allen, S. Munson-McGee, In: Climate Change and Biodiversity. and K.H. Havstad. 2004. Cross-scale T.E. Lovejoy and L. Hannah (eds.) Yale interactions, nonlinearities, and forecasting University Press, New Haven, CT. catastrophic events. Proceedings of the National Academy of Sciences of the United Parmesan, C. 2006. Ecological and States of America 101: 15130–15135. evolutionary responses to recent climate change. Annual Review of Ecology, Evolution, Peters, R.L. 1992. Conservation of biological and Systematics 37: 637–669. diversity in the face of climate change. p. 15–30. In: Global Warming and Biological Parmesan, C., and H. Galbraith. 2004. Diversity. R.L. Peters and T.L. Lovejoy (eds.) Observed Impacts of Global Climate Change Yale University Press, New Haven, CT. in the U.S. Pew Center on Global Climate Change, Arlington, VA. Peterson, A.T., M.A. Ortega-Huerta, J. Bartley, V. Sánchez-Cardero, J. Soberón, R.H. Parmesan, C., N. Ryrholm, C. Stefanescu, J.K. Buddemeier, and D.R.B. Stockwell. 2002. Hill, C.D. Thomas, H. Descimon, B. Huntley, Future projections for Mexican faunas L. Kaila, J. Kullberg, T. Tammaru, W.J. under global climate change scenarios. Tennent, J.A. Thomas, and M. Warren. 1999. Nature 416: 626–629.
Literature Cited 159 Peterson, G., C.R. Allen, and C.S. Holling. Rahmstorf, S., A. Cazenave, J.A. Church, 1998. Ecological resilience, biodiversity, J.E. Hasen, R.K. Keeling, D.E. Parker, and and scale. Ecosystems 1: 6–18. R.C.J. Somerville. 2007. Recent climate observations compared to projections. Peterson, G.D., G.S. Cummings, and S.R. Science 316: 709. Carpenter. 2003. Scenario planning: A tool for conservation in an uncertain world. Rauter, C.M., H.-U. Reyer, and K. Bollmann. Conservation Biology 17: 358–366. 2002. Selection through predation, snowfall, and microclimate on nest-site Pfeffer, W.T., J.T. Harper, and S.O. O’Neel. preferences in the Water Pipit Anthus 2008. Kinematic constraints on glacier spinoletta. Ibis 144: 433–444. contributions to 21st-century sea-level rise. Science 321: 1340–1343. Réale, D., A.G. McAdam, S. Boutin, and D. Berteaux. 2003. Genetic and plastic Phillips, S.J., R.P. Anderson, and R.E. responses of a northern mammal to climate Schapire. 2006. Maximum entropy change. Proceedings of the Royal Society B: modeling of species geographic Biological Sciences 270: 591–596. distributions. Ecological Modelling 190: 231–259. Reed, D.J. 2002. Sea-level rise and coastal marsh sustainability: Geological and Pounds, J.A., M.R. Bustamante, L.A. Coloma, ecological factors in the Mississippi delta J.A. Consuegra, M.P.L. Fogden, P.N. Foster, E. plain. Geomorphology 48: 233–243. La Marca, K.L. Masters, A. Merino-Viteri, R. Puschendorf, S.R. Ron, G. Arturo Sánchez- Refsgaard, J.C., J.P. van der Sluijs, A.L. Azofeifa, C.J. Still, and B.E. Young. 2006. Højberg, and P.A. Vanrolleghen. 2007. Widespread amphibian extinctions from Uncertainty in the environmental modeling epidemic disease driven by global warming. process—A framework and guidance. Nature 439: 161–167. Environmental Modelling & Software 22: 1543–1556. Prato, T. 2009. Evaluating and managing wildlife impacts of climate change under Rehnfeldt, G.E., N.L. Crookston, M.V. uncertainty. Ecological Modelling 220: Warwell, and J.S. Evans. 2006. Empirical 923–930. analyses of plant–climate relationships for the western United States. International Primack, R.B., I. Ibañez, H. Higuchi, S.D. Lee, Journal of Plant Sciences 167: 1123–1150. A.J. Miller-Rushing, A.M. Wilson, and J.A. Reist, J.D., F.J. Wrona, T.D. Prowse, M. Power, variability in phenological responses J.B. Dempson, J.R. King, and R.J. Beamish. toSilander warming Jr. 2009. temperatures. Spatial and Biological interspecific 2006. An overview of effects of climate Conservation 142: 2569–2577. change on selected arctic freshwater and Ambio 35: 381–387.
anadromous fishes.
160 Scanning the Conservation Horizon Literature Cited Reyes, E., M.L. White, J.F. Martin, G.P. Proceedings of the National Academy of Kemp, J.W. Day, and V. Aravamuthan. 2000. Sciences of the United States of America 102: Landscape modeling of coastal habitat 7465–7469. change in the Mississippi Delta. Ecology 81: 2331–2349. Root, T.L., J.T. Price, K.R. Hall, S.H. Schneider, C. Rosenzweig, and J.A. Pounds. 2003. Ricketts, T.H. 2001. The matrix matters: Fingerprints of global warming on wild Effective isolation in fragmented animals and plants. Nature 421: 57–60. landscapes. American Naturalist 158: 87–99. Root, T.L., and S.H. Schneider. 2002. Climate change: Overview and implications for Risbey, J., and M. Kandlikar. 2007. wildlife. p. 1–56. In: Wildlife Responses to Climate Change: North American Case the IPCC uncertainty assessment process. Studies. S.H. Schneider and T.L. Root (eds.) ClimaticExpressions Change of likelihood 85: 19–31. and confidence in Island Press, Washington, DC.
Robinson, D.C.E., S.J. Beukema, and L.A. Greig. 2008. Vegetation Models and Climate Change: Workshop Results. Prepared by ESSA Technologies Ltd., for Western Wildlands Environmental Threat Assessment Center, USDA Forest Service, Prineville, OR.
Root, T., and L. Hughes. 2005. Present and future phenological changes in wild plants and animals. p. 61–69. In: Climate Change and Biodiversity. T.E. Lovejoy and L. Hannah (eds.) Yale University Press, New Haven, CT.
Root, T.L. 1988a. Energetic constraints on avian distributions and abundances. Ecology 69: 330–339.
Root, T.L. 1988b. Environmental factors Alan G. Nelson associated with avian distributional boundaries. Journal of Biogeography 15: Running, S.W., and L.S. Mills. 2009. 489–505. Terrestrial Ecosystem Adaptation. Resources for the Future, Washington, D.C. Root, T.L., D.P. MacMynowski, M.D. Mastrandrea, and S.H. Schneider. 2005. Ryan, M.G., S. Archer, R. Birdsey, C. Dahm, L. Heath, J. Hicke, D. Hollinger, T. Huxman, species changes: Joint attribution. G. Okin, R. Oren, J. Randerson, and W. Human-modified temperatures induce Schlesinger. 2008. Land resources. p.
Literature Cited 161 75–120. In: The Effects of Climate Change Schlesinger, W.H., J.F. Reynolds, G.L. on Agriculture, Land Resources, Water Cunningham, L.F. Huenneke, W.M. Resources, and Biodiversity in the United Jarrell, R.A. Virginia, and W.G. Whitford. States. Synthesis and Assessment Product 1990. Biological feedbacks in global 4.3. Report by the U.S. Climate Change Science 304: 1043–1048. Science Program and the Subcommittee on Global Change Research. U.S. Department of Schneider,desertification. S.H., S. Semenov, A. Patwardhan, Agriculture, Washington, D.C. I. Burton, C.H.D. Magadza, M. Oppenheimer, A.B. Pittock, A. Rahman, J.B. Smith, A. Sah, J.P., M.S. Ross, J.R. Snyder, and D.E. Suarez, and F. Yamin. 2007. Assessing Ogurcak. 2010. Tree mortality following key vulnerabilities and the risk from climate change. p. 779–810. In: Climate slash pine (Pinus elliottii var. densa) forests Change 2007: Impacts, Adaptation and inprescribed the Florida fire Keys, and aUSA. storm International surge event in Vulnerability. Contribution of Working Journal of Forestry Research 2010: Article Group II to the Fourth Assessment Report ID 204795. doi:10.1155/2010/204795. of the Intergovernmental Panel on Climate Change. M.L. Parry, O.F. Canziani, J.P. Sala, O.E., F.S. Chapin III, J.J. Armesto, E. Palutikof, P.J. van der Linden, and C.E. Hanson (eds.) Cambridge University Press, Sanwald, L.F. Huenneke, R.B. Jackson, Cambridge, UK. A.Berlow, Kinzig, J. Bloomfield,R. Leemans, R.D.M. Dirzo, Lodge, E. Huber- H.A. Mooney, M. Oesterheld, N.L. Poff, M.T. Sykes, Schneider, S.H., and T.L. Root (eds.) 2002. B.H. Walker, M. Walker, and D.H. Wall. 2000. Wildlife Responses to Climate Change: Global biodiversity scenarios for the year North American Case Studies. Island Press, 2100. Science 287: 1770–1774. Washington, D.C.
Salafsky, N., R. Margolius, and K. Redford. Schumaker, N.H., T. Ernst, D. White, J. Baker, 2001. Adaptive Management: A Tool for and P. Haggerty. 2004. Projecting wildlife Conservation Practitioners. Biodiversity responses to alternative future landscapes Support Program, Washington, D.C. in Oregon’s Willamette Basin. Ecological Applications 14: 381–401. Sanford, E. 2002. Community responses to climate change: Links between temperature Schwartz, M.W., L.R. Iverson, A.M. Prasad, and keystone predation in a rocky intertidal S.N. Matthews, and R.J. O’Connor. 2006. system. p. 165–200. In: Wildlife Responses Predicting extinctions as a result of climate to Climate Change: North American Case change. Ecology 87: 1611–1615. Studies. S.H. Schneider and T.L. Root (eds.) Island Press, Washington, D.C. Scott, J.M., M. Murray, R.G. Wright, B. Csuti, P. Morgan, and R.L. Pressey. 2001 Saunders, S., C. Montgomery, T. Easley, and Representation of natural vegetation in T. Spencer. 2008. Hotter and Drier: The protected areas: Capturing the geographic West’s Changed Climate. Rocky Mountain range. Biodiversity and Conservation 10: Climate Organization and the Natural 1297–1301. Resources Defense Council. Natural Resources Defense Council, New York.
162 Scanning the Conservation Horizon Literature Cited Seager, R., M. Ting, I. Held, Y. Kushnir, J. Morlot, C.H.D. Magadza, H.-M. Füssel, A.B. Lu, G. Vecchi, H.-P. Huang, N. Harnik, A. Pittock, A. Rahman, A. Suarez, and J.-P. Leetmaa, N.-C. Lau, C. Li, J. Velez, and van Ypersele. 2009. Assessing dangerous N. Naik. 2007. Model projections of an climate change through an update of imminent transition to a more arid climate the Intergovernmental Panel on Climate in Southwestern North America. Science Change (IPCC) “Reasons for Concern.” 316: 1181–1184. Proceedings of the National Academy of Sciences of the United States of America 106: Shaffer, M., and B.A. Stein. 2000. 4133–4137. Safeguarding our precious heritage. p. 301–321. In: Precious Heritage: Status of Spracklen, D.V., L.J. Mickley, J.A. Logan, Biodiversity in the United States. B.A. Stein, R.C. Hudman, R. Yevich, M.D. Flannigan, L.S. Kutner, and J.S. Adams (eds.) Oxford and A.L. Westerling. 2009. Impacts of University Press, New York. climate change from 2000 to 2050 on
Shaver, G.R., J. Canadell, F.S. Chapin III, J. concentrations in the western United Gurevitch, J. Harte, G. Henry, P. Ineson, S. States.wildfire Journal activity of and Geophysical carbonaceous Research aerosol 114: Jonasson, J. Melillo, L. Pitelka, and L. Rustad. D20301. doi:10.1029/2008JD010966. 2000. Global warming and terrestrial ecosystems: A conceptual framework for Stein, B.A., L.S. Kutner, and J.S. Adams analysis. BioScience 50: 871–882. (eds.) 2000. Precious Heritage: The Status of Biodiversity in the United States. Oxford Sitch, S., B. Smith, I.C. Prentice, A. Arneth, University Press, New York. A. Bondeau, W. Cramer, J.O. Kaplan, S. Levis, W. Lucht, M.T. Sykes, K. Thonicke, and S. Steltzer, H., C. Landry, T.H. Painter, J. Venevsky. 2003. Evaluation of ecosystem Anderson, and E. Ayres. 2009. Biological dynamics, plant geography and terrestrial consequences of earlier snowmelt from carbon cycling in the LPJ dynamic desert dust deposition in alpine landscapes. vegetation model. Global Change Biology Proceedings of the National Academy of 9: 161–185. Sciences of the United States of America 106: 11629–11634. Skelly, D.K., L.N. Joseph, H.P. Possingham, L.K. Freidenburg, T.J. Farrugia, M.T. Stockwell, D.R., and D. Peters. 1999. The Kinnison, and A.P. Hendry. 2007. GARP modeling system: Problems and Evolutionary responses to climate change. solutions to automated spatial prediction. Conservation Biology 21: 1353–1355. International Journal of Geographical Information Science 32: 143–158. Smith, A.T., and M.L. Weston. 1990. Ochotona princeps. Mammalian Species Stoner, A.K., K. Hayhoe, and D.J. Wuebbles. 352: 1–8. 2009. Assessing general circulation model simulations of atmospheric teleconnection Smith, J.B., S.H. Schneider, M. Oppenheimer, patterns. Journal of Climate 22:4348–4372. G.W. Yohe, W. Hare, M.D. Mastrandrea, A. Patwardhan, I. Burton, J. Corfee-
Literature Cited 163 Stott, P.A., R.T. Sutton, and D.M. Smith. Tebaldi, C., and R. Knutti. 2007. The use of 2008. Detection and attribution of the multi-model ensemble in probabilistic Atlantic salinity changes. Geophysical climate projections. Philosophical Research Letters 35: L21702. Transactions of the Royal Society A 365: doi:10.1029/2008GL035874. 2053–2075.
Swain, P.B., and J.C. Kearsley. 2000. The Nature Conservancy (TNC). 2006. Classification of Natural Communities of Conservation by Design: A Strategic Massachusetts. Draft report. Massachusetts Framework for Mission Success. The Nature Natural Heritage and Endangered Species Conservancy, Arlington, VA. Program, Westborough. Thuiller, W., S. Lavorel, and M.B. Araújo. Tague, C., K. Heyn, and L. Christensen. 2009. 2005. Niche properties and geographical Topographic controls on spatial patterns extent as predictors of species sensitivity of conifer transpiration and net primary to climate change. Global Ecology and productivity under climate warming in Biogeography 14: 347–357. mountain ecosystems. Ecohydrology 2: 541–554. Torti, V.M., and P.O. Dunn. 2005. Variable effects of climate change on six species of North American birds. Oecologia 145: 486–495.
Trenberth, K.E. 2007. Warmer oceans, stronger hurricanes. Scientific American 297: 44–51.
Trenberth, K.E., A. Dai, R.M. Rasmussen, and D. Parsons. 2003. The changing character of precipitation. Bulletin of the American Meteorological Society 84: 1205–1207.
Turner, B.L. II, R.E. Kasperson, P.A. Matson, J.J. McCarthy, R.W. Corell, L. Christensen, N. Eckley, J.X. Kasperson, A. Luers, M.L. Martello, C. Polsky, A. Pulsipher, and A Schiller. 2003. A framework for vulnerability analysis in sustainability science. Proceedings of the National Academy of Sciences of the United States of America 100: 8074–8079.
USFWS U.S. Climate Change Science Program (U.S. CCSP). 2008a. Scientific Assessment of the Effects of Global Change on the
164 Scanning the Conservation Horizon Literature Cited United States. A report of the Committee U.S. Department of Agriculture (USDA) on Environment and Natural Resources, Forest Service, Rocky Mountain Research National Science and Technology Council, Station. 2010. Vulnerability of Individual Washington, D.C. Species to Climate Change: Vertebrate Species of the Middle Rio Grande Bosque, U.S. Climate Change Science Program New Mexico. Final report, Agreement No. (U.S. CCSP). 2008b. Preliminary Review of 201819H705, for the U.S. Fish and Wildlife Adaptation Options for Climate-Sensitive Service, Region 2, Albuquerque, NM. Ecosystems and Resources. A report by the U.S. Climate Change Science Program U.S. Environmental Protection Agency (U.S. and the Subcommittee on Global Change EPA). 2009. A Framework for Categorizing Research. S.H. Julius and J.M. West (eds.) the Relative Vulnerability of Threatened U.S. Environmental Protection Agency, and Endangered Species to Climate Change. Washington, D.C. EPA/600/R-09/011. National Center for Environmental Assessment, U.S. Climate Change Science Program (U.S. Washington, D.C. CCSP). 2009a. Thresholds of Climate Change in Ecosystems. Final Report, Synthesis U.S. Environmental Protection Agency (U.S. and Assessment Product 4.2. A report by EPA). 1997. Guiding Principles for Monte the U.S. Climate Change Science Program Carlo Analysis. EPA/630/R-97/001. Risk and the Subcommittee on Global Change Assessment Forum, U.S. Environmental Research. U.S. Geological Survey, Reston, VA. Protection Agency, Washington, D.C. http://www.epa.gov/NCEA/pdfs/montcarl. U.S. Climate Change Science Program (U.S. pdf (accessed March 5, 2010). CCSP). 2009b. Coastal Sensitivity to Sea-level Rise: A Focus on the Mid-Atlantic Region. U.S. Fish and Wildlife Service (U.S. FWS). Synthesis and Assessment Product 4.1. A 2005. Scientific Review of the Prescribed report by the U.S. Climate Change Science Fire Program at Blackwater National Program and the Subcommittee on Global Wildlife Refuge. Chesapeake Marshlands Change Research. U.S. Environmental NWR Complex and Fishing Bay Wildlife Protection Agency, Washington, D.C. Management Area, February 22, 2005. http://www.fws.gov/blackwater/pdf/ U.S. Climate Change Science Program (U.S. CCSP). 2009c. Best Practice Approaches 10, 2010). for Characterizing, Communicating, and scientificreviewfire.pdf (accessed August Incorporating Scientific Uncertainty in U.S. Fish and Wildlife Service (U.S. FWS). Climate Decision Making. Synthesis and 2008a. Endangered and threatened wildlife Assessment Product 5.2. A report by the and plants; determination of threatened Climate Change Science Program and status for the polar bear (Ursus maritimus) the Subcommittee on Global Change throughout its range. Federal Register 73: Research. M.G. Morgan et al. (eds.) National 28212–28303. Oceanic and Atmospheric Administration, Washington, D.C.
Literature Cited 165 U.S. Fish and Wildlife Service (U.S. FWS). as traced by organic carbon, biogenic silica 2008b. Endangered and threatened wildlife and stable isotope/trace element studies in and plants; status review for the Rio sediment cores. p. 109–113. In: Proceedings Grande cutthroat trout. Federal Register 73: of the 6th Biennial Long Island Sound 27900–27962. Meeting, Groton, CT; October 26–28, 2002.
U.S. Fish and Wildlife Service (U.S. FWS). VEMAP. 1995. Vegetation/ecosystem 2009a. Strategic Habitat Conservation: modelling and analysis project: Comparing The USFWS framework for landscape biogeography and biogeochemistry models conservation. http://www.fws.gov/ in an continental-scale study of terrestrial science/shc/index.html (accessed April 12, ecosystem responses to climate change and 2010). CO2 doubling. Global Biogeochemical Cycles 9: 407–437. U.S. Fish and Wildlife Service (U.S. FWS). 2009b. Endangered and threatened wildlife Vermeer, M. and S. Rahmstorf. 2009. Global and plants; revised designation of critical sea level linked to global temperature. Proceedings of the National Academy of (Euphydrydas editha quino). Federal Sciences 106: 21527-21532. Registerhabitat for 74: the 28776–28862. Quino checkerspot butterfly Visser, M.E., and C. Both. 2005. Shifts in U.S. Fish and Wildlife Service (U.S. FWS). phenology due to global climate change: 2010. Endangered and threatened wildlife The need for a yardstick. Proceedings of the Royal Society B 272: 2561–2569. to list the American pika as threatened orand endangered. plants; 12-month Federal finding Register on 75: a petition Vogel, C., S.C. Moser, R.E. Kasperson, and 6438–6471. G.D. Dabelko. 2007. Linking vulnerability, adaptation, and resilience science U.S. Global Change Research Program to practice: Pathways, players, and (USGCRP). 2009. Global Climate Change partnerships. Global Environmental Change Impacts in the United States. A report of 17: 349–364. the USGCRP. USGCRP, Washington, D.C. http://www.globalchange.gov/usimpacts/ Vos, C.C., H. Bavero, and C.J. Grashof- (accessed August 7, 2010). Bokdam. 2002. Corridors and species dispersal. p. 84–104. In: Applying Landscape Van Mantgem, N.L. Stephenson, J.C. Byrne, Ecology in Biological Conservation. K.J. L.D. Daniels, J.F. Franklin, P.Z. Fulé, M.E. Gutzwiller (ed.) Springer-Verlag, New York. Harmon, A.J. Larson, J.M. Smith, A.H. Taylor, and T.T. Veblen. 2009. Widespread increase Vrac, M., K. Hayhoe, and M. Stein. 2006. in tree mortality rates in the western United States. Science 323: 521–524. of seasonal circulation patterns over NorthIdentification America. and International inter-model Journal comparison of Varekamp, J.C., E. Thomas, F. Lugolobi, Climatology 27: 603–620. and M.R. Buchholtz . 2004. The paleo- environmental history of Long Island Sound
166 Scanning the Conservation Horizon Literature Cited Walker, B., C.S. Holling, S.R. Carpenter, and A. Kinzig. 2004. Resilience, adaptability and transformability in social-ecological systems. Ecology and Society 9(2): Article 5. http://www.ecologyandsociety.org/vol9/ iss2/art5/ (accessed August 7, 2010).
Walker, W.E., H. Harremoës, J. Rotmans, J.P. van der Sluijs, M.B.A. van Asselt, P. Janssen, and M.P. Krayer von Krauss. 2003. uncertainty management in model-based decisionDefining support.uncertainty: Integrated A conceptual Assessment basis for 4: 5–17.
Waltari, E., R.H. Hijmans, A.T. Peterson, Á.S. Nyári, S.L. Perkins, and R.P. Guralnick. 2007. Locating Pleistocene refugia: Comparing phylogeographic and ecological niche model predictions. PLoS One 2: e563. doi:10.1371/journal.pone.0000563.
Walther, G.-R., S. Beissner, and C.A. Burga. 2005. Trends in the upward shift of alpine plants. Journal of Vegetation Science 16: J&K Hollingsworth/USFWS 541–548. Webster, P.J., G.J. Holland, J.A. Curry, and H.-R. Chang. 2005. Changes in tropical Walther, G.-R., R.E. Post, P. Convey, A. cyclone number, duration, and intensity Menzel, C. Parmesan, T.J.C. Beebee, J.-M. in a warming environment. Science 309: Fromentin, O. Hoegh-Guldberg, and F. 1844–1846. Bairlein. 2002. Ecological responses to recent climate change. Nature 416: Werner, R.A., E.H. Holsten, S.M. Matsuoka, 389–395. and R.E. Burnside. 2006. Spruce beetles and forest ecosystems in south-central Alaska: Wang, M., J.E. Overland, V. Kattsov, J.E. A review of 30 years of research. Forest Walsh, X. Zhang, and T. Pavlova. 2007. Ecology and Management 227: 195–206. Intrinsic versus forced variation in coupled climate model simulations over the Arctic West, J.M., S.H. Julius, P. Kareiva, C. Enquist, during the twentieth century. Journal of J.J. Lawler, B. Petersen, A.E. Johnson, and Climate 20:1093–1107. M.R. Shaw. 2009. U.S. natural resources and climate change: Concepts and approaches Ward, J.R., K. Kim, and C.D. Harvell. for management adaptation. Environmental 2007. Temperature affects coral disease Management 44: 1001–1021. resistance and pathogen growth. Marine Ecology Progress Series 329: 115–121.
Literature Cited 167 Westerling, A.L., H.G. Hidalgo, D.R. Cayan, Worm, B., E.B. Barbier, N. Beaumont, J.E. and T.W. Swetnam. 2006. Warming and Duffy, C. Folke, B.S. Halpern, J.B.C. Jackson, earlier spring increases western U.S. forest H.K. Lotze, F. Micheli, S.R. Palumbi, E. Sala, wildlife activity. Science 313: 940–943. K.A. Selkoe, J.J. Stachowicz, and R. Watson. 2006. Impacts of biodiversity loss on ocean Wiens, J.A., and D. Bachelet. 2009. Matching ecosystem services. Science 314: 787–790. multiple scales of conservation with the multiple scales of climate change. Wormworth, J., and K. Mallon. 2007. Bird Conservation Biology 24: 51–62. Species and Climate Change: The Global Status Report Version 1.1. A report to World Wigley, T. 2008. MAGICC/SCENGEN Wide Fund for Nature. World Wide Fund for 5.3. User manual (version 2). http:// Nature-Australia, Sydney. www.cgd.ucar.edu/cas/wigley/magicc/ UserMan5.3.v2.pdf (accessed August 7, Yin, J., M.E. Schlesinger, and R.J. Stouffer. 2010). 2009. Model projections of rapid sea-level rise on the northeast coast of the United Williams, B.K., R.C. Szaro, and C.D. Shapiro. States. Nature Geoscience 2: 262 – 266. 2007. Adaptive Management: The U.S. Department of the Interior Technical Guide. Young, B.E., K.R. Hall, E. Byers, K. Gravuer, Adaptive Management Working Group, U.S. A. Ridder, and K. Szabo. In press. A natural Department of the Interior, Washington, D.C. history approach to rapid assessment of plant and animal vulnerability to climate Williams, S.E., L.P. Shoo, J.L. Isaac, A.A. change. In: Conserving Wildlife Populations Hoffmann, and G. Langham. 2008. Towards in a Changing Climate. J. Brodie, E. Post, an integrated framework for assessing the and D. Doak (eds.) University of Chicago vulnerability of species to climate change. Press, Chicago. PloS Biology 6: 2621–2626. Young, J.A., and R.R. Blank. 1995. Cheatgrass Wilson, E.O. 1992. The Diversity of Life. Norton, New York. p. 6–8. In: Proceedings of the California Exoticand wildfires Pest Plant in the Council, Intermountain Symposium West. Wilson, M., and B. Watts. 2009. Projected ’95. J.E. Lovich, J. Randall, and M.D. Kelly Impacts of Sea-level Rise on the Population (eds.) California Exotic Pest Plant Council, Size of Salt-marsh Breeding Birds within Sacramento. the Chesapeake Bay Region. Center for Conservation Biology, College of William Zhao, H.X., S.H. Wu, and L.G. Jiang. 2007. and Mary and Virginia Commonwealth Research advances in vulnerability University, Williamsburg. assessment of natural ecosystem response to climate change [in Chinese]. Ying Yong Wood, A.W., L.R. Leung, V. Sridhar, and D.B. Sheng Tai Xue Bao 18: 445–450. Lettenmaier. 2004. Hydrologic implications of dynamical and statistical approaches to downscaling climate model surface
Climatic Change 62: 189–216. temperature and precipitation fields.
168 Scanning the Conservation Horizon
901 E Street, Suite 400 Washington, DC 20004 202.797.6000 www.nwf.org
55%
Cert no. BV-COC-080403 Case Studies