Committee On Earth-Atmosphere Interactions: Understanding and Responding To Multiple Environmental Stresses, National Research Council

Understanding Multiple Environmental Stresses:
Report of a Workshop
Committee on Earth-Atmosphere Interactions:
Understanding and Responding to Multiple
Environmental Stresses, National Research Council
ISBN: 0-309-66576-0, 154 pages, 6 x 9, (2007)
This free PDF was downloaded from:
http://www.nap.edu/catalog/11748.html
Visit the National Academies Press online, the authoritative source for all books from the
National Academy of Sciences, the National Academy of Engineering, the Institute of
Medicine, and the National Research Council:
• Download hundreds of free books in PDF
• Read thousands of books online, free
• Sign up to be notified when new books are published
• Purchase printed books
• Purchase PDFs
• Explore with our innovative research tools
Thank you for downloading this free PDF. If you have comments, questions or just want
more information about the books published by the National Academies Press, you may
contact our customer service department toll-free at 888-624-8373, visit us online, or
send an email to [email protected].
This free book plus thousands more books are available at http://www.nap.edu.
Copyright © National Academy of Sciences. Permission is granted for this material to be

shared for noncommercial, educational purposes, provided that this notice appears on the
reproduced materials, the Web address of the online, full authoritative version is retained,
and copies are not altered. To disseminate otherwise or to republish requires written
permission from the National Academies Press.
Understanding Multiple Environmental Stresses: Report of a Workshop
Committee on Earth-Atmosphere Interactions:

Understanding and Responding to Multiple Environmental Stresses
Board on Atmospheric Sciences and Climate
Division on Earth and Life Studies
The NATIONAL ACADEMIES PRESS

Washington, D.C.
www.nap.edu
Copyright © National Academy of Sciences. All rights reserved.

THE NATIONAL ACADEMIES PRESS 500 Fifth Street, N.W. Washington, DC 20001
NOTICE: The project that is the subject of this report was approved by the Governing
Board of the National Research Council, whose members are drawn from the councils of
the National Academy of Sciences, the National Academy of Engineering, and the Institute
of Medicine. The members of the committee responsible for the report were chosen for
their special competences and with regard for appropriate balance.
This study was supported by the National Science Foundation and the National Aeronau-
tics and Space Administration under Grant No. ATM-0135923, the National Oceanic and
Atmospheric Administration under Contract No. 52-DGNA-1-90024, and the Environ-
mental Protection Agency under Purchase Order No. 2W-0373-NANX. Any opinions,
findings, and conclusions or recommendations expressed in this publication are those of
the author(s) and do not necessarily reflect the views of the agencies that provided support
for this project.
International Standard Book Number-13: 978-0-309-10331-2

International Standard Book Number-10: 0-309-10331-2
Cover: Satellite photo of smoke emanating from large wildfires along the Alaska-Canada
border and spreading across Alaska and into Siberia and the Arctic. Courtesy of NASA.
Additional copies of this report are available from the National Academies Press, 500 Fifth
Street, N.W., Lockbox 285, Washington, DC 20055; (800) 624-6242 or (202)334-3313 (in
the Washington metropolitan area); Internet, http://www.nap.edu.
Copyright 2007 by the National Academy of Sciences. All rights reserved.
Printed in the United States of America

The National Academy of Sciences is a private, nonprofit, self-perpetuating society of

distinguished scholars engaged in scientific and engineering research, dedicated to the
furtherance of science and technology and to their use for the general welfare. Upon
the authority of the charter granted to it by the Congress in 1863, the Academy has a man-
date that requires it to advise the federal government on scientific and technical matters.
Dr. Ralph J. Cicerone is president of the National Academy of Sciences.
The National Academy of Engineering was established in 1964, under the charter of
the National Academy of Sciences, as a parallel organization of outstanding engineers.
It is autonomous in its administration and in the selection of its members, sharing with
the National Academy of Sciences the responsibility for advising the federal government.
The National Academy of Engineering also sponsors engineering programs aimed at
meeting national needs, encourages education and research, and recognizes the superior
achievements of engineers. Dr. Wm. A. Wulf is president of the National Academy of
Engineering.
The Institute of Medicine was established in 1970 by the National Academy of Sciences
to secure the services of eminent members of appropriate professions in the examination of
policy matters pertaining to the health of the public. The Institute acts under the responsibil-
ity given to the National Academy of Sciences by its congressional charter to be an adviser
to the federal government and, upon its own initiative, to identify issues of medical care,
research, and education. Dr. Harvey V. Fineberg is president of the Institute of Medicine.
The National Research Council was organized by the National Academy of Sciences in
1916 to associate the broad community of science and technology with the Academy’s
purposes of furthering knowledge and advising the federal government. Functioning in
accordance with general policies determined by the Academy, the Council has become
the principal operating agency of both the National Academy of Sciences and the Na-
tional Academy of Engineering in providing services to the government, the public, and
the scientific and engineering communities. The Council is administered jointly by both
Academies and the Institute of Medicine. Dr. Ralph J. Cicerone and Dr. Wm. A. Wulf are
chair and vice chair, respectively, of the National Research Council.
www.national-academies.org


COMMITTEE ON EARTH-ATMOSPHERE INTERACTIONS:

UNDERSTANDING AND RESPONDING TO
MULTIPLE ENVIRONMENTAL STRESSES
ROSINA M. BIERBAUM (Co-chair), University of Michigan, Ann Arbor

MARY ANNE CARROLL (Co-chair), University of Michigan, Ann Arbor
CHRISTOPHER B. FIELD, Carnegie Institution of Washington, Stanford,
California
EDWARD L. MILES, University of Washington, Seattle
DONALD A. WILHITE, University of Nebraska, Lincoln
NRC Staff
CHRIS ELFRING, Study Director
MATTHEW M. RUSSELL, Associate Program Officer (until 9/2006)
DIANE GUSTAFSON, Administrative Coordinator

BOARD ON ATMOSPHERIC SCIENCES AND CLIMATE
ROBERT J. SERAFIN (Chair), National Center for Atmospheric Research,

Boulder, Colorado
M. JOAN ALEXANDER, NorthWest Research Associates/CORA, Boulder,
Colorado
FREDERICK R. ANDERSON, McKenna Long & Aldridge LLP,
Washington, D.C.
MICHAEL L. BENDER, Princeton University, New Jersey
ROSINA M. BIERBAUM, University of Michigan, Ann Arbor
MARY ANNE CARROLL, University of Michigan, Ann Arbor
CAROL ANNE CLAYSON, Florida State University, Tallahassee
WALTER F. DABBERDT, Vaisala Inc., Boulder, Colorado
KERRY A. EMANUEL, Massachusetts Institute of Technology, Cambridge
DENNIS L. HARTMANN, University of Washington, Seattle
PETER R. LEAVITT, Weather Information Inc., Newton, Massachusetts
JENNIFER A. LOGAN, Harvard University, Cambridge, Massachusetts
VERNON R. MORRIS, Howard University, Washington, D.C.
F. SHERWOOD ROWLAND, University of California, Irvine
THOMAS H. VONDER HAAR, Colorado State University/CIRA, Fort Collins
ROGER M. WAKIMOTO, National Center for Atmospheric Research,
Boulder, Colorado
Ex Officio Members
ANTONIO J. BUSALACCHI, JR., University of Maryland, College Park
NRC Staff
CHRIS ELFRING, Director
AMANDA STAUDT, Senior Program Officer
CURTIS MARSHALL, Program Officer
IAN KRAUCUNAS, Associate Program Officer
CLAUDIA MENGELT, Associate Program Officer
ELIZABETH A. GALINIS, Research Associate
LEAH PROBST, Research Associate
ROB GREENWAY, Senior Program Assistant
KATHERINE WELLER, Program Assistant
DIANE GUSTAFSON, Administrative Coordinator
ANDREAS SOHRE, Financial Associate
vi

Preface
Periodically the Board on Atmospheric Sciences and Climate (BASC) works

with its federal agency partners to select a topic for a special workshop, some-
times called our “summer study.” The purpose of the workshop is to provide an
opportunity for scientists, industry, and agency staff to explore current issues in
an interactive format. Sometimes these workshops address practical problems,
such as communicating uncertainties in weather forecasts (NRC, 2003), and other
times specialized technical issues, such as improving the physical parameteriza-
tions in coupled atmosphere-ocean-land models (NRC, 2005a). Often, such as in
this report, an issue is selected that might otherwise go unstudied due to the scale,
scope, or tractability of the problem.
The 2005 BASC workshop focused on multiple environmental stresses in the
earth-atmosphere system (see Appendix A for Statement of Task). Historically,
environmental problems have been studied one at a time and sector by sector (e.g.,
the impacts of air pollution on human health or the impacts of invasive species on
fisheries). Although this approach has enabled researchers to make good progress
in many areas in characterizing cause-effect environmental relationships that are
linear in nature and limited in scale, it does not consider the composite effects of
simultaneous environmental changes. Unless we consider robust options that solve
multiple problems and prevent new ones, we may be ineffective and inefficient in
our environmental efforts. Some of these issues are addressed in other National
Research Council reports (e.g., NRC, 1999, 2002).
vii

viii PREFACE
This workshop was intended as a step in identifying the types of near-term

and long-term research needed to understand multiple environmental stresses
and explore integrated strategies to address them. It was planned and facilitated
by a five-person steering committee and was held September 29-30, 2005, at
the Arnold and Mabel Beckman Center of the National Academies in Irvine,
California. More than 25 experts from a variety of disciplines and perspectives,
including both the natural and social sciences, attended, as well as managers and
stakeholders in various sectors and regions (see Appendix B for the workshop
agenda and Appendix C for the participant list). The participants were charged
to explore current understanding of multiple environmental stresses in the earth-
atmosphere system and to discuss the types of research needed to improve inte-
grated understanding and response strategies for these kinds of complex, nonlinear
problems. To focus the discussions, two case studies were selected and partici-
pants were assigned to come prepared with short talks on aspects of these cases;
other participants were assigned to lead discussion sessions to explore the issues
and generate ideas about research needs. This report is the steering committee’s
summary of these presentations and the associated discussions; abstracts of the
participants’ talks are included as Appendix D. The workshop was funded using
support provided from the National Science Foundation, National Oceanic and
Atmospheric Administration, National Aeronautics and Space Administration,
and Environmental Protection Agency.
On behalf of the Board on Atmospheric Sciences and Climate and the
National Academies, I would like to express great thanks to the steering committee
for its leadership and to all the workshop speakers and participants for their time
and thoughtful comments. Although a workshop by definition can only explore
issues and not provide truly detailed or deliberative recommendations, this work-
shop report should prove useful to researchers and agency program managers
looking for opportunities to address these complex issues.
Chris Elfring, Director

Board on Atmospheric Sciences and Climate
Following standard National Academies procedures for workshops, this report captures the discus-
sions and presentations that occurred during the two-day event; it does not contain recommendations.

Acknowledgments
This workshop report was written by the workshop steering committee based
on the presentations and discussions at the workshop, and we appreciate the input
from all the participants. In addition, this report has been reviewed in draft form
by individuals chosen for their diverse perspectives and technical expertise, in
accordance with procedures approved by the National Research Council’s Report
Review Committee. The purpose of this independent review is to provide candid
and critical comments that will assist the institution in making its published report
as sound as possible and to ensure that the report meets institutional standards
for objectivity, evidence, and responsiveness to the study charge. The review
comments and draft manuscript remain confidential to protect the integrity of the
deliberative process. We wish to thank the following individuals for their review
of this report:
M. Joan Alexander, NorthWest Research Associates/CORA, Boulder, Colorado

Eric J. Barron, University of Texas at Austin
Julio Betancourt, University of Arizona, Tucson
Scott C. Doney, Woods Hole Oceanographic Institution, Massachusetts
Kelly T. Redmond, Desert Research Institute, Reno, Nevada
William H. Schlesinger, Duke University, Durham, North Carolina
Although the reviewers listed above have provided many constructive com-
ments and suggestions, they were not asked to endorse the conclusions or rec-
ommendations nor did they see the final draft of the report before its release.
The review of this report was overseen by Elbert W. Friday, Jr., University of
ix

Acknowledgments
Oklahoma. Appointed by the National Research Council, he was responsible for

making certain that an independent examination of this report was carried out in
accordance with institutional procedures and that all review comments were care-
fully considered. Responsibility for the final content of this report rests entirely
with the authoring committee and the institution.

Contents
SUMMARY 1
1 INTRODUCTION 5
Defining the Concept of “Multiple Stresses,” 5
The Nature of the Problem, 7
What Assessments Conclude About Research Needs, 9
Introduction to the Case Studies, 9
Nonlinearities, Thresholds, and the Vulnerability-Resilience
Continuum, 10
2 DROUGHT 15
Context and Impacts, 19
Understanding Vulnerability and Response Strategies, 21
Drought Policy and Preparedness, 22
Research Needs, 24
3 ATMOSPHERE-ECOSYSTEM INTERACTIONS 27
Context and Impacts, 28
Impacts on Humans, Ecosystems, and Economies, 31
Policy Options: Adaptation and Mitigation, 32
Research Needs, 33
xi

xii Contents
4 LESSONS LEARNED FROM THE CASE STUDIES 36

Comprehensive Reginal Frameworks, 37
A Web of Integrated Sensors, 38
Regional Information Systems, 38
Framework for Process Studies, 39
Improving Our Predictive Capability, 39
Research Needs Related to Nonlinearities and Thresholds, 41
Research Needs Related to Increasing Resilience, 42
Research Needs Related to Regional Studies, 43
Near-Term Opportunities, 44
REFERENCES 49
APPENDIXES
A Statement of Task 53
B Workshop Agenda 54
C Workshop Participants 57
D Extended Speaker Abstracts 59
E Committee Biosketches 139

Summary
The research of the last decade has demonstrated that ecosystems and human
systems are influenced by multiple factors, including climate, land use, and the
by-products of resource use. Understanding the net impact of a suite of simultane-
ously occurring environmental modifications is essential for developing effective
response strategies. This suite of simultaneous influences, or multiple environ-
mental stresses, produces more than simply additive impacts. The term means
that there is a confluence and interaction of stresses that both accumulate and,
because of feedbacks, increase or become more complex. Multiple environmental
stresses in composite lead to qualitatively and quantitatively different outcomes
from single influences, and thus research that seeks to better understand these
multiple stresses requires thinking differently from that used in traditional, sec-
toral, or single-problem approaches.
There are no generally agreed upon methodologies for studying com-
plex systems of interconnected environmental influences that can have differ-
ent impacts in varied and sometimes subtle directions. Understanding multiple
stresses almost always requires consideration of multiple variables, nonlinear
processes, and a variety of spatial and temporal scales. We typically have only
a rudimentary understanding of the dynamics of interactions between different
environmental variables in complex systems, making it extremely difficult to
predict the combined effects of multiple interacting stresses. In addition to gaps
in the scientific understanding of multiple environmental stresses, there is a lack
of understanding of how to move from understanding to management and policy
decisions, and in particular how to devise options that make sense in the face of
significant uncertainties.

UNDERSTANDING AND RESPONDING TO MULTIPLE ENVIRONMENTAL STRESSES
The concept of multiple environmental stresses taken alone can seem vague,
and thus the steering committee organized this workshop around two examples to
provide different perspectives on multiple stress scenarios. The first case selected
was drought, a complex environmental condition that both is driven by multiple
environmental stresses and leads to multiple stresses across a wide range of time
and spatial scales. Drought is a normal climate variation that can vary in magni-
tude and intensity, and it provides a clear illustration of the feedbacks involved
both in the occurrence of the natural event and in the human activities that may
alter societal vulnerability (e.g., population growth, water management policies,
and changes in land cover). The second case selected focused on a wide range
of atmosphere-ecosystem interactions that taken together reflect characteristics of
multiple, simultaneous environmental stresses.
These two cases were selected because they offer very different problems,
scales, and lessons. Because of this, the presentations and discussions—and
the respective chapters in this report—are not perfectly parallel. Despite the
differences in approach, the workshop participants did identify some impor-
tant commonalities. As discussed in Chapter 4, the overarching lesson of the
workshop is that society will require new and improved strategies for coping
with multiple stresses and their impacts on natural and socioeconomic systems.
Improved communication among stakeholders, increased observations (especially
at regional scales), improved model and information systems, and increased infra-
structure to provide better environmental monitoring, vulnerability assessment,
and response analysis are all important parts of moving toward better understand-
ing of and response to multiple-stresses situations.
Workshop participants identified the development of comprehensive regional
frameworks for conducting environmental studies as a key part of understanding
multiple environmental stresses:
1. An integrated regional web of sensors that links existing observations into

a coherent framework
2. An integrated and comprehensive regional information system accessible
to a wide variety of researchers, operational systems, and stakeholders
3. Directed process studies designed to examine specific phenomena through
field study
4. Complex coupled system models at the spatial and temporal scales appro-
priate to study specific and integrated biologic, hydrologic, and socioeconomic
systems
5. A strong connection to significant regional issues and stakeholders.
Workshop participants identified the Regional Integrated Sciences and

Assessments program (RISA) as a possible model for such regional frameworks.
Overall, the degree to which progress is made on complicated environmental
problems is proportional to the degree we are able to implement these five

SUMMARY
organizing principles, and progress will be limited if we cannot. This will take
a major reorganization in how we approach multifactor environmental problems
and thus will be difficult.
As an outcome of the workshop, seven near-term opportunities for research
and infrastructure that could help advance our understanding of multiple stresses
and make this understanding useful to decision makers were proposed. These are
discussed in more detail in Chapter 4.
1. Building a new U.S. ground-based ecological measurement network to

understand multiple stresses at multiple scales
2. Enhancing global information systems and satellite observations
3. Synthesizing existing data to identify status and trends in ecosystems
4. Analyzing extreme events of record to identify nonlinearities and thresholds
for future change
5. Identifying technologies, management, or institutional mechanisms to
enhance resilience
6. Developing regional foci to understand vulnerability and coping capacity
in particular places
7. Increasing stakeholder involvement to help identify problems and solu-
tions and provide ideas on the kinds of information needed (and in what format)
to be useful for decision making.


Introduction1
DEFINING THE CONCEPT OF “MULTIPLE STRESSES”

Natural, managed, and socioeconomic systems often experience multiple
environmental stresses that can come together to cause impacts that are greater
than simply additive. Multiple environmental changes can also interact with each
other to cause additional unexpected stresses. The term “multiple stresses” is
simply a shorthand way of referring to scenarios where multiple environmental
influences are at work with various multidimensional interactions among them.
Understanding the net impact of a suite of simultaneously occurring environ-
mental modifications (e.g., elevated carbon dioxide; increased oxidants, reactive
nitrogen, and acid deposition; decreased stratospheric ozone; increased ultraviolet
radiation; higher mean temperature; changes in timing and availability of water;
loss of biodiversity, increases in invasive species, rising sea level; coastal develop-
ment and habitat fragmentation) is essential for developing predictive capabilities
and response strategies.
There are no generally agreed upon methodologies for studying complex sys-
tems of interconnected environmental influences that can have different impacts
in varied and sometimes subtle directions. Understanding multiple stresses almost
always requires consideration of multiple variables, nonlinear processes, and
a variety of spatial and temporal scales. However, we typically have only a
1Much of the material in this introductory chapter is taken from the workshop’s keynote address by
Dr. Eric Barron, University of Texas at Austin, who was asked to define terms and provide context
for the workshop.

rudimentary understanding of the dynamics of interactions between different

environmental variables in complex systems, making it extremely difficult to
predict the combined effects of multiple interacting stresses. Other scientific
challenges include understanding the multiple time and space scales on which
these interactions take place, developing indicators for “threshold” responses that
may lead to sudden and dramatic changes in societal or environmental structure
and function, characterizing and quantifying risks and vulnerabilities, and under-
standing the economic impact of multiple interacting changes. Another critical
challenge is proceeding with decision making given the many uncertainties and
limited prediction capabilities.
Given the highly interdisciplinary nature of multiple environmental stresses,
conducting research on the topic involves a variety of infrastructural and institu-
tional challenges. First, a broad range of observational and experimental/process
studies is needed for understanding integrated climate and ecosystem processes.
Second, coupled biophysical and biogeochemical models are needed to address
the dynamics of exchange of water, energy, carbon, and nitrogen on multiple
timescales, with biogeographic models that simulate the effects of climate and
other factors on vegetation distribution. New data and information systems may
be necessary to enable scientists to integrate knowledge across these disciplines.
Further, it appears likely that focusing on natural, managed, and socioeconomic
systems on a regional scale may provide a tractable approach to bringing together
the diverse researchers and knowledge needed to improve understanding of mul-
tiple environmental stresses.
In addition to gaps in the scientific understanding of multiple environmental
stresses, there is a lack of understanding of how to devise wise management
and policy approaches that address suites of problems and, in particular, options
that make sense in the face of uncertainty. These options can be technological,
managerial, or institutional and will require much more integrative research in
many disciplines. It will also require serious consideration of how we translate
information as well as a sense of what we do not know or cannot know, so that
it is useful to those who must make decisions. Adaptive management using the
best information available, while retaining flexibility to make changes, requires
managers to think differently, but it does increase resiliency to risk.
Effective communication between researchers and stakeholders is critical.
As research develops more sophisticated understanding of environmental systems
and how multiple stresses interact and compound, it becomes even more difficult
to translate this understanding to those who must use the information. Although
there is no easy answer to this dilemma, the solution appears to require extended
interactions, careful attention to including all perspectives, and frankness about
the level of certainty and uncertainty of the information.

INTRODUCTION
THE NATURE OF THE PROBLEM

The nature of the environmental issues facing any nation demands a capabil-
ity that allows us to enhance economic vitality, maintain environmental quality,
limit threats to life and property, and strengthen fundamental understanding of
Earth. In each case it is the ability to anticipate the future (e.g., forecasting an
impending storm, predicting water quality change in response to a new source
of pollutant) that makes information about the earth system truly useful. Reli-
able information about the future (i.e., forecasts or predictions) is essential when
addressing environmental issues. Thus, society requires greater access to and
greater confidence in both information and forecasts or projections in order to
weigh the advantages and risks of alternative courses of action. Such information
is a key commodity in enhancing economic vitality and societal well-being. What
stands in our way of providing this information?
Many of the driving forces that alter environmental quality are widely rec-
ognized and involve primarily weather and climate, patterns of land use and
land cover, and resource use with its associated waste products. A key feature
of most regions is that more than one driving force is changing simultaneously.
Consequently, most locations are characterized by multiple stresses. The effect of
a combination of environmental stresses is seldom simply additive. Rather, they
often produce amplified or damped responses, unexpected responses, or thresh-
old responses in environmental systems. Multiple, cumulative, and interactive
stresses are clearly the most difficult to understand and hence the most difficult
to manage.
In contrast to how the real world works, most research and policy focuses
on discrete parts of these complex problems. Basically, earth and environmental
sciences tend to focus on cause and effect, where we seek to understand how a
specific element of the system may respond to a specific change or perturbation
(e.g., the impact of acid rain on lake fisheries). The lack of methods to assess the
response of the system to multiple stresses limits our ability to assess the impacts
of specific human perturbations, to assess advantages and risks, and to enhance
economic and societal well-being in the context of global, national, and regional
stewardship.
The problem is not limited simply to moving from analysis of discrete parts
of complex problems to a more comprehensive analysis. First, economic vitality
and societal well-being are increasingly dependent on combining global, regional,
and local perspectives. A “place-based” imperative (i.e., site specific) for envi-
ronmental research stems from the importance of human activities on local and
regional scales, the importance of multiple stresses on specific environments,
and the nature of the spatial and temporal linkages between physical, biological,
chemical, and human systems. We find the strongest intersection between human
activity, environmental stresses, earth system interactions, and human decision
making in regional analysis coupled to larger spatial scales.

Second, although a decade of research on greenhouse gas emissions, ozone

depletion, and deforestation has answered some critical questions, the last decade
of effort has also revealed a number of new challenges. The most notable is the
challenge of creating integrated global observation capabilities and the compu-
tational and scientific limitations inherent in creating a truly integrated, global,
coupled system modeling capability suitable for assessing impacts and adapta-
tions. These problems are noteworthy in global change science, but they become
intractable at the scales of human decision making. A major part of the problem is
simply a matter of scale and the sheer volume of information required to combine
physical, biological, chemical, and human systems if the framework is global.
For example, whereas a global integrated observing system is challenging but
tractable and plays a fundamental role on the scale of a global circulation model,
it collapses under its own weight at higher spatial resolutions if we demand a
truly comprehensive data system involving the host of observations spanning
biology, hydrology, soils, weather, etc., required to address problems at the scales
of human decision making.
Recognition of the importance of developing a more integrated approach to
environmental research was made abundantly clear in the National Assessment
Synthesis Team’s report, Climate Change Impacts on the United States (NAST,
2000). The first recommendation for future research focused on developing a
more integrated approach to examining impacts and vulnerabilities to multiple
stresses. The report contains many examples where the key limitation to the
assessment of potential impacts on climate was a lack of knowledge of other
stressors. For example, changes in insect-, tick-, and rodent-borne diseases could
be clearly tied to weather and climate, but the number of other environmental
factors that could influence the disease vectors (e.g., the importance of land use
on disease hosts), transmission dynamics, and population vulnerabilities severely
limited our ability to make robust conclusions on how climate change might influ-
ence the distribution and occurrence of many infectious diseases in the future.
Why, with so many pressing problems demanding research attention, should
the United States give more focus to multiple stresses in environmental research?
The driving forces that alter environmental quality and integrity are often well
known, but most regions experience multiple simultaneous environmental
changes, and the combined effects of these changes are much more difficult to
understand and manage than the discrete issues that most research, analysis, and
policy focus on. This lack of appreciation for, and understanding of, multiple
stresses limits our ability to assess the impacts of specific human perturbations,
to assess advantages and risks, and to enhance economic and societal well-being
in the context of global, national, and regional stewardship. However, the solu-
tion to this problem is not simply increasing the scope of our analysis, but rather
developing a more focused and integrated approach to environmental research.

INTRODUCTION
WHAT ASSESSMENTS CONCLUDE ABOUT RESEARCH NEEDS

Many national and international assessments of environmental issues have
been conducted in the last two decades, such as the ozone assessments (WMO/
UNEP, 1985, 1989, 1991, 1994, 1998, 2002), the Intergovernmental Panel on Cli-
mate Change assessments (IPCC, 1990, 1995, 2001), the U.S. National Assess-
ment (NAST, 2000), and the Millennium Ecosystem Assessment (MEA, 2005).
Each of these multiyear efforts involving thousands of scientists, government
officials, and stakeholders has made significant contributions by summarizing the
state of science and characterizing remaining uncertainties. What is lacking, how-
ever, is a strategic research plan that could emanate from these assessments.
After assembling the data that are available, assessment researchers are
keenly aware of what was not there and which gaps are most critical to fill to
enhance understanding. Taking time at the end of assessments to characterize
what is known, what is not known, what is knowable in what time frame, and
what is most important to know to assist timely decision making would gener-
ate much-needed short-term and long-term research strategies. These strategies
would be valuable to governments and researchers in thinking about appropriate
budgets across agencies, across disciplines, across space, and over time to answer
societal questions.
Especially in times of tight budgets, all desired research cannot be conducted
simultaneously. Missing information about scientific processes, technological
promise, institutional mechanisms, ecological or socioeconomic thresholds, and
behavioral change all need to be developed, but some missing pieces of the
puzzle may be more crucial to pursue in order to find robust solutions. And some
questions are more tractable than others.
INTRODUCTION TO THE CASE STUDIES

The concept of multiple stresses, taken alone, can seem vague, and thus the
steering committee decided to focus on two concrete examples that were selected
to provide different perspectives on multiple-stresses scenarios. The first case
study selected for examination was drought, a complex environmental condi-
tion that both is driven by multiple environmental stresses and leads to multiple
stresses across a wide range of time and spatial scales. Drought is a normal
climate variation that can vary in magnitude and intensity, and while it is not the
only climate-induced generator of multiple stresses, it is a significant one and one
that provides clear illustration of the feedbacks involved. The second case study
selected for attention focused on a wide range of atmosphere-ecosystem-human
interactions that, taken together, reflect characteristics of multiple simultaneous
environmental stresses. For each case, workshop participants discussed the cur-
rent state of understanding, what can be learned from prior unexpected findings,
vulnerability and future socioeconomic impacts, and potential response strategies.

10 UNDERSTANDING AND RESPONDING TO MULTIPLE ENVIRONMENTAL STRESSES
Participants then stepped back from the examples in search of common lessons
that might be learned.
Drought, in general, originates from a deficiency of precipitation over
an extended period, resulting in a water shortage for some activity, group, or
environmental sector. Workshop participants discussed the current state of the
knowledge base concerning the causes, frequency, intensity, and predictability of
drought at multiple spatial scales within the continental United States and how
societal changes (e.g., increasing population) affect vulnerability to drought at
local and regional scales. The question of how the United States could facilitate
development of a risk-based drought management approach directed at increasing
resilience and decreasing vulnerability was highlighted.
The second case study looked at atmosphere-ecosystem interactions. In
the earth system, both the dynamics and composition of the atmosphere affect
the biosphere. In turn, uptake, storage, and emissions by the biosphere affect the
composition and dynamics of the atmosphere. Changing atmospheric conditions
(e.g., changes in chemical composition and physical characteristics) together con-
stitute multiple stresses to ecosystems, and the resultant ecosystem impacts and
atmospheric feedbacks are poorly understood. Workshop participants addressed
how global/climate change drivers affect atmospheric composition and dynamics
and subsequent atmosphere-ecosystem interactions, as well as the socioeconomic
impacts of climate change on agriculture and carbon cycling, capture, and seques-
tration as regards agriculture and forestry.
For both case studies the steering committee asked participants to explore
the historical record and identify unexpected findings that raise concern about
future responses to multiple stresses. In addition, the steering committee asked
participants to focus on areas where large uncertainties lie, processes that are
nonlinear and where predicting the integrated effect of multiple stresses is espe-
cially difficult, and on observed and potential thresholds (changes of state) that
may be beyond our current ability to predict. The steering committee also rec-
ognized important commonalities between the two cases and sought to draw
broader lessons about multiple stresses or multiple drivers. The workshop also
explored lessons learned from the Millennium Ecosystem Assessment (MEA,
2005) because this document provides a comprehensive summary of the state
of the world’s ecosystems and services and how they may change in the future.
Finally, workshop participants were asked to engage in a synthesis discussion
addressing tools, nonlinearities and thresholds, resilience, and the use of regional
studies.
NONLINEARITIES, THRESHOLDS, AND

THE VULNERABILITY-RESILIENCE CONTINUUM
In considering multiple stresses with respect to atmosphere-ecosystem inter-
actions and drought, we need to make explicit our conceptions of four criti-

INTRODUCTION 11
cal concepts: nonlinear responses, thresholds, vulnerability, and resilience. The

workshop organizers defined nonlinearity according to the approach adopted by
Rial et al. (2004), in which systems so characterized display significant imbal-
ances between inputs and outputs, primarily (although not exclusively) episodic
and abrupt rather than slow and gradual change, and multiple equilibria. In such
systems small changes in parameters often cause large changes in the behavior
of the system. Nonlinearity also often confronts, if not causes, thresholds (i.e.,
phase changes) in a system that are not easily reversible. The global climate
system is inherently nonlinear and quite complex and consequently generates a
wide variety of thresholds for natural ecosystems and human social systems. But
humans themselves now generate planetary-scale environmental effects, includ-
ing modification of the global climate system. The interaction of these two drivers
multiplies the stresses that ecosystems and human social systems must face.
Because the climate system is highly nonlinear and coordination and control of
human forcing is weak, thresholds are encountered quite often. These thresholds
present severe policy challenges. Thresholds also raise questions about vulner-
ability and resilience of both ecosystems and human social systems. Vulnerability
refers to magnitudes and rates of environmental change that exceed the capacity
of ecosystems and human social systems to cope and recover. These systems
then either break down or exhibit a variety of pathologies under those conditions.
Resilience, on the other hand, refers to a capacity to adapt to and cope with the
level of stress imposed, even if damage to the system results.
The application of these concepts can be illustrated with respect to the issue
of drought. In Figure 1-1, Wilhite and Buchanan-Smith (2005) illustrate the
important differences between societies that are either vulnerable to or resilient
in the face of drought. (Box 1-1 defines vulnerability, resilience, and other rel-
evant terms.) The figure shows that societal vulnerability begins with exposure
to drought in the absence of either risk-based drought management policies or
an early warning system. Vulnerability is then enhanced by multiple-stresses
problems in the form of marginalized groups who lack resources, options, and
ways of mitigating drought impacts. Vulnerability is also enhanced by previously
existing problems of dependence on the overexploitation of natural resources,
poverty, and violent conflict. As the drought event intensifies, society resorts to
crisis management in the face of potential disaster. How then is resilience to be
developed? The two most critical steps represent the development of a culture
of prevention by enacting risk-based drought policies and plans combined with
an early warning system. Building political capital means solving some of the
multiple-stresses problems that impinge on the society and retard adaptation to
drought. In addition, once drought mitigation actions have been implemented and
specific impacts avoided or reduced, serious attention needs to be paid to lessons
learned. This learning can then be fed back into the planning and early warning
subsystems. The vulnerability-resilience continuum is considered in more detail
in the next chapter.

Nature
Societal response to
living with risk
Factors contributing to vulnerability
Consequences,
reinforcement
FIGURE 1-1 Illustration of the difference between drought-vulnerable societies (A) and
drought-resilient societies (B).
SOURCE: Wilhite and Buchanan-Smith (2005). 1-1aCopyright 2005 from Drought and Water
Crises by D. Wilhite, ed. Reproduced by permission of Routledge/Taylor & Francis Group,
LLC. Modified from ISDR (2003).

INTRODUCTION 13
Nature
Societal response to
living with risk
Consequences,
reinforcement
1-1b

BOX 1-1
Key Definitions
Adaptation: Adjustment in natural or human systems in response to actual or

expected climatic stimuli or their effects, which moderates harm or exploits ben-
eficial opportunities. Various types of adaptation can be distinguished, includ-
ing anticipatory and reactive adaptation, private and public adaptation, and
autonomous and planned adaptation.
Adaptive capacity: The ability of a system to adjust to climate change (including

climate variability and extremes) to moderate potential damages, to take advan-
tage of opportunities, or to cope with the consequences.
Resilience: Amount of change a system can undergo without changing state.
Sensitivity: Sensitivity is the degree to which a system is affected, either adversely

or beneficially, by climate-related stimuli. The effect may be direct (e.g., a change
in crop yield in response to a change in the mean, range, or variability of tempera-
ture) or indirect (e.g., damages caused by an increase in the frequency of coastal
flooding due to sea level rise).
Vulnerability: The degree to which a system is susceptible to, or unable to cope

with, adverse effects of climate change, including climate variability and extremes.
Vulnerability is a function of the character, magnitude, and rate of climate variation
to which a system is exposed, its sensitivity, and its adaptive capacity.
SOURCE: IPCC (2001).

Drought
There is no universal definition of drought. It has been defined in many ways

by various disciplines because the characteristics of drought and its impacts
reflect the climate and societal characteristics of the region affected. Conceptu-
ally, drought is a deficiency of precipitation from expected or “normal” condi-
tions that extends over a season or longer period of time and where water is thus
insufficient to meet the needs of human activities and the environment. Four
types of drought are generally recognized: meteorological, agricultural, hydro-
logical, and socioeconomic. Drought risk is a product of a region’s exposure
to natural hazards and its vulnerability to extended periods of water shortage.
Drought is different from many other natural hazards in that it does not begin
and end swiftly. Its onset is gradual, and as intensity and duration increase, the
effects correspondingly become more widespread. This creeping phenomenon
is expressed through multiple indicators because impacts are nonstructural and
spread over large areas.
All meteorological droughts begin with a deficiency of precipitation over
time. But nature is not the only cause of droughts. Droughts can be socially
constructed or amplified when there is insufficient precipitation to meet needs
(e.g., needs may be greater than “normal” supply if unsustainable development
has occurred). Therefore, droughts are caused by changes in both supply of and
demand for water, and both are dynamic. Impacts of drought are multifaceted.
For instance, for agriculture, lack of precipitation affects the critical variables of
soil moisture and evapotranspiration, and these become the vectors of adverse
changes for society. For water supply for human uses, as the intensity, dura-
tion, and spatial extent of drought increase, the natural dimension decreases in
15

FIGURE 2-1 Percent area of the United States in severe and extreme drought, January
1985-July 2005.
SOURCE: Based on data from the National Climatic Data Center/NOAA.
significance and a high priority is placed on water resources management. Some

water shortages can be the result of overappropriation of supply so that “drought”
occurs even in years of “normal” precipitation.
On average approximately 15 percent of the United States is affected by
drought each year, based on the historical record from 1895 to now (National
Climatic Data Center/NOAA, 2005). Figure 2-1 illustrates recent drought trends.
Figure 2-2 illustrates longer-term trends for one location and shows a number of
periods noteworthy for their duration, severity, and spatial extent.
A still ongoing drought began in 1996 for large parts of the country. Approxi-
mately 35 to 40 percent of the country has been affected at one time or another by
severe to extreme drought during this period, and for some regions drought condi-
tions have persisted for five or more years. For example, parts of the Southeast,
particularly Georgia, North Carolina, South Carolina, and Florida, experienced
three to four consecutive years of drought between 1999 and 2002. In the West
much of the Southwest experienced five consecutive years of drought between
2001 and 2004, while much of Montana, Idaho, and surrounding states have
experienced severe drought for as many as seven consecutive years since 1999.
The most recent drought is particularly notable in that it was hotter than a similar
drought of the 1950s. In general, western North America has seen significant
warming over the last 100 years, particularly in the last couple of decades.
The recent period of unprecedented population increase in the western
United States coincided with one of the wettest periods on record. During the

DROUGHT 17
FIGURE 2-2 Precipitation reconstruction for the Bighorn Basin, Wyoming/Montana.

SOURCE: Compiled from data from Gray et al., 2004.
1990s, for example, the population of Nevada increased by 66 percent, Arizona

by 40 percent, Colorado by 30 percent, Idaho by 28 percent, and New Mexico by
20 percent. The rapidly expanding population in the region is exacerbating water
conflicts between multiple and competing users. It is also worth noting that the
most important management agreement in the West (the Colorado River Compact
of 1922) was based on an overestimation of the reliable average annual supply of
water (estimated at 16.4 million acre feet [MAF] when the true long-term aver-
age flow of the Colorado River is estimated at between 12 and 16 MAF, with the
current best estimate at 13.5 MAF) because the allocation was based on a short
observational record that represented one of the region’s wettest periods.
The paleoclimate record, seen through proxy records such as tree rings, coral,
and boreholes in ice, allows scientists to estimate precipitation amounts back
much farther than the past few hundred years of instrument records, and it is clear
that significant droughts are not solely a modern phenomenon (Cook et al., 2004;
Woodhouse and Overpeck, 1998). Droughts comparable to those of the 1930s,
1950s, or early 2000s have occurred in general once or twice per century over the
last 2000 years. Research by Clark et al. (2002) using lake sediment records from
Connie Woodhouse, personal communication, June 6, 2006.

Much of the material in this section is taken from the presentation of Dr. Jonathan Overpeck,
University of Arizona, at the workshop, September 29-30, 2005.

the northern Great Plains shows pronounced 100- to 130-year drought cycles
back at least 8,000 years.
Droughts of the 20th century were eclipsed by past droughts in terms of
annual maximum severity, duration of drought, and geographic extent of drought.
The paleoclimatic record indicates that droughts longer than a decade (i.e.,
“megadroughts”) were not rare, and that droughts affecting much of the western
United States have lasted as long as a century or more within the last 2,000 years
(Gray et al., 2004). The true range of “natural” drought variability is thus substan-
tially larger and more complex than suggested by the last century, when there are
accurate records of drought variations provided by modern instrumentation.
In general, most of the large droughts of the western United States have
affected more than one major river basin at a time, and some (e.g., a megadrought
in the late 16th century) apparently affected the United States from coast to coast
and from northern to southern borders. Another key aspect of drought variability
illuminated by the paleoclimatic record is that decades- to centuries-long hydro-
logical “regimes” (e.g., characterized by rare/short or frequent/longer droughts)
have begun and ended abruptly; transitions between drought regimes can take
place over years to decades, whereas the regimes themselves can be significantly
longer.
Great strides have been made in recent years with respect to understanding
the proximal cause of drought in North America. Drought in the southwestern
United States (e.g., 1950s and the recent drought) is known to be connected with
the El Niño/Southern Oscillation (ENSO), and dry winters are favored in La
Niña years. More recently, studies have confirmed that anomalous sea surface
temperature (SST) patterns, particularly in the tropics and the Indo-Pacific, can
explain both 20th-century and earlier droughts, and research has identified strong
statistical associations between decadal modes of Pacific and Atlantic variability
with decadal patterns of wet and dry conditions over North America (McCabe
et al., 2004; Hoerling and Kumar, 2003). Of course, the major challenge is to
explain the causes of the anomalous—and persistent—ocean conditions that lead
to North American drought.
There is increasing consensus that anthropogenic forcing will likely increase
the probability of drought in central and western North America. Exacerbating
this likelihood is the fact that temperatures are already rising significantly in the
American West, and snowpack is retreating in the same region. A lesson of the
paleoclimatic record is that anthropogenic forcing could trigger an abrupt transi-
tion into a more drought-prone climatic regime, thus increasing the frequency and
duration of drought. Given that these possibilities could materialize with or with-
Jonathan Overpeck, presentation at the workshop, September 29-30, 2006.

Ibid.
Ibid.
E.g., IPCC TAR,CLIVAR/PAGES/IPCC Workshop: A multi-millennia perspective on drought and
implications for the future, November 18-21, 2003, Tucson, Arizona.

DROUGHT 19
out significant future human-induced climate change, it makes sense to consider

no-regrets strategies to reduce vulnerability to drought in either case.
CONTEXT AND IMPACTS

As noted earlier, drought was selected for examination as a case study because
it both is driven by multiple environmental stresses and leads to multiple stresses
across a wide range of scales. Drought is, at its simplest, an imbalance between
water supply and water demand. Yet in reality, so many variables are present in
both sides of that equation that it is clearly a classic example of a multiple-stresses
scenario, where many factors combine in ways that are not entirely predictable.
Drought typically evolves slowly, and as it progresses the impacts accumulate and
expand in scope, extent, and intensity. From any one impact, there can be cascad-
ing impacts. This is clearly a challenge for decision makers who must plan for and
react to the accumulating impacts.
Both climate and society are dynamic, as are the relationships between
them (i.e., impacts and vulnerabilities). The different influences occur on vary-
ing scales (both time and space) and thus are challenging to predict with any
accuracy. Drought is indeed a normal component of climate variability, but as a
generator of multiple environmental stresses, it is the longer droughts—multiyear,
multidecadal, and even centennial—that are of greatest concern. The longer the
drought, the wider the range of ecosystem and societal effects that cascade from
primary to tertiary impacts as time lengthens and spatial scales widen.
Terrestrial ecosystems respond not only to temperature but particularly to
decreasing soil moisture that in turn induces woody plant mortality, rapid canopy
change, and increased soil erosion. Drought-induced tree mortality results in
heightened vulnerability to fires. Note that canopy fires are the largest type of
wildfire events in forests (Strauss et al., 1989). Increased temperature over long
periods of time can also facilitate larger infestations of pests that magnify tree
mortality and thereby expand the spatial scale of forest fires. Ecotones, or zones
of transition between ecosystems, appear to be most susceptible to rapid vegeta-
tion change under stress (Allen and Breshears, 1998). As reductions in herba-
ceous groundcover increase, the distribution of near-ground energy is altered and
can affect a wide range of ecosystem processes dependent on that energy. The
threat of nonlinear increases in soil erosion rises significantly as a consequence
of that shift in near-ground energy.
As drought intensifies over time, its spatial scale expands and its societal
ramifications deepen (Wilhite et al., 2005). Infrastructures are affected, as is the
supply of social services, and there emerge serious distributive consequences for
the less well off members of society. Trends in the spatial distribution of water,
Jonathan Overpeck, presentation at the workshop, September 29-30, 2005.

David Breshears, presentation at the workshop, September 29-30, 2005.

population size, demand, and competing uses all gain heightened significance and
increased stress at times of increasing water scarcity.
Rates of change in the timing and supply of water, as well as sequencing of
wet and dry years, can aggravate the impacts of drought. When wet and dry years
can alternate frequently, human and ecological systems adapt to the extremes of
variability; multiyear variability, on the other hand, presents the illusion of stabil-
ity and the human impacts can be greater because people plan inadequately. Not
only do human expectations differ based on timing and sequencing, but insects,
pathogens, and other pests also respond in nonlinear fashion to some hydrological
trends. Sequencing of water availability in the second half of the 20th century
and into the first decade of the 21st century is of particular significance to the
American West. For instance, snowpack provides up to 80 percent of the runoff,
and since the 1950s there has been a long-term trend in increasing temperature
and decreasing snowpack (Hamlet et al., 2005). The onset of spring and growing
season evapotranspiration, the timing of snowmelt and snowmelt discharge, and
the amount of recharge as the proportion of precipitation shifts from snow to rain
at critical elevations will all have effects (Stewart et al., 2005; Cayan et al., 2001;
Knowles et al., 2006). Increasing temperature, even if precipitation remains con-
stant, has the potential to dramatically alter the hydrology of river basins and the
severity of drought episodes. These types of trends are likely to lead to multiple
stresses on both ecosystems and human social systems, thereby exacerbating
competition between the two. Given these challenges, there is an acute need for
further development of our capacity to predict the onset of drought. We now
recognize that the true range of natural drought variability is substantially larger
and more complex than is suggested by the 20th-century instrumental record.
Research (Hoerling and Kumar, 2003) shows that while land surface feedbacks
with the atmosphere can be important to amplify or dampen drought in some
locations or seasons, these feedbacks are not always dominant in driving temporal
drought variability, and patterns in coupled atmosphere-ocean variability also
play an important role.
If the United States and other nations are to make progress in reducing
the serious consequences of drought, an improved understanding of the hazard
and its prediction and the full range of factors that influence vulnerability is
needed. Enhancing our knowledge of the hazard will require a complex, inte-
grated early warning system that incorporates climate, soil, and water supply
factors such as precipitation, temperature, soil moisture, snowpack, reservoir
and lake levels, groundwater levels, and streamflow. The implementation of the
National Integrated Drought Information System (NIDIS), currently underway
within the National Oceanic and Atmospheric Association (NOAA), represents a
multiagency and multiorganizational effort and was viewed by workshop partici-
pants as an important step in the development of an improved decision-support
system for the country.

DROUGHT 21
Because of the slow-onset nature of drought, the ability of resource managers

to adapt and/or impose alternative management practices in a timely fashion
would be greatly enhanced by more reliable seasonal climate forecasts. Unfortu-
nately, little skill currently exists to reliably predict the onset, severity, duration,
spatial extent, or end of a drought event a season or more in advance. This is a
critical research need. In addition to improved seasonal forecasts, there is need
for improved methods of probability risk assessments that rely on reconstructions
of past climates and account for climatic nonstationarity to calculate the occur-
rence and return probabilities of drought (Enfield and Cid-Serrano, forthcoming).
Improving seasonal forecasts and their application will also require collabora-
tion between scientists at research institutes, universities, and federal agencies.
The primary end users of these forecasts need to be involved in this process so
researchers understand their needs.
UNDERSTANDING VULNERABILITY AND RESPONSE STRATEGIES

Vulnerability to drought is dynamic and is influenced by a multitude of fac-
tors, including increasing population, regional population shifts, urbanization,
technology, government policies, land use and other natural resource manage-
ment practices, desertification or land degradation processes, water use trends,
and changes in environmental values (e.g., protection of wetlands or endangered
species). Therefore, the magnitude of drought impacts may increase in the future
as a result of an increased frequency of occurrence of the natural event (i.e.,
meteorological drought), changes in the factors that affect vulnerability, or a
combination of these elements. The development of a national drought policy
and preparedness plans at all levels of government that place emphasis on risk
management rather than following the traditional approach of crisis manage-
ment would be a prudent step for the United States to take. Crisis management
decreases self-reliance and increases dependence on government and donors.
The impacts of drought in recent years have been increasing, although there
is no systematic national assessment of drought impacts such as exists with other
natural disasters. The Federal Emergency Management Agency (FEMA) (1995)
estimated annual losses in the United States because of drought at $6 to $8 bil-
lion, making drought the most costly natural disaster in the country until Hurri-
cane Katrina in 2005. Losses from the 1988 drought have been estimated at more
than $39 billion. The National Drought Mitigation Center (Hayes et al., 2004) has
estimated that the losses associated with the 2002 drought exceeded $10 billion,
based on estimates made by 10 states. If these losses are extrapolated to include
all drought-affected states, they would be significantly higher. It is important to
note that these are estimates for a single drought year; major drought events often
occur over a series of years (e.g., 1999-2005).
The impacts of drought have also been growing in complexity. Historically,
the most significant impacts associated with drought have occurred in the agricul-

tural sector (i.e., crop and livestock production). There has been an expansion of
impacts in other sectors, particularly energy production, recreation and tourism,
transportation, forest and wildland fires, urban water supply, environment, and
human health. The recent drought years in the western United States, for example,
have resulted in financial impacts in nonagricultural sectors that have likely
exceeded those in agriculture. In addition to the direct impacts of drought, there
are significant indirect impacts that, in many cases, exceed in value the direct
losses associated with drought episodes. In addition to these human-focused
impacts, there are the effects on nonhuman systems that are even more difficult
to quantify.
DROUGHT POLICY AND PREPAREDNESS

In the past decade or so drought policy and preparedness have received
increasing attention from governments, international and regional organizations,
and nongovernmental organizations. Simply stated, a national drought policy
is a way to establish a clear set of principles or operating guidelines to govern
the management of drought and its impacts. The ideal policy is consistent and
equitable for all regions, population groups, and economic sectors and is con-
sistent with the goals of sustainable development and the wise stewardship of
natural resources. The overriding principle of drought policy is an emphasis on
risk management through the application of preparedness and mitigation mea-
sures. Preparedness refers to predisaster activities designed to increase the level
of readiness or improve operational and institutional capabilities for responding
to a drought episode. Mitigation is short- and long-term actions, programs, or
policies implemented during and in advance of drought that reduce the degree
of risk to human life, property, and productive capacity. These actions are most
effective if done before the event. Emergency response will always be a part
of drought management because it is unlikely that government and others can
anticipate, avoid, or reduce all potential impacts through mitigation programs. A
future drought event may also exceed the capacity of a region to respond. How-
ever, emergency response is best used sparingly and only if it is consistent with
longer-term drought policy goals and objectives.
A key component of a national drought policy is to reduce risk by develop-
ing better awareness and understanding of the drought hazard and the underly-
ing causes of societal vulnerability (Hayes et al., 2004). The principles of risk
management can be promoted by encouraging the improvement and application
of seasonal and shorter-term forecasts, developing integrated monitoring and
drought early warning systems and information delivery systems, developing
preparedness plans at various levels of government, adopting mitigation actions
and programs, and creating a safety net of emergency response programs that
ensure timely and targeted relief. The delivery of information in a timely manner

DROUGHT 23
is essential so that informed decisions can be made by resource managers and

others.
The traditional approach to drought management has been reactive, rely-
ing largely on crisis management. This approach has been ineffective because
response is untimely, poorly coordinated, and poorly targeted to drought-stricken
groups or areas. In addition, drought response is post-impact and relief tends to
reinforce existing resource management practices (i.e., it rewards poor resource
management and the lack of preparedness planning). Many governments are
striving to learn how to employ proper risk management techniques to reduce
societal vulnerability to drought and therefore lessen the impacts associated with
future drought events.
There are four key components in an effective drought risk reduction strategy
(O’Meagher et al., 2000): (1) the availability of timely and reliable information on
which to base decisions; (2) policies and institutional arrangements that encour-
age assessment, communication, and application of that information; (3) a suite
of appropriate risk management measures for decision makers; and (4) actions
by decision makers that are effective and consistent. In 1992 Australia adopted a
national drought policy that applied these components through three objectives:
(1) to encourage primary producers and other sections of rural Australia to adopt
self-reliant approaches to managing for climatic variability; (2) to maintain and
protect Australia’s agricultural and environmental resource base during periods of
extreme climate stress; and (3) to ensure early recovery of agricultural and rural
industries, consistent with long-term sustainable goals (O’Meagher et al., 2000).
Australia’s national drought policy is widely known and its philosophy adaptable
in other settings (Botterill, 2005).
In the United States there has been some progress in the development of pre-
paredness plans. The most noticeable progress has been at the state level, where
the number of states with drought plans has increased dramatically during the
past two decades. In 1982 only three states had drought plans in place; in 2005,
38 states had developed plans. The basic goal of state drought plans should be
to improve the effectiveness of preparedness and response efforts by enhancing
monitoring and early warning, risk and impact assessment, and mitigation and
response. Plans should also contain provisions (i.e., an organizational structure
or framework) to improve coordination within agencies of state government
and between local and federal government. Initially, state drought plans largely
focused on response efforts aimed at improving coordination and shortening
response time; today the trend is for states to place greater emphasis on mitigation
as the fundamental element of a drought plan. Thus, some plans are now more
proactive, adopting a more risk management approach to drought management.
Drought mitigation plans have three essential components, regardless of
whether they are developed at the state, national, regional, or local scale.

1. A comprehensive monitoring and early warning system provides the basis

for many of the decisions that must be made by a wide range of decision makers
as drought conditions evolve and become more severe. Equally important, early
warning systems need to be coupled to an effective delivery system that dis-
seminates timely and reliable information. As drought plans incorporate more
mitigation actions, it is imperative that these actions be linked to thresholds (e.g.,
reservoir levels, climate index values) that can serve as triggers for mitigation and
emergency response actions.
2. A critical step in the development of a mitigation plan is the conduct of a
risk assessment of vulnerable population groups, economic sectors, and regions
(Knutson et al., 1998). The purpose of the risk assessment is to determine who
and what is at risk and why. This is accomplished through an analysis of historical
and recent impacts associated with drought events.
3. After impacts have been identified and prioritized, the next step is to
identify appropriate mitigation actions that can help to reduce the risk of each
impact for future drought events.
One existing need is quantifying the advantages of a risk-based drought

mitigation planning effort over the crisis management approach so policy makers
see the advantages of committing resources up front to develop and implement
mitigation actions rather than waiting to deal with impacts during a crisis. In most
cases the costs associated with mitigation actions are minimal when compared
with the costs of drought, which often are in the billions of dollars.
The U.S. Congress has considered actions that could be taken in response
to recommendations issued in May 2000 by the National Drought Policy Com-
mission (NDPC) but has not moved on these discussions. One of the NDPC’s
recommendations strongly endorses drought planning at all levels of govern-
ment. Legislation has been introduced in the U.S. Congress that could lead to
the creation of a more permanent national drought council and a national drought
policy. Key components of this bill included an emphasis on risk management,
preparedness planning, and improvement of the nation’s drought monitoring sys-
tem and forecasting capabilities. A project, the National Integrated Drought Infor-
mation System (NIDIS), is intended to provide the foundation for development
of an improved drought monitoring system (Western Governors’ Association,
2004). NIDIS is one of the components of the National Drought Preparedness
Act, and authorization for NIDIS has been included in two bills introduced in
Congress (HR 5136, S2751) in spring 2006.
http://www.westgov.org/.

DROUGHT 25
RESEARCH NEEDS
During the workshop, once the formal presentations were complete, the
participants brainstormed a variety of possible research questions and needs. It
was noted that the ability to make projections about the future is what makes
knowledge powerful, and that in drought this means that much of the work is site
specific and situation dependent. Because investment strategies related to water
management and drought mitigation will differ depending on the size, cost, and
service life of the strategy or facility, such projections are particularly important
to policy makers. Integrated analysis of multiple stresses needs to replace cause-
effect-type analysis. Much needs to be done to link large and small scales so that
broad-based knowledge actually has practical application on the ground. There
are still major gaps in understanding and communicating to and from the user
community, and this inherently includes education and outreach. There is great
value to studying the paleo record both because this gives us a basis for under-
standing megadrought and because looking at the past gives a greater time span
over which relationships between duration, frequency, severity, extent, and loca-
tion can be defined and quantified at the continental scale. There is also a need to
consider the impacts on the services provided by ecosystems.
Finally, participants discussed possible steps to improve our capability to
integrate science knowledge so that we are better able to deal with multiple
stresses in decision support, with a focus on research needs. For the drought case
study, the participants listed the following as steps that could advance our under-
standing of the multiple-stresses components and interactions for drought:
• Implement the National Integrated Drought Information System, empha-

sizing a partnership between federal and nonfederal agencies and organizations.
This would improve monitoring and early warning systems to provide better
and more timely and reliable information to decision makers; address data gaps
in drought monitoring and enhance networks, particularly for soil moisture,
snowpack, and groundwater; and develop new monitoring and assessment tools/
products that will provide resource managers at all levels with proper decision-
support tools at higher resolution.
• Help the scientific and policy communities and resource managers better
understand drought as a complex natural hazard. For example, more effort could
be made to communicate information on probabilities of single- and multiple-year
drought events to natural resource managers and planners, policy makers, and the
public in ways and formats that make sense to different audiences. Paleoclimate
and historical climate research could be done to better understand past droughts
at the regional scale. Additional research could be done on nonlinear processes
and thresholds to provide improved prediction capability about the connections
between tree mortality, energy and water budget shifts, and soil erosion.
• Improve the reliability of seasonal climate forecasts and train end users
on how to apply this information to improve resource management decisions

with the goal of reducing drought risk. One part of this might be to develop
more competitive research grant programs to fund research on drought predic-
tion. In particular, there is a need for enhanced observations and research on both
the paleoclimate record and the drought-related dynamics of ocean-atmosphere
coupling. Another idea might be to form a consortium of scientists to encourage
collaboration on drought prediction. New funding mechanisms might be needed
that explicitly encourage multidisciplinary research bridging the gap between
physical and biological science and human needs. Finally, it might be useful to
develop a network of scientists and end users to assess the practical needs of end
users and how forecast information can be communicated more effectively to the
user community to maximize its application.
• Assess the economic, social, and environmental impacts associated with
drought. Unless we improve our understanding of human behavior, the best inten-
tioned plans will continue to produce less than desired results. The inadequate
assessment of drought costs continues to be a significant problem in commu-
nicating the importance of drought mitigation to the management and policy
communities. More accurate assessments of the true impacts of drought would
provide greater justification for investments in mitigation actions at the local,
state, and regional levels. Finally, work could be done to improve early assess-
ments of drought impacts through the application of appropriate models (i.e.,
crop, hydrological).
• Assess the science and technology needs for improving drought planning,
mitigation, and response at the local, state, tribal, regional, and national levels. To
do this, it might be necessary to evaluate current drought planning models avail-
able to governments and other authorities for developing drought mitigation plans
at the state and local levels of government and require plans to follow proposed
standards or guidelines. Efforts could be made to identify improved triggers (i.e.,
links between climate/water supply indicators/indices and impacts) for the phase-
in and phase-out of drought mitigation and response programs and actions during
drought events. Work could be done to develop vulnerability profiles for various
economic sectors, population groups, and regions and to identify appropriate
mitigation actions for reducing vulnerability to drought for critical sectors.
• Increase awareness of drought, its impacts, trends in societal vulnerability,
and the need for improved drought management. This might include initiating
K-12 drought/water awareness programs/curricula or launching public aware-
ness campaigns for adult audiences, directed at water conservation and the wise
stewardship of natural resources.
• Design more focused and systematic education and outreach programs for
stakeholders based on information derived from periodic surveys of their inter-
ests. From the results of such surveys, design workshops tailored to the specific
interests of different combinations of stakeholders with the objective of producing
decision-support tools on a continuing basis.

Atmosphere-Ecosystem Interactions
The atmosphere and the earth’s ecosystems are parts of a coupled system.
For a large variety of processes, forcing from one partner in the interaction
elicits one or more responses from the other partner, which in turn elicits other
responses from the first. This bidirectional coupling gives atmosphere-ecosystem
interactions the potential to be among the most complex in the natural world. The
increasing involvement of human actions as important drivers introduces a broad
new suite of responses and interactions. Historically, most of the study of drivers
and responses in atmosphere-ecosystem interactions has started with single-factor
investigations, building on the infrastructure, concepts, and tools of particular dis-
ciplines. Over the last several decades new knowledge has continued to accumu-
late in the traditional disciplines, but more and more of the breakthroughs are at
the borders of traditional disciplines. Climate dynamics, hydrology, atmospheric
chemistry, ecology, oceanography, and geomorphology function increasingly as a
single superdiscipline, often called earth system science. In the future continued
progress in this new superdiscipline is likely to require effective collaboration
with or integration of a wide range of human sciences, from agronomy and civil
engineering to economics and government.
The potential importance of bidirectional interactions is long acknowledged
but relatively little studied, at least until recently. For example, Ahhrenius’s
calculations (1896) of climate forcing from coal combustion identified key com-
ponents in anthropogenic warming, and in the 19th century the claim that rain
follows the plow was a powerful inducement for agricultural expansion in the
western United States. Following the introduction of climate models, insights on
bidirectional coupling began to emerge. Studies by Charney et al. (1975, 1977)
27

on the role of vegetation (or lack of vegetation) in modulating the climate of the
Sahara are classic foundations for the modern science of atmosphere-ecosystem
interactions. Later studies on the role of vegetation in the climate of the Amazon
basin (Salati and Vose, 1984; Shukla et al., 1990; Dickinson and Henderson-
Sellers, 1988; Lean and Warrilow, 1989) began to bring human actions into the
science of atmosphere-ecosystem interactions. At about the same time, analyses
of deforestation indicated its potentially large contribution to climate forcing
through the carbon cycle (Woodwell et al., 1983). Also around this time a series
of breakthroughs established the role of chemicals released from plants and from
human processes in modulating the chemistry of the atmosphere (ozone hole,
biogenic volatile organic components).
Since these early discoveries, understanding the nature and implications of
atmosphere-ecosystem interactions has been one of the central goals in earth
system science. It is also increasingly clear that understanding atmosphere-
ecosystem interactions is one of the fundamental prerequisites for designing a
path to a sustainable future.
CONTEXT AND IMPACTS

The earth, oceans, atmosphere, and human actions need to be considered as
a single, coupled system for a thorough understanding of climate, ecosystems,
hydrology, or atmospheric chemistry. At small spatial and temporal scales, the
coupling ceases to be of first-order importance. But at larger scales of space
and time, the coupling between the atmosphere, land ecosystems, and oceans is
always relevant and often dominant.
In the coupled earth system, components respond differently to different
forcings. Responses are often nonlinear and often have threshold-type character-
istics, with little response over a wide range of forcing, followed by a transition
to a fundamentally new state over a short time or a narrow range of forcing.
Understanding the locations of these thresholds and the mechanisms controlling
them is among the most important challenges in earth system science. The lack
of an obvious response to initial forcing can lead to the incorrect conclusion that
a component of the system is insensitive to the altered environment.
Many of the behaviors of parts of the earth system have clear threshold
responses. Wildfires, for example, almost never occur until temperature, humid-
ity, fuel load, and fuel moisture enter the permissive range. But when all the
environmental conditions are compatible with sustaining a wildfire, risks increase
rapidly. This wildfire threshold could have important implications for Amazon
rainforests if the future is warmer and drier (Nepstad et al., 1999). It could
also interact in an important way with anthropogenic burning given the recent
evidence that aerosols from Amazon fires can decrease rainfall (Andreae et al.,
2004). And in a clear example of a feedback, reduced rainfall over tropical land
masses during El Niño events has been shown to encourage more biomass burn-

ATMOSPHERE-ECOSYSTEM INTERACTIONS 29
ing in the tropics, which in turn yields higher annual concentrations of carbon
dioxide (CO2) in the atmosphere; knowing this could lead resource managers to
greater policing in El Niño years in an attempt to reduce the extent of burning by
agriculturalists and reduce carbon emission.
Other important examples of threshold come from the response of temperate
forest ecosystems to warming or the deposition of atmospheric nitrogen. In
controlled ecosystem experiments nitrogen inputs produce little change over
several years, but the nitrogen excess eventually reaches a point where the system
collapses. In response to warming the initial response is a large increase in soil
warming, followed by a sudden decline when the ecosystem runs out of easily
decomposable material.
Some of the important thresholds in earth system responses can operate in
more than one direction. One good example of this is the relationship of atmo-
spheric ozone to levels of volatile organic compounds (VOCs) in the atmosphere.
Depending on the ratio of VOCs to nitrogen oxides (NOx), an increase in VOCs
could lead to a large decrease or a large increase in ozone production. Changes
in land use can behave as thresholds. Often, dramatic changes in land use
follow changes in policy, price supports, or transportation infrastructure. If con-
sequent changes in local climate make the changes in land use difficult to reverse
(Dickinson and Henderson-Sellers, 1988; Lean and Warrilow, 1989; Shukla et al.,
1990), the changes that occur across a narrow threshold can be locked in place.
Interactions between the earth, oceans, and atmosphere often involve the
simultaneous action of diverse mechanisms. Terrestrial and ocean carbon balance
provide beautiful examples of the overall fluxes controlled by a large number
of individual mechanisms. In the oceans, temperature interacts with alkalinity,
salinity, and dissolved inorganic carbon to control CO2 solubility (Sabine et al.,
2004). Biological processes are also important contributors to the carbon balance
of the oceans, with potentially subtle changes in the composition of the producer
and consumer communities leading to substantial effects on the downward trans-
port of particulate carbon (Sabine et al., 2004). On longer timescales the delivery
of mineral nutrients from upwelling or from the delivery of windborne dust plays
an important role.
On land, diverse processes contribute to the overall carbon balance (Pacala
et al., 2001). The current carbon balance of the United States has large influ-
ences due to land use change, CO2 fertilization, nitrogen deposition, ozone,
and climate. The early optimism that future terrestrial carbon dynamics might
be modeled as a simple response to atmospheric CO2 (Bacastow and Keeling,
1973) has been replaced by an appreciation that drivers from human actions,
Jerry Melillo, presentation at the workshop, September 29-30, 2005.

Ibid.
Alex Guenther, presentation at the workshop, September 29-30, 2005.
PatriciaRomero-Lankao, presentation at the workshop, September 29-30, 2005.
Jerry Melillo, presentation at the workshop, September 29-30, 2005.

atmospheric composition, climate, and ecological processes all interact, with

contrasting relative importances in different settings.
Climate change is expected to influence the capacities of the land and oceans
to act as repositories for anthropogenic carbon dioxide and in turn provide a
feedback that affects climate further. Modeling experiments show that carbon sink
strengths vary with the rate of fossil fuel emissions, so carbon storage capaci-
ties of both land and oceans decrease and climate warming increases with faster
emissions (Fung et al., 2005).
The bidirectional nature of earth-atmosphere interactions has important
implications for a wide range of earth system processes. Coupling plays a central
role in both carbon-climate and climate-albedo feedbacks. Many experiments and
simulations indicate that, depending on the starting point, a warmer climate can
lead to either a loss or gain of ecosystem carbon, (Mack et al., 2004; Shaver,
et al., 2006). If in response to warming, ecosystems lose carbon, then atmo-
spheric carbon increases, producing a positive feedback on the initial warming.
If warming leads to an increase in ecosystem carbon (with more carbon in plants
and soils), then the feedback is negative. A number of model experiments now
explore the role of land and ocean feedbacks in modulating the climate forcing
from atmospheric CO2. The general conclusion from these studies is that the
terrestrial feedback is positive (in the direction of exaggerating warming) (Cox
et al., 2000; Friedlingstein, 2004; Fung et al., 2005), although the magnitude of
the feedback is uncertain. The CO2 sensitivity of the climate model used in these
simulations plays an important role in determining the strength of the feedback,
as does the tendency of the ocean to take up the carbon released from the land.
The magnitude of the current uncertainty is large. With comparable forcing from
anthropogenic CO2 emissions, equally credible models end the 21st century
with atmospheric CO2 levels differing by more than 200 ppm, a quantity of CO2
greater than the total released by fossil fuel combustion to date.
Albedo-climate feedbacks may be equally important. New evidence indicates
that warming in the Arctic is already leading to increased abundances of shrubs,
which lead to an increase in the absorption of solar radiation, especially in the
spring, and reinforce the warming (Chapin et al., 2005). Simulation results indi-
cate that historical land use in the central United States has cooled the climate.
The lack of historical cooling reflects the combined effects of this albedo effect,
plus other processes that have counteracted it. Combining effects on albedo and
carbon storage, increasing forest biomass in the temperate latitudes tends to pro-
duce a net warming, while reforestation or afforestation in the tropics tends to
produce a net cooling (Gibbard et al., 2005).
Ibid.
Scott Doney, presentation at the workshop, September 29-30, 2005.
Ibid.

Humans exert and respond to a wide range of stresses in the coupled earth-
ocean-atmosphere system. Almost all studies of natural science components of
global change considered human drivers as a fixed set of boundary conditions,
and analyses of human responses viewed changes in climate or air quality as
givens. While these are clearly simplifications, research teams simply did not
have the breadth of expertise or the technical tools to tackle truly integrated
approaches. A few teams have recently made bold attempts to integrate human
actions and the natural sciences in an interactive framework. For example, the
scenarios developed for the Millennium Ecosystem Assessment use a modeling
framework that attempts to integrate changes in agricultural demand with changes
in climate, leading to, among other things, projections of deforestation and
prices of major agricultural crops (MEA, 2005). Recent Massachusetts Institute
of Technology Emissions Prediction and Policy Analysis simulations address
interactions among climate, ozone, crops, and the economy in a coupled frame-
work. They have also looked at air pollution, human health, and the economy as
a coupled system. Consistent with the early stage of this research, many of the
potentially most important drivers of change in patterns of human action have
not been explored with coupled models. Specifically, the impacts of HIV/AIDS
and other major epidemics could have major impacts on future human activity.10
The fundamentally important distribution of wealth, opportunity, and independent
decision making11 was a focus of the Millennium Ecosystem Assessment,12 but
its exploration with coupled models is just beginning.
Clearly, atmosphere-ecosystem interactions unfold through diverse processes,
across a range of scales, and with nonlinearities. We have some understanding
of a variety of the mechanisms involved, but there are many uncertainties. Much
uncertainty relates to the impacts of global change on humans, ecosystems, and
economies; interactive effects among these sectors have the potential to amplify
or suppress the initial effects, sometimes by a large multiplier. As with the
drought case study, this is a scenario where varied impacts can accumulate and
expand in scope, extent, and intensity. From one impact there can be cascading
impacts.
IMPACTS ON HUMANS, ECOSYSTEMS, AND ECONOMIES

Atmosphere-ecosystem interactions unfold through diverse processes. Cli-
mate, air pollution, droughts, and fires are all sensitive to controlling mechanisms
that have atmospheric components, ecosystem components, and components
that arise specifically from the interactions between them. Though most of these
John Reilly, presentation at the workshop, September 29-30, 2005.

10William Easterling, presentation at the workshop, September 29-30, 2005.
11Patricia Romero-Lankao, presentation at the workshop, September 29-30, 2005.
12Harold Mooney, presentation at the workshop, September 29-30, 2005.

mechanisms are understood in outline form, many of the details are unknown.
Much of the reason that the range of uncertainty related to impacts of global
changes on humans, ecosystems, and the economy is so large is that the interac-
tive effects have the potential to amplify or suppress the initial effects, sometimes
by a large multiplier.
Several kinds of human factors can exaggerate vulnerability to the impacts
modulated by atmosphere-ecosystem interactions. Poverty, lack of control over
one’s destiny, and an extremely unequal distribution of wealth all tend to decrease
coping capacity, increase vulnerability, degrade ecosystem services, and increase
the challenge of finding effective paths toward solutions.13 In contrast, human
factors that stimulate technical innovation, distribute control, and encourage
local decision making can decrease vulnerability while increasing ecosystem
services.14
Atmosphere-ecosystem interactions introduce potentially important uncer-
tainties into a large suite of future global changes. Characterizing these uncertain-
ties and, where possible, reducing them, is one of the central challenges of global
change research. Still, it is important to recognize that unknowns in the realm of
human actions increase the uncertainties even further.15 For a truly useful under-
standing of the range of global change processes, we need to develop useful ways
to more effectively integrate earth, atmospheric, and human processes. 16
POLICY OPTIONS: ADAPTATION AND MITIGATION

Atmosphere-ecosystem interactions have important impacts not because
they result in new phenomena but because they modulate a wide range of earth
and atmospheric processes. Especially in a context with multiple, simultaneous
interaction drivers, this modulation can be of primary importance. Atmosphere-
ecosystem interactions have the potential to amplify the impacts of minor
processes or suppress the impacts of major ones. They are the dominant sources
of uncertainty in some processes and a major source of uncertainty in others.
Increased understanding of these interactions is a major theme in global change
research.
In many settings increased understanding of atmosphere-ecosystem interaction
can play a central role in designing effective strategies for adapting to or mitigat-
ing impacts of global change. A central need is a thorough enough understand-
ing of these interactions to map the locations of thresholds, especially those that
cause positive feedbacks in global change responses. Examples of threshold
responses in these interactions are increasingly well developed. For instance, it
13Ibid.
14Ibid.
15Patricia Romero-Lankao, presentation at the workshop, September 29-30, 2005.
16Harold Mooney, presentation at the workshop, September 29-30, 2005.

is well known that warming that leads to increases in the abundance of Arctic
shrubs, which when the shrubs become common enough, decreases local albedo
and amplifies warming (Chapin et al., 2005). Another example is the effect of
nitrogen fertilization. While increasing deposition of reactive nitrogen (typically
as NO3 and NH4+) can lead to increased uptake of CO2, nitrogen fertilization
typically also results in emissions of nitrogen gasses (e.g., NH 3 and N2O, an even
more powerful greenhouse gas than CO2) (e.g., Vitousek et al., 1997). Also, nitro-
gen fertilization leading to nitrogen saturation of terrestrial ecosystems results in
the loss of nutrient cations, causing reduced productivity locally and eutrophi-
cation of aquatic systems downstream (Vitousek et al., 1997). Interactions can
also be multistep. For example, the effect of biogenic volatile organic carbon on
tropospheric ozone can lead to an indirect impact of high temperatures on crop
yields and forest growth.17 The negative effects of elevated ozone are, somewhat
surprisingly, not suppressed by growing trees in elevated CO2 (Karnosky et al.,
1999).
In some cases a thorough understanding of atmosphere-ecosystem inter
actions can provide an insurance policy against adaptations that fail to accomplish
their goals or that have undesirable side effects. For example, recent evidence
shows that especially in midlatitude forests the warming caused by decreased
albedo can be larger than cooling from carbon storage, providing an impor-
tant caveat in the motivation for broad reforestation efforts in the midlatitudes
(Gibbard et al., 2005; Feddema et al., 2005).
In other cases an atmosphere-ecosystem interaction can serve as an effective
foundation for successful mitigations even if they are unintentional. Increases
in plant growth and ecosystem carbon in response to elevated atmospheric CO 2
(Prentice et al., 2001) provide a classical example of negative feedback on
atmospheric carbon (see Figure 3-1). Another example concerns the possibility
that an ocean acidified enough, in response to high atmospheric CO2, to start
dissolving carbonate may dramatically increase its rate of CO 2 uptake (Sabine
et al., 2004).
In sum, atmosphere-ecosystem interactions do not establish a single set of
issues for adaptation and mitigation. Instead, they appear almost as a large suite
of risks and opportunities. Positive feedbacks have the potential to increase vul-
nerability, especially when responses cross thresholds. Negative responses have
the potential to amplify the utility of adaptation and mitigation measures. In gen-
eral, thorough understanding is critical, as the nature, direction, and magnitude
of likely feedbacks are rarely clear.
17Alex Guenther, presentation at the workshop, September 29-30, 2005.

Coupled Carbon-Nitrogen Dynamics
FIGURE 3-1 There is a strong feedback between decomposition and plant growth, and
soil mineral nitrogen is the primary source of nitrogen for plant growth. This can result in
a shift from being a carbon source to carbon sink under warming.
3-1 Doney
SOURCE: Scott Doney, presentation at the workshop, September 29-30, 2006.
RESEARCH NEEDS
The literature is increasingly rich with examples of important atmosphere-
ecosystem interactions, but few, if any, are thoroughly understood. The commu-
nity with the expertise to address questions in atmosphere-ecosystem interactions
is small. Investigators in this area need to combine a research-level understand-
ing of atmospheric processes with a sophisticated knowledge of terrestrial and
marine ecosystems. For investigators not equipped to tackle the coupled system,
collaborations are an essential tool, though interdisciplinary collaborations are
often difficult and complex. Because some of the interactions unfold only on
long timescales or large spatial scales, we need experimental, observational, and
simulation techniques to explore the range of possibilities. This kind of work will
require multifactor experiments.
From the workshop discussions participants identified the following inter
actions that could benefit from increased multifactor research. Participants noted
that for most of these questions, key elements of a comprehensive understanding
are in place, but resources for thorough study have been lacking. These items

are not prioritized or necessarily similar in scope but rather reflect the workshop
participants’ on-site thinking.
• How do CO2 and a warming climate interact to affect soil moisture,

ocean acidification, and the carbon balance of ecosystems (both terrestrial and
marine)?
• What is the role of biogenic VOCs in generating ozone, and what is the
role of ozone in degrading crop yields?
• What is the role of changes in surface albedo in changing climate? What
are the carbon-cycle implications of this?
• What is the role of air pollutants in degrading crop yields? How do these
effects change with warming?
• How might human pressures for increased food production lead to an
expansion of agricultural land, and what are the costs in ecosystem services for
the converted land?
• How do human decisions about cropping, land use or cover change, and
urbanization influence atmosphere-ecosystem interactions?
• How do greenhouse gas increases and associated warming, land use, and
air pollution interact with biodiversity?
• What is the relative role of extreme events and average conditions in
establishing the impacts of atmosphere-ecosystem interactions? Under what con-
ditions do atmosphere-ecosystem interactions enhance the value of investments
in adaptation and mitigation?
• How do drivers relate with stresses to produce certain vulnerabilities/
adaptive capacities?
• How do socioeconomic, institutional, and environmental processes influ-
ence environmental change and adaptive capacity (livelihoods, migration)?
• How does information get to those who need it? What kinds of informa-
tion are most useful to decision makers, resource managers, and others who could
benefit?
• How can we develop improved tools and strategies for addressing multiple
environmental stresses, such as improved observational and modeling capabilities,
integrated sensors, regional information systems, and predictive capabilities?
• How can societal resilience to multiple environmental stresses be improved?
How can adaptive management approaches be developed and implemented?

Lessons Learned from the Case Studies
This workshop looked at two case examples, using presentations and discus-
sion, to explore current understanding of multiple environmental stresses in the
earth system and to discuss the types of research needed to improve integrated
understanding of these kinds of complex, nonlinear problems. Understanding
multiple stresses is challenging because it almost always requires consideration
of multiple variables and larger, more complex spatial scales. Yet without a
more sophisticated understanding of the impacts of a suite of environmental
stresses, we cannot make the kind of progress necessary to improve our predictive
capability and response strategies.
The overarching lesson of the workshop discussions is that a thorough
understanding of the integrated effects of—and future vulnerability to—mul-
tiple stresses to natural and socioeconomic systems requires improved use of
existing tools and strategies and, in addition, the development of improved tools
and strategies—such as observational, modeling, and information systems infra-
structure—to support environmental monitoring, vulnerability assessment, and
response analysis and that the entire process needs significant involvement of
stakeholders.
During the workshop, the National Ecological Observing Network (NEON)
was mentioned as an example of the type of nationally networked research, com-
munication, and informatics infrastructure needed to provide more comprehen-
sive and interdisciplinary measurements and experiments. References were also
made to other possible infrastructure, such as
http://www.neoninc.org/.
36

LESSONS LEARNED FROM THE CASE STUDIES 37
• the National Phenology Network (USA-NPN), designed to facilitate

systematic collection and free dissemination of phenological data from across the
United States to support research concerning interactions among plants, animals,
and the lower atmosphere, especially the long-term impacts of climate change;
• the Ocean Research Interactive Observatory Networks (ORION), designed
to provide high-frequency, continuous time-series measurements in broad-scale
spatial arrays needed to define the links among physical, biological, chemical,
and geological variables in the oceans and provide spatially coherent data to study
processes and enable modeling efforts;
• the Ameri-Flux Network, designed to provide continuous observations
of ecosystem-level exchanges of CO2, H2O, energy, and momentum spanning
diurnal, synoptic, seasonal, and interannual timescales;
• the proposed Integrated and Sustained Ocean Observing System (IOOS);
• the International Global Ocean Observing System (GOOS); and
• the Global Earth Observation System of Systems (GEOSS).
Observing systems alone cannot solve the puzzle of understanding multi-

stress environmental problems, but they are a necessary component because they
provide the data needed to characterize the environment and determine trends.
There is a real need for careful attention to the systematic creation and valida-
tion of long-term, consistent climate data records (NRC, 2004a). The following
paragraphs describe some of the other tools and strategies highlighted during the
workshop.
Comprehensive Regional Frameworks

Many participants advocated development of comprehensive regional frame-
works for environmental studies as outlined during the workshop’s keynote
address by Dr. Eric Barron. The vision Dr. Barron proposed included
1. an integrated regional web of sensors that link existing observations into

a coherent framework and enable new observations to be developed within an
overall structure;
2. an integrated and comprehensive regional information system accessible
to a wide variety of researchers, operational systems, and stakeholders;
3. directed process studies designed to examine specific phenomena through
field study to address deficiencies in understanding;
http://www.uwm.edu/Dept/Geography/npn/.
http://www.orionprogram.org/.
http://public.ornl.gov/ameriflux/.
http://www.ocean.us/.
http://www.ioc-goos.org/.
http://www.earthobservations.org/.

4. a regional, high-resolution modeling foundation for constructing increas-

ingly complex coupled system models at the spatial and temporal scales appropri-
ate for the examination of specific and integrated biological, hydrological, and
socioeconomic systems; and
5. a strong connection to significant regional issues and stakeholders.
Workshop participants returned repeatedly to the Regional Integrated Sciences

and Assessments (RISA) program as a possible model for such regional frame-
works. (RISAs are discussed in greater detail later in this chapter.)
A Web of Integrated Sensors

Current U.S. observation strategy tends to focus on the measurement of
discrete variables at a specific set of locations designed to serve the different
needs of weather forecasting, pollution monitoring, hydrological forecasting, or
other specific mission objectives. This mission focus results in a diverse set of
networks supported by a number of different federal agencies, states, or regional
governments. A host of environmental issues motivates additional new observa-
tions. However, these new observations are frequently viewed independently of
an integrated observing strategy. In addition, it is difficult to identify sufficient
financial support for regular and consistently repeated observations. These factors
severely limit our ability to integrate physical, biological, chemical, and human
systems.
Creating comprehensive regional observing networks would allow us to
(a) link observing systems into a web of integrated sensors building upon exist-
ing weather and hydrological stations and remote sensing capability; (b) create
the agreements across a set of more limited agencies and federal, state, and local
governments needed to create a structure to the observing system; (c) provide a
compelling framework that encourages or demands the integration of new obser-
vations into a broader strategy; and (d) create strong linkages between research
and operational observations that result in mutual benefit. Clearly, these steps
will be difficult to achieve given political dynamics and constrained budgets, but
these sorts of comprehensive approaches are needed to improve our capability
to respond and ensure flexibility over time. The importance of continued support
for and maintenance of existing environmental networks should not be under
estimated as a foundation for what needs to happen in the future.
Regional Information Systems

Efforts to create comprehensive information systems increasingly reflect
federal and state mandates to make data more accessible and useful to the public
http://www.climate.noaa.gov/cpo_pa/risa/.

and to ensure that our investments in research yield maximum societal benefit.
Development of a tractable regional digital database is feasible and enables par-
ticipation of universities; federal, state, and local governments; and industry in
an endeavor for which immediate benefit for a state or region can be evident. For
this to happen efficiently these information systems will need to take advantage
of existing facilities and infrastructure, augmenting selectively as needed and,
overall, improving the connections that bind the pieces together. This will take
careful and detailed planning and a strong commitment to implementation.
Framework for Process Studies

Process studies are a critical element of scientific advancement because they
are designed to probe uncertainties in knowledge about how the earth system
functions. The objective is to use field study to address deficiencies in our under-
standing. Targeted process studies improve our ability to quantify thresholds,
understand nonlinear interactions of multiple environmental factors, and decipher
long timescale responses to climate change. The benefit of these intensive studies
is maximized when they can be coupled with a highly developed, integrated set
of sensors, with readily accessible spatial and temporal data within a regional
information system, and with a predictive model framework that readily enables
the entrainment and testing of new information from process studies.
Improving Our Predictive Capability

Prediction is central to the translation of knowledge about the earth system
into economic benefit and societal well-being. Although there is still substan-
tial room for improvement, over the last several decades we have experienced
enormous increases in our ability to forecast weather and to project climate and
climate variability into the future. The demand for new forecasting products
involving air quality, energy demand, water quality and quantity, ultraviolet
radiation, and human health indexes is growing rapidly. Environmental issues
will demand an even greater capability to integrate physical, biological, chemical,
and human systems in order to examine the response of critical regions or cases
to multiple stresses.
Global weather and climate models provide the strongest physical foundation
for more comprehensive predictive capability. The numerical models that under-
pin this type of forecast are increasingly becoming the framework for the addition
of new numerical formulations designed to predict air quality, the water balance
for river forecast models, and a host of other variables, including the migration of
forests under climate change conditions. As we attempt to produce predictions at
the scale of human endeavors, mesoscale models (capable of predicting synoptic
weather systems) and downscaling of coarser resolution model output are increas-
ingly becoming the focal point of weather and climate studies because of their

BOX 4-1
Examples of Tools and Strategies
After both cases were presented and discussed, workshop participants took
part in brainstorming to identify examples of important observations, modeling
tools, and research strategies for improving understanding of and response to
multiple-stresses problems. This box captures those ideas to illustrate the range
of possible actions; they are listed as presented and not prioritized or standardized
in style or scope.
Observations
• Maintain existing environmental networks and expand observing infrastructure

(e.g., NEON) while maintaining a balance between high-cost, large platforms/
instruments and low-cost, small platforms/instruments.
• Expand aircraft fleet to include smaller/lower cost aircraft. Consider using
remotely piloted vehicles.
• Expand use of observing systems (flux towers and aircraft transects) for measur
ing net ecosystem exchange and developing/testing models.
• Expand use of remote sensing (satellite or aircraft) to access the increasing
suite of ecological measurements and molecular biology tools.
• Focus on intersections and gradients.
• Expand field manipulations to develop process understanding; generate a plan
for manipulation study priorities.
• Generate the capability to manipulate multiple variables and conduct experi-
ments over appropriate spatial and temporal scales.
• Ensure consistency of long-term measurements because long-term data are
needed to identify nonlinearities.
• Generate integrated, accessible data archive/information systems.
• Generate global databases of multiple stresses (georeferenced) for large-scale
models; include a database of spatial economic data.
Models
• Integrated hierarchy of models.

• Models that integrate human and natural components to study nonlinearities
and thresholds.
• Extend model focus to effects on ecosystem trophic levels.
• Use ensemble runs to assess uncertainty and probability.
• Use inverse modeling to test end products of complex models.
• Station-specific data projection model based on historical information with the
ability to perturb the model using appropriate climate forcing functions.
• Statistical tools that clearly display the interaction between data and forecast
procedures and forecast validation.
• Develop model scenarios that provide better understanding of catastrophic
events (e.g., war, pests, disease).
continued

BOX 4-1 Continued
Strategies
• Integrate datasets and provide information delivery system to users.

• Map multiple stresses using multiple overlays in geographic information systems
(GISs) to identify “critical zones.”
• Create reliable indexes of leading vulnerability indicators that combine suites
of stressors that are known to interact and might give a first-order monitoring
capability (i.e., a Dow Jones-type index for the environment as a publicity tool).
• Use video/computer games as a multiple-stresses decision-support tool (e.g.,
use place-based scenarios to explore megadrought thresholds, drought scenarios,
ocean or atmosphere scenarios).
• Make better use of existing data, such as by linking qualitative and quantita-
tive information in impact assessments, expanding use of spatial representation
(GIS) to visualize data, and improving decision analysis and understanding of
uncertainties.
• Co-develop scenarios with those who use the information for decision making,
working together in effective partnerships.
• Work with stakeholders at the design point so that research meets user
needs.
• Work to better communicate risks.
• Use seemingly simple systems that exhibit complicated behavior that could be
used by managers to illustrate probabilities (e.g., flood frequency analysis).
• Develop better models for drought planning, improve tools for municipalities,
and generate methods for testing drought plans.
• Identify implications of changes in states or processes.
• Survey to identify ecosystem shifts/state changes/changing carbon or moisture
levels (look to see if they do go through thresholds).
• Prioritize wetlands protection.
• Use regional monies to leverage efforts.
potential to make predictions on the scale of river systems, cities, agriculture, and
forestry. Development of a mesoscale numerical prediction capability that meshes
with regional sensor webs and information systems would facilitate development
of tractable coupled models, initiate experimental forecasts of new variables, and
enable assessment of the outcomes associated with multiple stresses.
RESEARCH NEEDS RELATED TO

NONLINEARITIES AND THRESHOLDS
Workshop participants frequently highlighted lessons learned about non
linearities in the climate system and the difficulties associated with quantifying

the effects of multiple interacting stresses and technological change. Although

current models are useful in identifying single-variable nonlinearities, few models
are sufficiently comprehensive to provide quantitative predictions of the effects
of multiple environmental stresses. While coupled modeling systems developed
in the future are expected to be useful in the identification of nonlinearities,
it was thought that nonlinearities are currently best identified by long-term
observations.
The thresholds considered by participants to have the highest priority for
study include climate/pest interactions resulting in changes in functional types of
natural vegetation; megadrought (climate threshold, ecosystem thresholds, human
thresholds, cascading effects); and interaction between ecosystems, climate
change, and air pollution. Suggestions regarding how best to approach the study
of thresholds included studies involving initial system observations followed by
single-variable and multiple environmental stresses experiments and modeling;
studies focused on ecotones, zones where marginality of nutrients, predators,
climate, land use, economics, and policies create unstable land uses that are espe-
cially sensitive to small variations in drivers; and studies of extreme conditions
(e.g., air pollution in megacities) where changes in state may be observed. Partici-
pants also encouraged development of threshold typology, identification of a core
set of controlling (and dynamic) variables that determine system behavior, assess-
ment of the interaction of fast and slow variables (as related to the threshold);
assessment of the degree to which a system may be capable of self-organization;
and assessment of the ability to build and increase the capacity for learning and
adaptation. Threshold-focused research needs to study both direct and indirect
effects, linking thresholds and impact occurrence to indicators/indices, study of
the full probability space of observations and model outputs, and consideration
of new opportunities that are likely to result from globalization.
RESEARCH NEEDS RELATED TO INCREASING RESILIENCE

Workshop participants encouraged the following approaches to increasing
resilience to multiple environmental stresses:
• use of models;
• improvement of models for response planning;
• identification of additional water storage;
• consideration of new conservation strategies;
• maintenance of biodiversity;
• improved communication of environmental capacities and limitations to
local officials;
• improved understanding of adaptive or buffering capacity, which is deter-
mined by the types of capital available (natural, social, human, cultural, and
produced);

• full leveraging of adaptive capacity, technological capacity, and expertise;

• in-parallel focus on combinations of social and environmental stresses and
combinations of environmental stresses;
• use of war games as a scenario-based tool for informing decision making,
examination of historical analogs that share important similarities with contem-
porary (and anticipated future) stresses;
• incorporation of “normal” (rather than fair to good) socioeconomic condi-
tions and civil and regional wars in scenarios for sustainable development; and
• linking results of models to resilience, improvement of public awareness
of related issues, and elimination of nonsustaining financial incentives.
Moreover, a number of steps were suggested for the creation of vigorous and
continuous links between researchers and decision makers, including
• incorporation of the variety of time and space scales and the diversity of
variables that are important to decision makers;
• emphasis on the education of the user in the meaning and significance of
climate and land use information in order to promote greater use and more robust
applications;
• ensuring mutual information exchange and feedback;
• focus on communication and accessibility of information;
• continuous evaluation and assessment of the use and effectiveness of the
services;
• employment of active mechanisms to enable the transition from research
discovery to useful products; and
• employment of a variety of methods of education and outreach.
RESEARCH NEEDS RELATED TO REGIONAL STUDIES

During the workshop the argument was made that the ability to anticipate
the future is what makes knowledge powerful. The knowledge we seek concerns
the role and effects of multiple stresses in the context of atmosphere-ecosystem
interactions. These interactions include climate variability and change over a wide
range of time/space scales, land use/land cover changes, human social systems,
waste products and streams, and the combined effects of all the above on natural
ecosystems. This knowledge must perforce be place based (i.e., site or region
specific) because context is important. Integrated assessments of multiple stresses
across a variety of time/space scales are required in which the impacts and deci-
sions are place based but the drivers of impacts are drawn from a much larger
scale. We can summarize by saying it is critical to link the large-scale drivers to
place-based contexts with a focus on multiple stressors and put the knowledge to
work—that is, connect in partnerships with real stakeholders and decision makers
from the place where the work is done.

Obstacles to realizing the vision above include lack of the following:
• integrated observing systems;

• a common modeling framework;
• a foundation of process studies geared toward prediction;
• an integrated data and information system; and
• systematic, vigorous connections to real users of the information and
decision makers.
One program in existence comes close to meeting the challenge described

above and that is the Regional Integrated Sciences and Assessments (RISA)
program of NOAA’s Office of Global Programs. As described at the workshop,
the RISA program was launched in 1995 with a pilot project at the University
of Washington (the Climate Impacts Group). The program now consists of eight
regional teams distributed across the United States, each with a focus on the role
of climate variability and the projected impacts of climate change in defined
sectors of human socioeconomic activity and on specified ecosystems. Each
program is required to establish links and partnerships with stakeholders and
decision makers so that research results can be translated into usable knowledge
and decision-support tools that are specific to the subregion. Emphasis is placed
not only on assessing the climate sensitivity of different activities and ecosystems
but also on their vulnerabilities to climate variability and change and on policies
and programs that would increase the resilience of these subregions to climate-
related risks of varying magnitude.
So far RISA teams cover the Pacific Northwest, the Southwest, the Colorado
River system, California, Hawaii, the Southeast (Florida and Georgia), the
Carolinas, and New England. Although the basic template and objectives for each
team are the same, there is considerable variation in the way the teams imple-
ment the vision because each team is grounded in a particular place in which the
mix of concerns varies along with constituents, capabilities, and climate-related
risks. These teams document what and how climate drivers function in specific
places, what impacts they typically exert on various kinds of natural systems and
socioeconomic activities that are sensitive to climate variability, and what levels
of risk each subregion faces, inter alia. The crucial questions, not surprisingly,
shift from place to place. So, for instance, “Will winter snowpack and spring
streamflow be above or below normal this year?” might be a critical question
in the Pacific Northwest, but it will have no meaning in Florida where “Will it
freeze?” is definitely one of the critical questions. In the western United States
water is the central issue and will be even more so under scenarios of climate
change because the entire West consists largely of snowmelt-driven systems. No
matter what their foci, all subregions are now faced with the necessity of trying to
understand what their vulnerabilities to anthropogenic climate change are, what
the magnitudes and rates of change might be, and how best to adapt to and cope
with these changes over time.

The RISA teams clearly demonstrate how useful such an approach can be.
But to date this is still a small program and a long way from fulfilling the vision
of cohesive observations, data management, data access, carefully designed pro-
cess studies across regions and subregions nested in a framework for developing
regional predictive models of the effects of multiple stresses and translating
the research outputs in a series of vigorous and continuing connections with
stakeholders.
NEAR-TERM OPPORTUNITIES
Looking overall at the workshop presentations and discussions, a great range
of issues and opportunities were explored. As a final step, the steering committee
reviewed the information generated and identified seven near-term opportunities
for advancing our understanding of multiple stresses and making this understand-
ing useful to decision makers.
1. A Ground-Based Measurements Network. There is a real need for com-

prehensive ground-based measurements of important ecological indicators such
as biodiversity, species composition, ecosystem functioning, ecological aspects of
biogeochemical cycles, and other elements. This information over time will allow
improved understanding of the ecological implications of climate change, the
evolution of infectious diseases, invasive species, and land use change over time
and across large spatial scales (NRC, 2003). The National Ecological Observatory
Network (NEON) that has been under development is an example of the kind of
system that could contribute the types of information needed.
2. Global Information Systems and Satellite Observations. In 2005 mem-
bers of the Group on Earth Observations (GEO), which includes 60 countries and
the European Commission, agreed to a 10-year implementation plan for a Global
Earth Observation System of Systems (GEOSS). Anticipated foci and socioeco-
nomic benefits include
• sustainable agriculture and reduced desertification;

• disaster reduction and improved ecosystem management and protection;
• biodiversity conservation;
• improved weather information, forecasting, and warning;
• adaptation to climate variability and change;
• improved water resource management;
• understanding of environmental factors affecting human health and well-
being; and
• improved management of energy resources (NRC, 2005b).
GEOSS could be configured to assist with detecting and understanding mul-

tiple stresses and extreme events. For example, one of the key components of the

International Earth Observing System (IEOS), the primary contribution by the

United States to GEOSS, would be a National Integrated Drought Information
System (NIDIS). The goal of NIDIS is to develop an integrated drought informa-
tion system that would enhance the ability of users to assess on a timely basis
the risks and potential impacts associated with drought through the availability of
appropriate decision-support tools. Other aspects of NIDIS focus on the develop-
ment of a comprehensive drought early warning and delivery system, an enhanced
research environment that emphasizes impact mitigation and improved predictive
capabilities, and a framework for interacting with and educating stakeholders
and the public on causes of drought, preparedness strategies, and how drought
impacts human and natural systems. NIDIS is considered to be an invaluable
resource in helping water managers and policy makers at all levels deal with
the increasing impacts of drought and water resources management in a climate
change environment.
IEOS/GEOSS could also provide longer-term forecasting, especially for
severe weather events, such as Hurricane Katrina, and an all-hazards, all-media
alerting system. (In the case of wildfires, hikers, for example, would get immedi-
ate messages on their cell phones to evacuate areas.) Furthermore, as improved
forecasting will help with the distribution of resources in warm or cold years
and in extreme wet or dry seasons, IEOS/GEOSS could be very important to the
water and energy sectors.
3. Synthesis of Data. The Heinz Center’s10 State of the Nation’s Ecosystems
project—done in concert with federal, corporate, and academic partners—is char-
acterizing data gaps and data integration needs by sector (urban, forests, coasts,
etc.) in order to produce indicators on the improvement or degradation of U.S.
resources. This multiyear effort could include a section on the composite effect of
multiple stresses on ecological and urban sectors and identify missing data most
needed to characterize the status of trends of ecosystem health.
4. Nonlinearities and Thresholds. There is a rich historical record of
responses to extremes of record (droughts, floods, hurricanes). Overlaying those
conditions on the socioeconomic and ecological conditions of today—conduct-
ing “what if” scenarios to see if today’s communities and natural resources could
cope with, for example, the drought of the 1930s or a direct hit of hurricane
Andrew on Miami—would be extremely valuable. Similarly, scenarios recon-
structing extreme events of the past could study if an increase in temperature
and a change in water availability would lead to breakpoints or thresholds in the
ecological or economic realms (e.g., as observed by Breshears et al., 2005).
IEOS, a global system of missions made up of EOS (Earth Observing System) satellites together
with other Earth observation missions from NOAA (National Oceanic and Atmospheric Administra-
tion), Europe, and Japan.
10http://www.heinzctr.org.

5. Resilience. Within the assessment processes, institutional, technological,

and economic options that offer insurance or appear the most robust to a suite of
environmental changes could be identified. These could include such measures
as changed cropping patterns, water conservation, germplasm preservation, park
design, and habitat connectivity, within the context of long-term resiliency to a
changing climate. In addition, early warning systems for various environmental
indicators (droughts, floods, tropical cyclones, wildfires) could be established in
pertinent regions. Finally, development of a compendium of best practices for
coping with extreme events and deployment of appropriate preparedness pro-
grams would enhance resilience.
6. Regional Foci. Understanding the impacts of climate change in a particu-
lar place in concert with the other environmental stressors operating in that region
is key to developing wise coping options. The Regional Integrated Sciences and
Assessments program and the regional studies begun under the U.S. National
Assessment process are models of this nascent type of analysis, and an increase
in this kind of activity is greatly needed.
7. Stakeholder Involvement. Connecting stakeholders to an ongoing
research effort directly aimed at producing usable knowledge of value to stake-
holders requires long-term partnerships, trust that researchers will actually stay
the course, thorough familiarity on both sides about what each is doing, consider-
able effort expended by the research teams to gain knowledge about the decision
context and the needs of the different types of stakeholders, and appreciation by
the stakeholders of the added value the results of the research can offer to their
concerns. All of this takes time and resources. The RISA teams, for example,
have used periodic systematic surveys, annual workshops custom-tailored to the
specific interests of different combinations of stakeholders (e.g., water resource
managers, forest fire managers, fisheries managers, farmers, coastal managers),
and the co-production of specific decision-support tools as ways to build in true
stakeholder involvement.
Research and experience to date show that overemphasizing climate forecasts
per se is counterproductive. Users have a decided preference for deterministic
forecasts and lack of understanding of probabilistic forecasts to an extent that
only the technically advanced early adopters find probabilistic climate forecasts
to be useful. For others a softer approach is more useful and more readily under-
stood. This approach is grounded on the fact that all stakeholders really want to
understand to what extent climate is responsible for the underlying variation in
the resources they use or manage or the economic activities in which they are
engaged. Once researchers recognize this fact, it is possible to have fruitful, long-
term relationships that evolve. However, each party to the relationship has to be
committed to learning from the other, and the researchers need to strive to pro-
duce information and decision tools that are useful and timely to the stakeholders.
However, it should be understood that stakeholders cannot define the totality of
the research agenda for the simple reason that often the stakeholders do not and

cannot be expected to know what they need to know about the dynamics of the
climate system. So the research agenda must be balanced; it cannot be the product
of curiosity alone but rather it must be defined to meet certain ends that can be
transferred to the decision maker.

References
Allen, C. D., and D. D. Breshears. 1998. Drought-induced shift of a forest-woodland ecotone: Rapid
landscape response to climate variation. Proceedings of the National Academy of Sciences
U.S.A. 95:14839-14842.
Andreae, M. O., D. Rosenfeld, P. Artaxo, A. A. Costa, G. P. Frank, K. M. Longo, and M. A. F. Silva-
Dias. 2004. Smoking rain clouds over the Amazon. Science 303:1337-1342.
Arrhenius, S. 1896. On the influence of carbonic acid in the air upon the temperature of the ground. The
London, Edinburgh and Dublin Philosophical Magazine and Journal of Sciences 41:237‑276.
Bacastow, R., and C. D. Keeling. 1973. Atmospheric carbon dioxide and radiocarbon in the natural
carbon cycle. II. Changes from A.D. 1700 to 2070 as deduced from a geochemical reservoir.
Pp. 86-135 in G. M. Woodwell and E. V. Pecan, eds. Carbon and the Biosphere. Springfield,
Virginia: U.S. Department of Commerce.
Botterill, L. 2005. Late twentieth century approaches to living with uncertainty: The national drought
policy. In From Disaster Response to Risk Management: Australia’s National Drought Policy, L.
C. Botterill and D. A. Wilhite, eds. Dordrecht, The Netherlands: Springer Publishers.
Breshears, D. D., N. S. Cobb, P. M. Rich, K. P. Pricee, C. D. Allen, R. G. Balice, W. H. Romme,
J. H. Kastensf, M. L. Floyd, J. Belnapl, J. J. Anderson, O. B. Myers, and C. W. Meyer. 2005.
Regional vegetation die-off in response to global-change-type drought. Proceedings of the
National Academy of Sciences U.S.A. 102:15144-15148.
Cayan, D. R., S. A. Kamerdiener, M. D. Dettinger, J. M. Caprio, and D. H. Peterson. 2001. Changes
in the onset of spring in the western United States. Bulletin of the American Meteorological
Society 82:399-415.
Chapin, F. S., M. Sturm, M. C. Serreze, J. P. McFadden, J. R. Key, A. H. Lloyd, A. D. McGuire, T. S.
Rupp, A. H. Lynch, J. P. Schimel, J. Beringer, W. L. Chapman, H. E. Epstein, E. S. Euskirchen,
L. D. Hinzman, G. Jia, C. L. Ping, K. D. Tape, C. D. C. Thompson, D. A. Walker, and J. M.
Welker. 2005. Role of land-surface changes in Arctic summer warming. Science 310:657-660.
Charney, J., P. H. Stone, and W. J. Quirk. 1975. Drought in the Sahara: A biogeophysical feedback.
Science 187:434-435.
Charney, J. G., W. J. Quirk, S-H. Chow, and J. Kornfield. 1977. A comparative study of effects of albedo
change on drought in semiarid regions. Journal of the Atmospheric Sciences 34:1366‑1385.
49

50 REFERENCES
Clark, J. S., E. C. Grimm, J. J. Donovan, S. C. Fritz, D. R. Engstrom, and J. E. Almendinger. 2002.

Drought cycles and landscape responses to past aridity on prairies of the northern Great Plains,
USA. Ecology 83(3):595-601.
Cook, E. R., C. A. Woodhouse, C. M. Eakin, D. M. Meko, and D. W. Stahle. 2004. Long-term aridity
changes in the western United States. Science 306:1015-1018.
Cox, P. M., R. A. Betts, C. D. Jones, S. A. Spall, and I. J. Totterdell. 2000. Acceleration of global
warming due to carbon-cycle feedbacks in a coupled climate model. Nature 408:184-187.
Dickinson, R. E., and A. Henderson-Sellers. 1988. Modelling tropical deforestation: Study of
GCM land-surface parameterizations. Quarterly Journal of the Royal Meteorological Society
114:439‑462.
Enfield, D. B., and L. Cid-Serrano. Forthcoming. Projecting the risk of future climate shifts. Inter-
national Journal of Climate.
Feddema, J. J., K. W. Oleson, G. B. Bonan, L. O. Mearns, L. E. Buja, G. A. Meehl, and W. M.
Washington. 2005. The importance of land-cover change in simulating future climates. Science
310:1674-1678.
FEMA (Federal Emergency Management Agency). 1995. National Mitigation Strategy. Washington,
D.C.: FEMA.
Friedlingstein, P. 2004. Climate carbon cycle interactions. Pp. 217-224 in The Global Carbon Cycle:
Integrating Humans, Climate, and the Natural World, C. B. Field and M. R. Raupach, eds.
Washington, D.C.: Island Press.
Fung, I., S. C. Doney, K. Lindsay, and J. John. 2005. Evolution of carbon sinks in a changing climate.
Proceedings of the National Academy of Sciences U.S.A. 102:11201-11206.
Gibbard, S., K. Caldeira, G. Bala, T. J. Phillips, and M. Wickett. 2005. Climate effects of global land
cover change. Geophysical Research Letters 32:L23705.
Gray, S. T., C. L. Fastie, S. T. Jackson, and J. L. Betancourt. 2004. Tree-ring-based reconstruc-
tion of precipitation in the Bighorn Basin, Wyoming, since 1260 A.D. Journal of Climate
17(19):3855‑3865.
Hagman, G. 1984. Prevention Better than Cure: Report on Human and Natural Disasters in the Third
World. Stockholm: Swedish Red Cross.
Hamlet, A. F., P. W. Mote, M. P. Clark, and D. P. Lettenmaier. 2005. Effects of temperature and
precipitation variability on snowpack trends in the western United States. Journal of Climate
18(21):4545-4561.
Hayes, M. J., M. D. Svoboda, C. L. Knutson, and D. A. Wilhite. 2004a. Estimating the economic
impacts of drought. Proceedings of the 14th Conference on Applied Climatology, Seattle,
Washington, January 11-15.
Hayes, M. J., O. V. Wilhelmi, and C. L. Knutson. 2004b. Reducing drought risk: Bridging theory and
practice. Natural Hazards Review 5:106-113.
Hoerling, M., and A. Kumar. 2003. The perfect ocean for drought. Science 299:691-694.
IPCC (Intergovernmental Panel on Climate Change). 1990. First Assessment Report. Cambridge,
England: Cambridge University Press.
IPCC. 1995. Second Assessment Report: Climate Change. Cambridge, England: Cambridge Univer-
sity Press.
IPCC. 2001. Third Assessment Report: Climate Change. Cambridge, England: Cambridge University
Press.
ISDR (International Strategy for Disaster Reduction) Ad Hoc Discussion Group on Drought. 2003.
Living with Risk: An Integrated Approach to Reducing Society’s Vulnerability to Drought.
Geneva, Switzerland: ISDR.
Jackson, R. B., E. G. Jobbagy, R. Avissar, S. B. Roy, D. J. Barrett, C. W. Cook, K. A. Farley, D. C. le
Maitre, B. A. McCarl, and B. C. Murray. 2005. Trading water for carbon with biological carbon
sequestration. Science 310:1944-1947.

REFERENCES 51
Karnosky, D. F., B. Mankovska, K. Percy, R. E. Dickson, G. K. Podila, J. Sober, A. Noormets, G.

Hendrey, M. D. Coleman, M. Kubiske, K. S. Pregitzer, and J. G. Isebrands. 1999. Effects of
tropospheric O3 on trembling aspen and interaction with CO2: Results from an O3-gradient and
a FACE experiment. Journal of Water, Air and Soil Pollution 116:311-322.
Knowles, N., M. D. Dettinger, and D. R. Cayan. 2006. Trends in snowfall versus rainfall in the west-
ern United States. Journal of Climate 19(18):4545-4559.
Knutson, C., M. J. Hayes, and T. Phillips. 1998. How to Reduce Drought Risk. Lincoln, Nebraska:
National Drought Mitigation Center.
Lean, J., and D. A. Warrilow. 1989. Simulation of the regional climatic impact of Amazon deforesta-
tion. Nature 342:411-413.
Mack, M. C., E. A. G. Schuur, M. S. Bret-Harte, G. R. Shaver, and F. S. Chapin. 2004. Ecosystem car-
bon storage in arctic tundra reduced by long-term nutrient fertilization. Nature 431:440-443.
McCabe, G. J., M. A. Palecki, and J. L. Betancourt. 2004. Pacific and Atlantic Ocean influences on
multidecadal drought frequency in the United States. Proceedings of the National Academy of
Sciences U.S.A. 101:4136-4141.
MEA (Millennium Ecosystem Assessment). 2005. Strengthening Capacity to Manage Ecosystems
Sustainability for Human Well-Being. Washington, D.C.: Island Press.
NAST (National Assessment Synthesis Team). 2000. Climate Change Impacts on the United States:
The Potential Consequences of Climate Variability and Change. U.S. Global Change Research
Program. Cambridge, England: Cambridge University Press.
Nepstad, D. C., A. Veríssimo, A. Alencar, C. Nobre, E. Lima, P. Lefebvre, P. Schlesinger, C. Potter,
P. Moutinho, E. Mendoza, M. Cochrane, and V. Brooks. 1999. Large-scale impoverishment of
Amazonian forests by logging and fire. Nature 398:505-508.
NRC (National Research Council). 1999. Our Common Journey: A Transition Toward Sustainability.
Washington, D.C.: National Academy Press.
NRC. 2002. Science for the Sustainability Transition. Washington, D.C.: The National Academies
Press.
NRC. 2003. Communicating Uncertainties in Weather and Climate Information: A Workshop
Summary. Washington, D.C.: The National Academies Press.
NRC. 2004a. Climate Data Records from Environmental Satellites. Washington, D.C.: The National
Academies Press.
NRC. 2004b. NEON: Addressing the Nation’s Environmental Challenges. Washington, D.C.: The
National Academies Press.
NRC. 2005a. Earth Science and Applications from Space: Urgent Needs and Opportunities to Serve
the Nation. Washington, D.C.: The National Academies Press.
NRC. 2005b. Improving the Scientific Foundation for Atmosphere-Land-Ocean Simulations: Report
of a Workshop. Washington, D.C.: The National Academies Press.
O’Meagher, B., M. Stafford Smith, and D. H. White. 2000. Approaches to integrated drought risk
management. Pp. 115-128 in Drought: A Global Assessment (Volume 2), D. A. Wilhite, ed.
London, U.K.: Routledge Publishers.
Pacala, S. W., G. C. Hurtt, D. Baker, P. Peylin, R. A. Houghton, R. A. Birdsey, L. Heath, E. T.
Sundquist, R. F. Stallard, P. Ciais, P. Moorcroft, J. P. Caspersen, E. Shevliakova, B. Moore, G.
Kohlmaier, E. Holland, M. Gloor, M. E. Harmon, S. M. Fan, J. L. Sarmiento, C. L. Goodale,
D. Schimel, and C. B. Field. 2001. Consistent land- and atmosphere-based U.S. carbon sink
estimates. Science 292:2316-2319.
Prentice, I. C., G. D. Farquhar, M. J. R. Fasham, M. L. Goulden, M. Heimann, V. J. Jaramillo, H.
S. Kheshgi, C. Le Quéré, R. J. Scholes, and D. W. R. Wallace. 2001. The carbon cycle and
atmospheric carbon dioxide. Pp. 183-237 in Climate Change 2001: The Scientific Basis. Con-
tribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel
on Climate Change, J. T. Houghton, Y. Ding, D. J. Griggs, M. Noguer, P. J. van der Linden,
X. Dai, K. Maskell, and C. A. Johnson, eds. Cambridge, United Kingdom, and New York, NY,
USA: Cambridge University Press.

52 REFERENCES
Rial, J. A., R. A. Pielke Sr., M. Beniston, M. Claussen, J. Canadell, P. Cox, H. Held, N. de Noblet-
Ducoudre, R. Princc, J. F. Reynolds, and J. D. Salas. 2004. Nonlinearities, feedbacks and critical
thresholds within the earth’s climate system. Climatic Change 65:11-38.
Sabine, C. L., M. Heiman, P. Artaxo, D. C. E. Bakker, C-T. A. Chen, C. B. Field, N. Gruber, C.
LeQuéré, R. G. Prinn, J. E. Richey, P. Romero-Lankao, J. A. Sathaye, and R. Valentini. 2004.
Current status and past trends of the carbon cycle. Pp. 17-44 in The Global Carbon Cycle:
Integrating Humans, Climate, and the Natural World. C. B. Field and M. R. Raupach, eds.
Salati, E., and P. B. Vose. 1984. Amazon Basin: A system in equilibrium. Science 225:129.
Shaver, G. R., A. E. Giblin, K. J. Nadelhoffer, K. K. Thieler, M. R. Downs, J. A. Laundre, and E.
B. Rastetter. 2006. Carbon turnover in Alaskan tundra soils: Effects of organic matter quality,
temperature, moisture and fertilizer. Journal of Ecology 94:740-753.
Shukla, J., C. Nobre, and P. Sellers. 1990. Amazon deforestation and climate change. Science
247:1322-1325.
Stewart, I. T., D. R. Cayan, and M. D. Dettinger. 2005. Changes toward earlier streamflow timing
across western North America. Journal of Climate 18:1136-1155.
Strauss, D., L. Bednar, and R. Mees. 1989. Do one percent of forest fires cause ninety-nine percent
of the damage? Forest Science 35:319-328.
Vitousek, P. M., J. Aber, R. W. Howarth, G. E. Likens, P. A. Matson, D. W. Schindler, W. H. Slesinger,
and G. D. Tilman. 1997. Human alteration of the global nitrogen cycle: Causes and conse-
quences. Issues in Ecology, No. 1, Spring.
Western Governors’ Association. 2004. Creating a Drought Early Warning System for the 21st
Century: The National Integrated Drought Information System. Denver, Colorado.
Wilhite, D. A. 1991. Drought planning: A process for state government. Water Resources Bulletin
27(1):29-38.
Wilhite, D. A., and M. Buchanan-Smith. 2005. Drought as hazard: Understanding the natural and
social context. Pp. 3-32 in Drought and Water Crises: Science, Technology, and Management
Issues, D. A. Wilhite, ed. Boca Raton, Florida: CRC Press.
WMO/UNEP (World Meteorological Organization/United Nations Environment Programme). 1985.
Atmospheric Ozone: Assessment of Our Understanding of the Processes Controlling its Present
Distribution and Change: 1985. Report No. 16, 3 volumes. Geneva, Switzerland: WMO.
WMO. 1989. Scientific Assessment of Stratospheric Ozone: 1989. Global Ozone Research and Moni-
toring Project Report No. 20. Geneva, Switzerland: WMO.
WMO. 1991. Scientific Assessment of Ozone Depletion: 1991. Global Ozone Research and Monitor-
ing Project, Report No. 25. Geneva, Switzerland: WMO.
WMO. 1994. Scientific Assessment of Ozone Depletion: 1994. Geneva, Switzerland.
Woodhouse, C. A., and J. T. Overpeck. 1998. 2000 years of drought variability in the central United
States. Bulletin of the American Meteorological Society 79:2693-2714.
Woodwell, G. M., J. E. Hobbie, R. A. Houghton, J. M. Melillo, and B. Moore. 1983. Global deforesta-
tion: Contribution to atmospheric carbon dioxide. Science 222:1081-1086.

Appendix A
Statement of Task
This workshop will use key presentations, case examples, panelist discussion,
and open discussion to explore current understanding of multiple environmental
stresses in the earth-atmosphere system on natural, managed, and socioeconomic
systems, and discuss the types of research needed to improve integrated under-
standing of these kinds of complex, nonlinear problems.
Workshop presenters and participants are asked to discuss the following
questions:
• For the case examples and in general, what is the state of the knowledge
base related to the coupling between physical, chemical, ecological, and human
systems? What research has been done or is being done in the area of multiple
stresses?
• For the case examples and in general, how do multiple stresses interact
on different temporal and spatial scales? What does this imply for management
options?
• For the case examples and in general, what are the potential nonlinearities
in response to multiple stresses?
• How can we improve our capability of integrating scientific knowledge so
that we are better able to deal with complex multiple stresses, including uncer-
tainty, in decision-support systems?
• What research, conducted at the regional or sectoral level, might best
promote analysis of multiple stresses and provide information useful to decision
makers?
53

Appendix B
Workshop Agenda
Understanding and Responding to Multiple Environmental Stresses:

A Workshop
September 29-30, 2005
The Arnold and Mabel Beckman Center

100 Academy
Irvine, California
Thursday, September 29
8:30 a.m. Welcome and Introductions
Mary Anne Carroll, University of Michigan, and Rosina Bierbaum,
University of Michigan (Steering Committee Co-chairs)
Workshop Objectives
Chris Elfring (Director, BASC)
9:00 a.m. Keynote Address: What Do We Mean by “Multiple

Environmental Stresses” and Why Do We Care?
Eric Barron, Pennsylvania State University
9:45 a.m. Introduction to Case Study I: Drought as an Example of a

Multiple Stress Scenario
Donald Wilhite, University of Nebraska (Steering Committee)
54

APPENDIX B 55
10:30 a.m. Break
11:00 a.m. Understanding Hazard and Predictability of Drought

Facilitator: Donald Wilhite, University of Nebraska (Steering
Committee)
• Jonathan Overpeck, University of Arizona
• Kelly Redmond, Western Regional Climate Center, Desert
Research Institute
• Phil Pasteris, National Water and Climate Center, U.S.
Department of Agriculture
• Discussion and Questions
12:30 p.m. Lunch
1:30 p.m. Understanding Vulnerability and Response Strategies to

Drought
Facilitator: Jonathan Overpeck, University of Arizona
• William Easterling, Pennsylvania State University
• David Breshears, University of Arizona
• Roger Pulwarty, NOAA/CIRES/Climate Diagnostics Center
• Donald Wilhite, University of Nebraska (Steering Committee)
3:30 p.m. Break
4:00 p.m. Case Study I Wrap-up: Research Needs to Understand and

Respond to the Multiple Environmental Stresses that Lead to
Drought
Facilitator: Edward Miles, University of Washington (Steering
Committee)
5:30 p.m. Adjourn
6:30 p.m. Dinner
7:30 p.m. Dinner Speaker: The Millennium Ecosystem Assessment (MEA)

and How It Helps Us Understand the Impacts of Multiple
Stresses
Harold Mooney, Stanford University
8:30 p.m. Adjourn

56 APPENDIX B
Friday, September 30
8:00 a.m. Introduction to Case Study II: Atmosphere-Ecosystem
Interactions and Lessons About Multiple Stresses
Jerry Melillo, Marine Biological Laboratory
8:45 a.m. Understanding Atmosphere-Biosphere Interactions

Facilitator: Mary Anne Carroll, University of Michigan (Steering
Committee)
• Ronald Prinn, Massachusetts Institute of Technology
• Scott Doney, Woods Hole Oceanographic Institution
• Alex Guenther, National Center for Atmospheric Research
• John Reilly, Massachusetts Institute of Technology
• Jerry Melillo, Marine Biological Laboratory
10:30 a.m. Break
10:45 a.m. Understanding Vulnerability of Ecosystem Goods and Services

and Response Strategies
Facilitator: Rosina Bierbaum, University of Michigan (Steering
Committee)
• William Easterling, Pennsylvania State University
• David Breshears, University of Arizona
• Patricia Romero Lankao, Autonomous Metropolitan University,
Xochimilco, Mexico
• Richard Norgaard, University of California, Berkeley
12:30 p.m. Lunch
1:30 p.m. Case Study II Wrap-up: Research Needs to Understand and

Respond to Multiple Environmental Stresses in Atmosphere-
Ecosystem Interactions
Facilitator: Christopher Field, Carnegie Institution (Steering
Committee)
2:30 p.m. Synthesis of Workshop Findings: Multiple Environmental

Stresses—Knowledge Gaps and Future Directions
Facilitators: Mary Anne Carroll, University of Michigan, and
Rosina Bierbaum, University of Michigan (Steering Committee
Co-chairs)
4:30 p.m. Adjourn

Appendix C
Workshop Participants
Eric Barron, Pennsylvania State University

David Breshears, University of Arizona
Antonio Busalacchi, University of Maryland, College Park
Scott Doney, Woods Hole Oceanographic Institution
William Easterling, Pennsylvania State University
Jay Famiglietti, University of California, Irvine
Kristie Franz, University of California, Irvine
Alex Guenther, National Center for Atmospheric Research
Diana Josephson, National Center for Atmospheric Research
Chester Koblinsky, NOAA
Jane Leggett, EPA
Jerry Melillo, Marine Biological Laboratory
Harold Mooney, Stanford University
Richard Norgaard, University of California, Berkeley
Jonathan Overpeck, University of Arizona
Gi-Hyeon Park, University of California, Irvine
Phil Pasteris, USDA/National Water and Climate Service
Ronald Prinn, Massachusetts Institute of Technology
Roger Pulwarty, NOAA/CIRES
Kelly Redmond, Desert Research Institute
John Reilly, Massachusetts Institute of Technology
Patricia Romero Lankao, Autonomous Metropolitan University, Xochimilco,
Mexico
Richard Rosen, NOAA
57

58 APPENDIX C
Dongryeol Ryu, University of California, Irvine

Thomas Vonder Haar, Colorado State University
Committee
Rosina Bierbaum, University of Michigan
Mary Anne Carroll, University of Michigan
Christopher Field, Carnegie Institution of Washington
Edward Miles, University of Washington
Donald Wilhite, University of Nebraska
Staff
Chris Elfring
Liz Galinis
Ian Kraucunas

Appendix D
Extended Speaker Abstracts
ADDRESSING MULTIPLE STRESSES

Eric J. Barron
The University of Texas at Austin
The nature of the environmental issues facing any nation demands a capability
that allows us to enhance economic vitality, maintain environmental quality, limit
threats to life and property, and strengthen fundamental understanding of the earth.
In each case it is the ability to anticipate the future (e.g., a forecast of an impend-
ing storm, a prediction of the water quality change in response to a new source
of pollutant) that makes information about the earth system truly useful. Reliable
information about the future (i.e., predictions) is the key to addressing environ-
mental issues. However, society requires greater access to and greater confidence
in both information and forecasts or projections in order to weigh the advantages
and risks of alternative courses of action by private and public decision makers.
Such information is a key commodity in enhancing economic vitality and societal
well-being. What stands in our way of providing this information?
The Nature of the Problem

The driving forces that alter environmental quality and integrity are widely
recognized, involving primarily weather and climate, patterns of land use and
Presented in the order given during the workshop.
59

60 APPENDIX D
land cover, and resource use with its associated waste products. But a key feature
of most regions is that more than one driving force is changing simultaneously.
Consequently, most locations are characterized by multiple stresses. The effect of
a combination of environmental stresses is seldom simply additive. Rather, they
often produce amplified or damped responses, unexpected responses, or thresh-
old responses in environmental systems. Multiple, cumulative, and interactive
stresses are clearly the most difficult to understand and hence the most difficult
to manage.
In contrast, most research, analysis, and policy are based on studies that
examine discrete parts of these complex problems. Basically, earth and environ-
mental sciences tend to focus on cause and effect, where we seek to understand
how a specific element of the system may respond to a specific change or per-
turbation (e.g., acid rain on lake fisheries). The lack of an ability to assess the
response of the system to multiple stresses limits our ability to assess the impacts
of specific human perturbations, to assess advantages and risks, and to enhance
economic and societal well-being in the context of global, national, and regional
stewardship.
However, the problem is not limited simply to moving from analysis of dis-
crete parts of complex problems to a more comprehensive analysis.
First, economic vitality and societal well-being are increasingly dependent
on combining global, regional, and local perspectives. A “place-based” impera-
tive for environmental research stems from the importance of human activities
on local and regional scales, the importance of multiple stresses on specific envi-
ronments, and the nature of the spatial and temporal linkages between physical,
biological, chemical, and human systems. We find the strongest intersection
between human activity, environmental stresses, earth system interactions and
human decision making in regional analysis coupled to larger spatial scales.
Second, a decade of research on greenhouse gas emissions, ozone depletion,
and deforestation has clarified many critical unanswered questions. However,
the last decade of effort has also revealed a number of challenges, most notably
the challenge of creating integrated global observational capabilities and the
computational and scientific limitations inherent in creating a truly integrated,
global, coupled system modeling capability suitable for assessing impacts and
adaptations. These problems are noteworthy in global change science, but they
become intractable at the scales of human decision making. A major part of
the problem is simply a matter of scale combined with the sheer information
required to combine physical, biological, chemical, and human systems if the
framework is global. For example, whereas a global integrated observing system
is challenging but tractable and plays a fundamental role on the scale of a global
circulation model, it collapses under its own weight at higher spatial resolutions if
we demand a truly comprehensive data system involving the host of observations
spanning biology, hydrology, soils, weather, etc., required to address problems at

APPENDIX D 61
the scales of human decision making. For this reason, we have never fully real-
ized the objective of “earth system science.”
Recognition of the importance of developing a more integrated approach to
environmental research was abundantly clear in the U.S. National Assessment of
Climate Change Impacts on the United States (NAST, 2000). The first recom-
mendation for future research focused on developing a more integrated approach
to examining impacts and vulnerabilities to multiple stresses. There were many
examples in which the key limitation to the assessment of potential impacts to
climate change was a lack of knowledge of other stressors. For example, changes
in insect-, tick-, and rodent-borne diseases could be clearly tied to weather and
climate, but a number of other environmental factors that could influence the
disease vectors (e.g., the importance of land cover/land use on disease hosts),
transmission dynamics, and population vulnerabilities severely limited our ability
to make robust conclusions on how climate change might influence the distribu-
tion and occurrence of many infectious diseases in the future.
Addressing Societal Needs

The need for society to enhance economic vitality, while maintaining envi-
ronmental quality and limiting threats to property and life, should drive the
environmental research and operational enterprise. These societal needs lead to a
vision that requires a focus on multiple stresses. To address this vision, we need
to develop a comprehensive regional framework for environmental science. This
vision includes five elements:
1. an integrated regional web of sensors, including physical, chemical, bio-

logical, and socioeconomic factors, that link existing observations into a coher-
ent framework and enable new observations to be developed within an overall
structure;
2. an integrated and comprehensive regional information system, accessible
to a wide variety of researchers, operational systems, and stakeholders;
3. directed process studies designed to examine specific phenomena through
field study to address deficiencies in understanding;
4. a regional, high-resolution modeling foundation for constructing increas-
ingly complex coupled system models at the spatial and temporal scales appropri-
ate for the examination of specific and integrated biological, hydrological, and
socioeconomic systems; and
5. a strong connection to significant regional issues and stakeholders.
These five elements are described in detail below:
1. A Web of Integrated Sensors. The current U.S. observation strategy

appears to be haphazard when viewed in the overall context of environmental

62 APPENDIX D
problems. The reason is clear. The observations are driven by different mission
needs and tend to focus on the measurement of discrete variables at a specific
set of locations designed to serve the different needs of weather forecasting,
pollution monitoring, hydrological forecasting, or other objectives. The mission
focus results in a diverse set of networks that are supported by a large number of
different federal agencies, states, or regional governments. Increased awareness
of a host of environmental issues drives demand for additional new observations.
However, these new observations are frequently viewed independently of any
overall structure or integrated observing strategy. Operational needs and research
or long-term monitoring needs are also often independent. Importantly, regular
and consistently repeated observations present added challenges in garnering suf-
ficient financial resources. The end result is almost certainly fiscally inefficient,
and undoubtedly limits our ability to integrate physical, biological, chemical, and
human systems.
The limitations of the current observing strategy are widely recognized, and
they have spurred efforts to develop global observing systems for global change,
climate, and oceanic and terrestrial systems in the international arena. These
efforts are commendable and must be encouraged, but they are also extremely
challenging because of the breadth of measurements, nations, capabilities, and
policies that are involved.
In contrast, at a regional level in the United States we have the potential to
(a) link observing systems into a web of integrated sensors building upon exist-
ing weather and hydrological stations and remote sensing capability; (b) create
the agreements across a set of more limited agencies and federal, state, and local
governments needed to create a structure to the observing system; (c) provide a
compelling framework that encourages or demands the integration of new obser-
vations into a broader strategy; and (d) create strong linkages between research
and operational observations that result in mutual benefit. The result is likely to
create new efficiencies through the development of measurement systems that
are more comprehensive, rather than a suite of separately funded, disconnected
systems. The result is also likely to result in greater scientific benefit to society
and greater understanding due to the co-location or networking of many different
measurement capabilities. The demonstration of fiscal efficiency and improved
capability and resulting benefit are likely to create a significant additional impetus
for developing national and global integrated observing systems.
2. Regional Information Systems. Society has amassed an enormous amount

of data about the earth. New satellite systems and other observational capabili-
ties promise enormous increases in the availability of earth data. Fortunately,
technological innovations are allowing us to capture, process, and display this
information in a manner that is multiresolution and four-dimensional. The major
challenges involve data management; the storage, indexing, referencing, and
retrieving of data; and the ability to combine, dissect, and query information. The

APPENDIX D 63
ability to navigate this information, seeking data that satisfy the direct needs of a
variety of users, is likely to spark a new “age of information” that will promote
economic benefit and engender new research directions and capabilities to inte-
grate physical, biological, chemical, and human systems.
The efforts to create comprehensive information systems increasingly reflect
federal and state mandates to make data more accessible and useful to the public
and to ensure that our investments in research yield maximum societal benefit.
The development of a global digital database is again an enormous challenge. In
contrast, a regional or state focus becomes a logical test bed, enabling the par-
ticipation of universities; federal, state, and local governments; and industry in
the development of a regional information system that is tractable and for which
immediate benefit for a state or region can be evident. Again, the demonstration
of capability and resulting benefit are likely to create a significant additional
impetus for developing national and global information systems.
3. Framework for Process Studies. Process studies are a critical element of

scientific advancement because they are designed, through focused observations
and modeling, to probe uncertainties in knowledge about how the earth system
functions. In many cases mismatch between model predictions and observations
can drive targeted investigations to limit the level of error. Frequently, efforts to
couple different aspects of the earth system (e.g., the atmosphere and land-surface
vegetation characteristics) prompt targeted exploration because the level of under-
standing is still rudimentary. The objective is to use field study to address defi-
ciencies in our understanding. The benefit of these intensive studies is maximized
when they can be coupled with a highly developed, integrated set of sensors, with
readily accessible spatial and temporal data within a regional information sys-
tem, and with a predictive model framework that readily enables the entrainment
and testing of new information from process studies. Hence, a regional focus is
empowered by process studies that are directly tied to answering specific ques-
tions designed to assess the impacts of specific human perturbations, to assess
advantages and risks, and to enhance economic and societal well-being.
4. Predictive Capability. The value of reliable advanced information is

widely recognized. For example, we have considerable experience with the ben-
efit of reliable weather forecasts, whether it involves the planning of a day around
a precipitation forecast, the protection of life and property that stems from severe
weather warnings, or the economic benefit of weekly, seasonal, or interannual
forecasts used by climate- or weather-sensitive industries such as agriculture,
forestry, fisheries, construction, or transportation. For example, millions of dollars
in commodity markets can be saved by regional utilities with advanced weather
or seasonal forecasts, and El Niño forecasts a season in advance can substantially
modify agricultural practice.

64 APPENDIX D
Prediction is central to the translation of knowledge about the earth system

into economic benefit and societal well-being. Over the last several decades we
have experienced enormous increases in our ability to forecast weather and to
project climate and climate variability into the future. The demand for new fore-
casting products, involving air quality, energy demand, water quality and quantity,
ultraviolet radiation, and human health indexes is also growing rapidly, and as we
demonstrate feasibility and benefit, society is likely to demand a growing number
of new operational forecast products on prediction timescales of days to decades
into the future. Further, we already clearly sense that environmental issues will
demand an even greater capability to integrate physical, biological, chemical, and
human systems in order to develop the predictive capability needed to examine
the response of critical regions or cases to multiple stresses.
Global weather and climate models provide the strongest physical founda-
tion for more comprehensive predictive capability. The numerical models that
underpin this type of forecast are increasingly becoming the framework for the
addition of new numerical formulations designed to predict air quality, the water
balance for river forecast models, and a host of other variables, including the
migration of forests under climate change conditions. As we attempt to produce
predictions at the scale of human endeavors, mesoscale models (capable of pre-
dicting synoptic weather systems) and downscaling of coarser resolution model
output are increasingly becoming the focal point of weather and climate studies
because of their potential to make predictions on the scale of river systems, cities,
agriculture, and forestry. Enormous potential exists if we can institutionalize a
mesoscale numerical prediction capability that meshes with regional sensor webs
and information systems. Such a capability enables a strategy and implementation
capability for building tractable coupled models, initiating experimental forecasts
of new variables, assessing the outcomes associated with multiple stresses, and
taking advantage of the discipline of the forecasting process to create a pow-
erful regional prediction capability. This capability, built upon the numerical
framework of weather and climate models, can be extended to air quality, water
quantity and quality, ecosystem health, human health, agriculture, and a host of
other areas.
It is time to bring a demanding level of discipline to the forecasting of a wide
variety of environmental variables. The objective is to stimulate the interplay
between improvements in observation, theory, and practice needed to develop
capabilities of broad value to society. The discipline of forecasting is dependent
on four steps: (1) collection and analysis of observations of present conditions;
(2) use of subjective or quantitative methods to infer future conditions; (3) assess-
ment of the accuracy of the prediction with observations; and (4) analysis of the
results to determine how methods and models can be improved. At a minimum we
are capable of bringing a much greater level of structure and discipline into our
predictions of the future, ranging from specific forecasts to statistical ensembles

APPENDIX D 65
that include a measure of expected accuracy to an assessment of the range of

possibilities.
5. Creation of a Vigorous and Continuous Link to Users and Decision

Makers. We need to create a vigorous connection between the research and
decision makers by (a) incorporating the variety of space and timescales and the
diversity of variables that are important to decision makers; (b) emphasizing
the education of the user in the meaning and significance of climate and land
use information in order to promote greater use and more robust applications;
(c) ensuring mutual information exchange and feedback; (d) focusing on com-
munication and accessibility of information; (e) continuously evaluating and
assessing the use and effectiveness of the services; (f) employing active mecha-
nisms to enable the transition from research discovery to useful products; and
(g) employing a variety of methods of education and outreach.
Summary
The above structure is inherently a hybrid between research and operational
functions. Both benefit from the level of integration of observations and infor-
mation, the targeted process studies, and the model development capability. An
emphasis on a region-specific predictive capability will drive the development of
new understanding and new suites of comprehensive interactive high-resolution
models that focus on addressing societal needs. A key objective is to bring a
demanding level of discipline to “forecasting” in a broad arena of environmental
issues. Common objectives and an integrated framework will also engender new
modes and avenues of research and catalyze the development of useful opera-
tional products. With demonstrated success, the concepts of integrated regional
observation and information networks, combined with comprehensive models,
will grow into a national capability that far exceeds current capabilities. Such a
capability is designed to address a broad range of current and future regional and
global environmental issues.
Eric Barron is dean of the Jackson School of the University of Texas at Austin.
Prior to assuming his current position he served as dean of the College of Earth
and Mineral Sciences and Distinguished Professor of Geosciences at the Penn-
sylvania State University. His research interests are in the areas of climatology,
numerical modeling, and earth history. Dr. Barron received his Ph.D. in ocean-
ography from the University of Miami.

66 APPENDIX D
DROUGHT: OBSERVATIONS, MECHANISMS,

POTENTIAL SURPRISES AND CHALLENGES
Jonathan Overpeck
University of Arizona
INTRODUCTION
Drought is one of the most challenging and costly environmental concerns
confronting society. Over time, many institutions have been developed to deal
with drought, but there is still a regular multi-billion dollar annual drought impact
in the United States alone. Moreover, a complete consideration of drought pro-
vides a sobering view of the future. The purpose of this presentation is to provide
an overview of drought variability and related issues.
Drought is a concern worldwide, but by necessity this presentation will focus
on North America only. Nonetheless, many of the North American lessons can
inform climate-society debates elsewhere—particularly in Africa, where there is
also a rich history of observations, as well as research on the causes and impacts
of drought variability.
THE INSTRUMENTAL RECORD OF DROUGHT

Drought is always affecting some part of North America, and even a single-
year drought has impact and is thus important. The longer droughts are of greater
concern, with the droughts of the 1930s, 1950s, and 1999 to the present being
the droughts of record and of most concern. These longer droughts all affected
large portions of North America. The most recent drought is particularly notable
in that it was hotter than the similar drought of the 1950s. In general, western
North America has seen significant warming over the last 100 years, particularly
in the last couple of decades.
It is notable that the recent period of unprecedented population increase in
the western United States coincided with one of the wettest periods on record.
It is also troubling that Colorado River water was divided up among basin states
during the one of the other wettest periods: in total, 16.5 MAF of water was
allocated per year, even though the measured flow during the early 20th century
wet period was less than 15 MAF per year.
THE PALEOCLIMATIC RECORD OF DROUGHT

Even a quick glance at the paleoclimate record of drought makes it clear that
the instrumental record provides a biased record of drought due to a shortness of
record. Droughts comparable to the 1930s, 1950s or early 2000s have occurred,

APPENDIX D 67
in general, once or twice per century over the last 2,000 years. The true long-term
average flow of the Colorado is well below what is allocated to western states and
Mexico. All droughts of the 20th century were eclipsed by past droughts, both
in terms of annual maximum severity, duration, and geographic extent. It is clear
from the paleoclimatic record that droughts longer than a decade (i.e., “mega-
droughts”) were not that rare and that droughts impacting much of the western
United States have lasted as long as a century or more within the last 2,000 years.
Paleoclimatic research has shown that at least one lake in the Sierra Nevada of
western North America went dry for decades on more than one occasion—some-
thing that has not occurred since Europeans settled the West. The true range of
“natural” drought variability is thus substantially larger and more complex than
suggested by the last century of instrumental drought variation.
In general, most of the large droughts of the western United States have been
large enough to affect more than one major river basin at a time, and some (e.g.,
the late 16th-century megadrought) apparently impacted the United States from
coast to coast and from northern to southern borders. Many (all?) long-duration
droughts moved spatially from year to year and usually included years with nor-
mal or above-normal rainfall.
Another key aspect of drought variability illuminated by the paleoclimatic
record is that decades- to centuries-long hydrological “regimes” (e.g., character-
ized by rare/short or frequent/longer droughts) have begun and ended abruptly—
transitions between drought regimes can take place over years to decades, whereas
the regimes themselves can be significantly longer.
MECHANISMS OF DROUGHT VARIABILITY

Great strides have been made recently with respect to understanding the
proximal cause of drought in North America. We have long known that drought
in the southwestern United States (e.g., 1950s and the recent drought) have con-
nections with ENSO and that dry winters are favored in La Niña years. More
recently, modeling studies have confirmed that anomalous sea surface tempera-
ture (SST) patterns, particularly in the tropics, and particularly in the Indo-Pacific,
can explain both 20th-century and earlier droughts. Moreover, research has now
also identified strong relationships between decadal modes of Pacific and Atlantic
variability with decadal patterns of wet and dry conditions over North America.
Of course, the major challenge is to explain the causes of the anomalous—and
persistent—ocean conditions that lead to North American drought.
Although land-surface feedbacks can play key roles in amplifying or damp-
ing drought, it appears that these feedbacks are not dominant in driving temporal
drought variability. More work is needed on the role these feedbacks play, but
only after recognizing that coupled atmosphere-ocean variability is likely more
important.

68 APPENDIX D
POTENTIAL SURPRISES
It is just a matter of time until North America gets hit with a decadal mega-
drought. This is likely regardless of how large any anthropogenic impact might
be. Thus warned, why would anyone be surprised? This shortcoming of human
nature needs to be understood and overcome.
ANTHROPOGENIC VERSUS NATURAL DROUGHT INFLUENCES

The IPCC (TAR and a recent CLIVAR-PAGES-IPCC workshop convened
by Overpeck and Trenberth) concluded that anthropogenic forcing will likely
increase the probability of drought in central and western North America. Exacer-
bating this likelihood is the fact that temperatures are already rising significantly
in the West, and snowpack is already retreating in the same region (as a result of
the warming). A lesson of the paleoclimatic record is that anthropogenic forcing
could trigger an abrupt transition into a more drought-prone climatic regime,
thus increasing the frequency and duration of drought. Given that these possibili-
ties could materialize with or without significant future human-induced climate
change, it makes sense to consider no-regrets strategies to reduce vulnerability
to drought in either case.
Most recently it has been proposed that unprecedented warming in the tropi-
cal Indian and western Pacific Oceans—likely a result of global warming—is the
cause of the ongoing western North American drought. This hypothesis is well
supported by modeling studies and is troubling in that the anomalous warming
could become the norm in the future in years without El Niño events. If dry con-
ditions intensify this winter, it could be a sign that we are in what could turn out
to be the first megadrought since Europeans settled North America.
CHALLENGES AND OPPORTUNITIES

Needless to say, there is a great deal of additional insight that could be
revealed through a more comprehensive study of past drought and megadrought.
Can we define the full range of possible drought sufficiently for successful no-
regrets adaptation? What are the empirical links between drought and variability
in sea surface temperatures? What sets up the anomalous SSTs? Are the anoma-
lies forced or generated by internal variability in the coupled system? What are
the potential triggers of abrupt hydrological change? Are there early warning
signs to be monitored for? Can droughts and megadroughts be predicted? Are
surprises avoidable?
Although the mechanisms of drought cannot be understood fully without
much improved observations (particularly paleo), the most important challenge
might be in the area of ocean and coupled atmosphere-ocean modeling. Whereas
the instrumental record of drought can be simulated fairly well given specified
(observed) SST fields, the same drought record cannot be simulated with state-

APPENDIX D 69
of-the-art coupled models. Clearly, the development of more realistic coupled

models must be an urgent priority.
Dealing with drought requires the ability to understand regional climate
variability over all seasons and to eventually be able to deliver reliable seasonal
to interannual climate outlooks. At present, advances are limited by modeling
regional-scale processes, and this limitation is in turn related by the lack of good
regional-scale climate monitoring. Thus, there is an increased need for denser
in situ climate observing in the topographically complex western epicenter for
U.S. drought.
BOTTOM LINE?
Even if we develop an ability to predict drought, we must work with stake-
holders in society (particularly in central and western North America) to develop
adaptation strategies that reduce vulnerability to megadrought. There is little
doubt that such a drought will occur at some time in the future and that anthropo-
genic climate change will exacerbate the situation. There is also little doubt that
stakeholders and institutions are ill prepared for the inevitable megadrought.
Jonathan Overpeck is both a professor and director of the Institute for the Study
of Planet Earth at the University of Arizona. Dr. Overpeck’s research focuses
on global and regional climate dynamics. His research aims to reconstruct and
understand the full range of climate system variability, recognize and antici-
pate possible “surprise” behavior in the climate system, understand how the
earth system responds to changes in climate forcing, and detect and attribute
environmental change to various natural (e.g., volcanic, solar) and nonnatural
(e.g., greenhouse gases or tropospheric aerosol) forcing mechanisms. His work
also focuses on improving the use of climate knowledge by decision makers in
society.

70 APPENDIX D
UNDERSTANDING HAZARD AND

PREDICTABILITY OF DROUGHT
Kelly T. Redmond
Desert Research Institute
What are the regional differences in complexities of drought monitoring?

What drought monitoring and data needs have been identified for the United
States?
At first blush, drought seems like a simple concept. However, the many ways
in which it is expressed show how complex a phenomenon it really can be. The
following personal observations elaborate on this point.
THE DEFINITION OF DROUGHT

Over the years I have attended a large number of drought status meetings,
and even if the subject is not discussed right away, sooner or later the conversa-
tion turns to a definition of drought. Many definitions are geared toward a spe-
cific economic sector, geographic province, or other situation. The definitions of
drought attached to these situations are seldom general or readily generalizable,
thus rather arbitrary, and thus not very satisfactory. There is a strong predilection
to view drought almost solely in terms of its climatic and physical drivers. Most
definitions refer to extended periods and to deficient precipitation. But what are
the durations and amounts, and perhaps other attributes, that these are referenced
to? And is there a single such referent?
It seems unavoidable that the context of the situation where the question is
being posed has to be brought into play. And if context matters, then we are led
to questions of “deficient from which standpoint?” Water deficiencies can occur
when water income is not great enough, or when water outgo is too great. Some
drought indices address a mismatch between supply and demand of water (e.g.,
the Palmer index family), typically in the soil, but a large number do not. Other
measures, like the Standardized Precipitation Index (SPI), focus on supply, with
the implication that this dominates. Technically, the SPI is not a drought index
itself but rather one of many tools to evaluate drought. More generally, though, it
seems that drought has come into existence when the amount of water available is
not keeping up with expectations or needs. These expectations, in turn, are based
on long-term expectations of supply balanced against long-term expectations of
demand. What constitutes “long-term” has to do with the timescales for getting
into or out of difficulty. These can of course vary quite widely, from hours to
centuries, though we generally mean days to months or years, depending on the
issue or sector.
A definition is not of much use if it is not always applicable. Nearly every
concept of drought involves a water budget (water supply versus water demand)

APPENDIX D 71
that has somehow come to be “out of balance” over some extended time. For
these reasons, I have been led toward a simpler and more widely applicable
definition of drought:
insufficient water to meet needs.
The main point here is that both supplies and needs (demand) change, so that
for a given rate of supply, if demand goes up, the likelihood of shortages aris-
ing from typical fluctuations in supply goes up. When Albuquerque’s population
consisted of 19 people, the deficiency in supply that led to shortages was different
from the deficiency in supply that resulted in shortages when there are 600,000
people. Both supply and demand are dynamic.
Furthermore, there are many demands for water from natural systems (veg-
etation and wildlife and fish), and any useful comprehensive definition of drought
has to encompass all biological systems. We could add a clause that refers to the
needs of such biological systems, but a simple definition is much more preferable
from the standpoint of elegance and clarity of thought.
Even if supply is reduced, if there is still sufficient water to meet all needs,
then there is no drought. If supply remains constant, and demand is increased,
there can be drought where there was none before. Such imbalances in supply
and demand express themselves as impacts. In essence, if there is no impact,
there is no drought. Thus, an inescapable corollary that accompanies this line of
reasoning is that
drought is defined by its impacts.
In routine assessment of drought, such as for the U.S. Drought Monitor, this
is the approach commonly taken. This kind of definition is harder to deal with
for those who like concrete, definite numbers, because it is “soft” and situation
dependent. But I am unable to come up with any example of drought that is not
situation dependent.
Expressing this somewhat differently, an adequate definition or framework
for conceptualizing drought ought to work equally well in Death Valley and the
Olympic Mountains, in Greenland or Kentucky or Panama or Kihei or Aconcagua.
Biological systems in all those places must address water supply and demand.
REGIONAL DIFFERENCES IN THE CHARACTER OF DROUGHT

It is exactly the regional differences in how we view and describe drought
that led to the foregoing train of thought. The nature of the lags in the physi-
cal system is of fundamental importance. These lags can be natural or human
caused. When there are such lags (from storage mechanisms that depend on water
phase—liquid or frozen—or from different types of buffering mechanisms—deep

72 APPENDIX D
aquifers, reservoirs, porous rock, etc.), generally a system of physical infrastruc-

ture, conventions, legal mechanisms, cultural and social practices, and the like
will have grown up around them.
In the United States the most striking difference is of course the difference
between snowmelt-driven systems and rain-driven systems. These are strongly
tied to the seasonal variations in the timing of precipitation, and to temperature,
since this influences the phase of new precipitation (liquid or frozen) and the
melt and runoff characteristics. Temperature is very important for all of these
processes and also affects the demand through the phenological stage. Though not
often treated as such, temperature is a hydrological element. All of these factors
are elevation dependent and can thus vary over quite small scales (less than a
kilometer)—amount, seasonality, phase, vegetative demand, sublimation, and
others. To cope with these factors, an elaborate system for tracking and manag-
ing water in the West has arisen, in part codified into law, and a large number
of dams or other storage mechanisms have also arisen to retain water until it is
needed. Furthermore, water flows downhill, so gravitational considerations are a
necessity (storage upstream always has more flexibility). Also, somewhat sepa-
rately, because of the mining heritage, a system of water rights has arisen in the
West that is different from that in the East. These factors fold together in intricate
and complicated (i.e., nonlinear) ways, with thus the potential for many kinds of
interesting behavior.
Another regional factor is the distance between the supply of water and the
demand for that same water, often hundreds of miles away, and many months,
even years, afterward. The water that supplies desert cities typically arises in
locations not even visible from those cities. This water crosses a variety of legal
and political boundaries, necessitating a variety of agreements, compacts, and
understandings. There are great differences regionally between rain-dominated
and more humid environments such as in the eastern United States and the snow-
dominated and very arid environments found in the western United States.
In addition there are significant differences among states and across inter-
national boundaries in how groundwater is addressed, particularly whether it is
considered (legally) to be linked with surface water, even though it is very well
known that physically this is the case. Recognizing and addressing this linkage
could affect some key interest groups in ways that would not provide incentive
to change.
Drought impacts are experienced across a continuum of spatial scales, many
of them at the scale of an individual person or household or small business. We
have much less information at those scales and often have none. This applies to
the causative physical factors as well as to the individual impacts. If we are lack-
ing these local impacts, we are not in a position to aggregate upward and identify
impacts that are cumulatively large but individually small. In part this difficulty
stems from the difficulty of sampling to this level. Such sampling takes time,
resources, skilled personnel, and coordination so that other analogous results can

APPENDIX D 73
be compared and added. In general there are woefully few resources to address
information needs at the local level. In addition, such capabilities vary widely
from state to state; more uniformity is needed. This should be a priority of a
national integrated approach to drought, such as with NIDIS (National Integrated
Drought Information System). We are also lacking in mechanisms to record and
track such information, but recent progress has been reported by the National
Drought Mitigation Center in developing tools to assist with this cumbersome
but important problem.
Until we can obtain comprehensive assessments of the full impacts of
drought, across all scales, we cannot provide the documentation that decision
makers often require to back up requests for utilization of public funds.
MULTIPLE STRESSES
There are of course many factors at least as important as climate that govern
whether there is enough water for a particular purpose. Trends in spatial distri-
bution and numbers of people and population demographics greatly affect water
use. There are many competing uses and needs for water by people (municipal
and industrial, transportation, dilution and conveyance of waste, hydropower,
recreation, traditional cultures) and by larger ecosystems (fish and wildlife,
endangered species, silt deposition and delta health, benefits from flow fluctua-
tion, marine and estuarine systems, delivery of fresh water affecting ocean circu-
lation, nutrient transport and deposition). Water quantity and timing affect water
quality. Each issue has optimal water flow and quality characteristics and requires
that constraints be met for satisfactory function. These constraints cannot usually
be both individually and collectively optimized simultaneously, thus requiring
global optimization over all issues and thus compromises.
This is basically an issue of parts versus wholes. We can identify and describe
each stress in great detail as a separate subject. It is when they combine that all
sorts of interesting behaviors become possible. However, we should not lose sight
of the fact that we deal with multiple stresses all the time and we generally make
it through life anyway.
COMPLEXITY AND NONLINEAR DYNAMICS

Because we have multiple stresses interacting at the same time, and
because these interactions are highly nonlinear, we are in effect automatically
preregistered in a giant experimental game from which we cannot withdraw.
Emergent phenomena abound, and we should not be surprised to see unexpected
or unpredictable things occur. We cannot say in detail what those are going to
be; it is rather simply an issue that we should be prepared to be surprised at
any time.

74 APPENDIX D
Because of this, it seems that the intertwined subjects of complexity, non-

linear dynamics, and chaos (and allied concepts) have a great deal to say about
the generalized behavior we would expect to encounter. One of my own difficul-
ties has been to try to discern where the knowledge we have gained about this
subject can be supplied in some useful, practical, specific, or otherwise helpful
way. In other words, where are the “insertion points” for the knowledge and
insights developed by this field, in trying to cope with drought in a multi-stress
environment? As with the stock market (itself a highly complex and nonlinear
phenomenon), “past performance may not be an adequate predictor of future
performance.”
Many of our response mechanisms do not seem to adequately allow for
low-probability, high-consequence phenomena that form part of the risk port-
folio. What we really have is a large number of probability distributions (and
some of them with heavy tails, such as precipitation) interacting in many ways,
with nonzero odds of occasionally rather spectacular outcomes. By definition,
we call these “surprises” but we should not be surprised (in a general way) to
see (specific) surprises. We know they are likely to arise, but we do not know
with any detail what they will consist of. The typical city fire department can be
confident that a house will catch fire, but it cannot say which house. There are
many military analogs, because in that arena many consequences of surprise are
generally not favorable, but much thought and training are devoted to maintaining
the flexibility to address them when they do happen.
Naturally we should try to anticipate as many types of outcomes from inter-
acting stresses as possible, in a deterministic fashion, but it is simply hopeless to
guess them all. Thus it is guaranteed that surprise will always be present.
All of this plays into our tendency to formulate management plans, and most
especially, laws, that are unnecessarily (often in order to gain their approval) rigid
and not reactive to new or developing information.
BOUNDARY CONDITIONS AND

INITIAL CONDITIONS FOR DROUGHT
By boundary conditions I am referring to the general circumstances “exter-
nal” to a particular water balance determination, analogous to a boundary value
problem in mathematics. In truth, in a connected system, there is hardly anything
that is really “external,” but that is being glossed over for now. As these condi-
tions change, the degree or scope or other properties of some current drought
situation are likely to be assessed differently. The history of how a current situa-
tion came into being may make a considerable difference as to what needs to be
done, for two situations with similar current conditions.
One boundary condition that may not be stationary in time is demand, which
is often proportional to population in some manner. A rather typical assumption
made is that drought status is driven more by supply than demand. And it is true

APPENDIX D 75
that the relative (annual) variation of precipitation is typically two to five times
larger than the relative variation of evapotranspiration. The temporal variation of
demand is not zero, however. In addition, population growth, at whatever rate,
can, for example, negate assumptions about the stasis of a system of interest.
Groundwater contamination, as one example, can drastically alter supply,
even if the source of contamination has been slowly building or gathering. A
contaminated plume that is many years old may show up all at once. Groundwater
flow through fractured media can exhibit many properties of Levy diffusion and
related “burst” behavior. The impacts of such behaviors are often regarded as
“surprises” and thus as pathologies, whereas a more careful analysis would have
allowed for their possible existence.
Certain climate conditions (prolonged drought or prolonged moisture) can tip
a system into a new “basin of attraction” in terms of state variables. For example,
slow withdrawal from a water table can suddenly kill plant life when the roots
can no longer reach deep enough. Hysteresis can then occur, because the reverse
trajectory is not possible (dead plants do not become undead when the roots are
reached as the water table replenishes). Slow changes in demand can make it
more likely that some extreme condition occurs more often than before, increas-
ing the likelihood that simultaneous occurrence of multiple extreme conditions
takes a system to coordinates in state space that it has not hitherto visited.
RATES MATTER
We are accustomed to the processes in various systems proceeding at certain
ranges of rates. If sustainability is the goal, then there needs to be a matching of
rates of supply versus demand, for all resources that are being consumed, or even
merely used. In many cases groundwater withdrawal is occurring at far faster rates
than the rates of aquifer recharge. This is especially true where we are mining
groundwater that is hundreds or thousands of years old. Furthermore, recharge
rates are highly variable and episodic in time (in various settings recharge can
be steady or bursty), and in arid and mountainous environments especially, they
are extremely variable in space, in both constant ways (mountains do not move,
except in California) and time-dependent ways (monsoon thunderstorms, for
example).
With systems that exhibit intermittent or bursty or transient behavior, the
averaging time over which rates are computed can be very important. A heavy
monsoon rainstorm in the desert may rain 5 inches per hour, but only for 10 min-
utes. The net effect of several such storms might later be expressed as “3 inches
per year.”
In addition, another rate comes into play. This is the rate at which our per-
ceptions change as the circumstances in our vicinity change. Nearly all human
beings live more in the past than in the present; we are always behind in our
thinking. In an ideal world we would probably actually be living in a projected

76 APPENDIX D
future (assuming we could do that correctly) and making our decisions from that
perspective. But in reality our perceptions often lag the real situation because of
learning times, communication delays, the need to spend time on other things,
perhaps a certain unwillingness to be totally up to date, and the like. As an
example, it seems likely that the typical native westerner is not fully aware of the
full dimensions of sprawl in their favorite vicinity. Most of us are likely making
decisions that reflect our understanding of how things were a minute or a week
or a year ago, or two, or five, or 10, or 50. We probably could not cope if we
endeavored to keep abreast of every development, so there may be elements of
psychological defense mechanisms at work that have withstood the tests imposed
by evolution.
The point here is that communication and the transmission of learning cannot
occur instantly everywhere, every part of a system has imperfect knowledge of
the status of the other parts, and the differences in such rates are a reflection of the
contingent nature of history. These differences lead to differences in strategies for
addressing whatever stresses are on the plate at the moment.
SEQUENCING
The order in which things occur makes a difference. We have heard this
from a number of drought and water managers. A single wet year in a sequence
of dry years can yield much different overall consequences than if that wet year
had been first or last. In some systems the drought clock can be partially or fully
reset by a well-timed recharge. To help create a “worst case” scenario, the study
in the early 1990s of severe and sustained drought on the Colorado River initiated
the dry period with the lowest observed runoff years in succession.
Furthermore, the sequencing of different facets of climate can have great
consequences, such as a particular moisture regime (wet or dry) accompanied or
followed by a particular temperature regime (hot or cold). Insects, pathogens, and
other pests often show spectacular responses to such combinations. The mormon
crickets that have been prevalent in the northern Great Basin in recent years are
favored or hurt by specific sequences of weather and climate anomalies. Drought
and warm temperatures are permitting bark beetle behavior on an unparalleled
scale in the northern Rocky Mountains of the United States and Canada in the
last few years. For the latter, threshold factors, such as the ability to squeeze in
two generations per year instead of one, are one example where highly nonlinear
processes can qualitatively change the manner in which a system works.
Political action on complex problems is very often affected by essentially
random accidents of timing.

APPENDIX D 77
PREDICTABILITY AND PROBABILITY

With respect to the climate system, we have piecewise understanding of its
internal workings, its current status, the status of boundary conditions, and the
ability to model it to some level of accuracy. But all of these are imperfect. Some
of these imperfections can be addressed through diligence and expenditure of
resources, and others represent fundamental limitations inherent in what we can
know. Predictions can be made better by improving the former, but are always
subject to the limitations imposed by the latter. Prediction is only possible for
patchy combinations of circumstances, and we can only know some of those
combinations.
All of our understanding points to Nature as being fundamentally probabi-
listic. Thus, even though it is difficult, we should learn to work and think in this
mode as much as we can. In particular, we should be thinking in terms of the
probability distributions of our confidence in our observational database, and
in the descriptions of how we model how the system evolves from one state to
the next.
This represents a huge challenge for public education. In like manner, we
have known about quantum mechanics for 80 years, but most of the public has
a very poor understanding of this subject and does not even know it exists, even
though it is the fundamental basis for how things function.
In atmospheric science, including both weather and climate forecasting,
the ensemble approach has been increasingly applied and has resulted in better
forecasts. This in effect constitutes a kind of poor man’s method for sampling the
possible states of a system, given the observations and the understanding of some
of their error characteristics, and for sampling the physical parameterizations that
represent how certain processes work. The same set of equations is run many
times over but with random differences imposed on the input data, reflecting
observational uncertainty, and in more complex approaches (“superensembles”),
averaging across models that have a variety of approaches to parameterization
of some particular process. The outcome of a typical set of forecast runs is that
there are many possible solutions, with tendencies to cluster in certain ways.
There are methodologies that can arrive at a consensus forecast but still retain
the uncertainty and that furnish confidence information of value to other parties
trying to utilize these results.
This ensemble methodology seems particularly valuable whenever there are
a large number of degrees of freedom in a system. The climate system certainly
has this property, and the drought/impacts “system” has far more than that. Fur-
thermore, many of the latter are not subject to physical parameterization and are
thus probably better described by using probability distributions. This method
seems to offer many possibilities beyond the world of meteorology and climate
where it originated and is still being refined.
The goal is to obtain better knowledge of the probability distributions of pos-
sible outcomes and assess what level of risk is attached to each outcome in the

78 APPENDIX D
distribution. With this better understanding of what outcomes are possible, the
costs and benefits associated with each outcome, and development of a weighted
average outcome, can be evaluated. The net effect is that we do not end up
making definitive and definite conclusions about what is or is not the case. We
need to be neither too confident nor too diffident in what we believe, and we need
probabilistic tools to help us with these assessments. Probabilities and risk-based
approaches can lead to better decisions, including the tenacity with which we
defend or suggest certain courses of action in the face of uncertain results.
The suggestion from this observer is that such approaches have the potential
for reducing our usually undue confidence in the results we arrive at and the
actions we recommend based on those results. This does not preclude the ability
to utilize probabilistic guidance to help crystallize our thinking and make rapid,
decisive, and definitive decisions when needed.
Some elements of this discussion were raised in the article “The Depiction
of Drought: A Commentary,” by Kelly T. Redmond, Bulletin of American Meteo-
rological Society, 2002, 83(8):1143–1147.
Kelly T. Redmond is regional climatologist and deputy director for the Western
Regional Climate Center located at the Desert Research Institute in Reno, Nevada.
His research and professional interests span every facet of western U.S. climate
and climate behavior, its physical causes and behavior, how climate interacts with
other human and natural processes, and how such information is acquired, used,
communicated, and perceived. Dr. Redmond received his Ph.D. in meteorology
from the University of Wisconsin, Madison.

APPENDIX D 79
DROUGHT MONITORING COMPLEXITIES IN THE WEST

Philip Pasteris
U.S. Department of Agriculture
“What can be learned about the impacts and interactions of multiple stresses
from records of western U.S. precipitation and climate observations and their
relation to drought and water supply forecasting?”
BACKGROUND
“The future ain’t what it used to be.” Y. Berra
The western United States was built with, and is highly dependent on, water
captured from mountain snowpacks that may be hundreds or even thousands
of miles away from population centers and agriculture. The reservoirs built to
meet the needs of agriculture, power generation, municipal water supply, and a
variety of other uses were conceived, and in some cases built, nearly 100 years
ago when populations were scarce, industry demand for power was in its infancy,
and endangered species legislation did not exist.
The West’s climate and mystique have lured settlers for over 150 years;
however, it is ironic that drought was the genesis of one of the largest migrations
during the 1930s. The population of the West was only 11.9 million in 1930 but
grew 15 percent to 13.7 million by 1940 during the drought years. The West
was able to absorb this increase; however, a pattern of western migration had
begun. Significant reservoir construction from 1930 to 1970 resulted in increased
water availability and inexpensive power, allowing irrigated agriculture, increased
populations, and industry to gain a secure foothold.
Between 1980 and 2000 the population of the West grew 47 percent, from
42.2 million to 62 million, with no significant increase in water storage infrastruc-
ture. By 2002 agriculture represented 25.5 million western acres, generating an
annual products-sold market value of $51.1 billion.
Recent energy prices have ignited interest in nonfossil renewable energy gen-
eration. Agencies such as the Bonneville Power Administration, which markets
power produced from 31 federal dams in the Columbia Basin, have established
operating plans based on climate and streamflow records for the period 1929-1978.
Climate variability, combined with a projected 8 percent increase in regional firm
energy demands, from an estimated 23,300 average megawatts in 2005 to 25,200
average megawatts in 2012 will have a direct impact on power availability and
western economies. Columbia Basin hydroelectric dams, which rely on winter
snowpack accumulation and spring and summer melt cycles, provide 73 percent
of the region’s energy (BPA, 2003).

80 APPENDIX D
Recent Trends in Western U.S. Streamflow Variability

and Persistence and Potential Impacts on Drought
Forecast streamflows and reservoir rule curves govern daily, monthly, and
water year reservoir operations for power generation, navigation, flood control,
and endangered species management. Many rule curves were developed using
data from the middle of the past century and may not fully represent the recent
streamflow trends and variability. A recent publication in the Journal of Hydro
meteorology, “A Recent Increase in Western U.S. Streamflow Variability and
Persistence” (Pagano and Garen, 2005), investigated trends in western U.S.
streamflow. Variability and persistence can combine to amplify stress in drought-
prone areas or, in extreme cases, produce drought in previously drought-free
areas.
From the abstract: “April-September streamflow volume data from 141
unregulated basins in the western United States were analyzed for trends in year-
to-year variability and persistence. Decadal time-scale changes in streamflow
variability and lag-1 year autocorrelation (persistence) were observed.
“The significance of the variability trends was tested using a jackknife pro-
cedure involving the random resampling of seasonal flows from the historical
record. As shown in Figure 1, the 1930s-1950s was a period of low variability
and high persistence, the 1950s-1970s was a period of low variability and anti-
persistence, and the period after 1980 showed high variability and high persis-
tence. In particular, regions from California and Nevada to southern Idaho, Utah,
and Colorado have recently experienced an unprecedented sequence of consecu-
tive wet years along with multiyear extreme droughts.
“These various streamflow characteristics are not necessarily varying on
the same time scales or coincidentally; increases in variability have preceded
increases in autocorrelation by approximately 5-10 years, which have in turn
preceded increases in skewness by another five years. Nonetheless, the various
phenomena have become ‘in phase,’ making the most recent 20 years the only part
of the record that is highly variable, highly persistent, and highly skewed. This
triple alignment is perhaps the most challenging scenario for water managers.
One possible scenario involves a series of consecutive wet years that overwhelm
reservoirs and inflate stakeholder expectations about the amount of water avail-
able. An extended stretch of dry years exhausts storage reservoirs and does not
give them a chance to recover. Smaller reservoirs that do not have multiple-year
storage capacity would be especially vulnerable. In comparison, individual dry
years interspersed among wet years are tolerable.
“These decadal oscillations also have implications for water supply forecast-
ing. Statistical streamflow forecasting techniques that use persisted spring and
summer streamflow as a predictive variable for next year’s flows will lead the
forecaster astray when the climate regime switches between positive and nega-
tive autocorrelation. The changes in persistence and variability are undoubtedly
linked to changes in precipitation and temperature and not changes in basin char-

APPENDIX D 81
FIGURE 1 A. Time series of the fraction of western U.S. streamflow stations reporting
statistically significant increases (solid) or decreases (dashed) in 20-year moving window
variance compared to the period of record. B. Fraction of stations reporting lag-1 year
autocorrelation of greater than 0.3 (solid) or less than –0.3 (dashed). All data are plotted
at the end year of the 20-year moving window.
acteristics or soil properties. It is unknown at this time whether procedures that

use antecedent autumn streamflow (e.g., September-November) as a predictive
variable to index the effects of soil moisture are also vulnerable to this effect. The
causes of the current triple alignment are unknown.”
Seasonality of Snowpack and Drought in the West

and Recent Events
Recent publications by Mote (2005) highlight the shifts from traditional
snow accumulation during the winter to a trend to warmer spring temperatures

82 APPENDIX D
and declines in springtime snow water equivalent (SWE) in much of the North
American West over the period 1925-2000, especially since mid-century. The
Pacific Northwest has experienced two years of extremely low snowpacks in
the past five years, 2001 and 2005. The 2001 snowpack deficits resulted in the
second-lowest Columbia Basin streamflow on record, and the 2005 snowpacks,
while not as low as 2001 basin-wide, did set new records in the Cascades of
Washington and Oregon. In contrast to 2001 and 2005, an above-average snow-
pack on March 1 fell victim to record warmth and dryness over a two-month
period (Pagano et al., 2004).
Warmer and wetter springs kick-start the growing season, and with poten-
tially low snowpacks this can be problematic if water rights are called later in
the growing season. The shift to earlier spring runoff in the West documented by
Stewart et al. (2004) will pose challenges for water managers through the rest of
this century.
After a six-year drought in the Great Basin, an enormous single-year snow-
pack recharged soil moisture and resulted in significant spring runoff. Is this an
aberration, or will a long-term drought reestablish itself in the region? Can a
probability of occurrence be quantified for the next water year? Can rapid shifts
in climate from abundance to drought be forecasted with reliability? What will
convince users that this can be done?
DATA AVAILABILITY and Requirements for

Monitoring Drought, Climate, and Water Supply
The entire world has been transformed by the Internet. Within the last
20 years canary yellow teletype paper placed on a clipboard is now readily avail-
able for all to download and use. In the West, the SNOTEL data are downloaded
several million times per year, giving customers the ability to develop and run
site-specific or regional models to meet a wide variety of user needs beyond what
can be done by traditional federal or state partners.
In addition to the Internet, affordable computers with spreadsheet and
graphics software can replicate the work done in the past by mainframes.
DATA GAPS
• Does a data gap really count if there is no need for the data (e.g., unpopu-
lated areas, without agriculture)?
• What is a temporal gap? How often should a station report to meet user
needs? Does technology support frequent observations?
• If data are poor, is that a data gap?

APPENDIX D 83
SNOWPACK MONITORING
The SNOTEL dataset (~25 years) is a relatively new dataset compared to
COOP network or paleo data, but it fills a very important vertical data void in
the West. The oldest datasets (snow courses) are monthly or biweekly during the
winter and extend back to the 1930s. Long-term snowpack records are a critical
component of climate change research.
There is a critical need to provide quality control on all SNOTEL data. A
project to provide quality control on SNOTEL temperature data will be completed
in early 2006.
Remote monitoring is not cheap. SNOTEL sites require maintenance annu-
ally, or more often in some areas. A significant computer/communications invest-
ment is also necessary.
Plans to automate 900 manual snow courses with SNOTEL automation are
in place and about a dozen snow courses or new sites/year are automated.
WATER SUPPLY FORECASTING

Statistically based methods are still in use and provide reasonable results
given the calibration dataset. However, statistical methods do not handle late-
season events (heavy spring rains) or early-season forecasts during October-
December due to lack of snowpack.
Improvements are underway to do a better job of visualizing the data and
incorporating new prediction techniques. In addition, simulation modeling can
help distribute flows during critical/extreme hydroclimatic events. Simulation
models may require more real-time middle- and lower-elevation stations to prop-
erly represent hydrological conditions. In any event, “clean datasets” are needed
to calibrate either model.
There is also a need to integrate climate forecasts with hydrological models
to account for climate variability. This will be a challenge since the user com-
munity needs to understand the relationship of uncertainty between the climate
and hydrological forecasts.
SUMMARY COMMENTS
In conclusion, increased climate and streamflow variability present an ever-
growing challenge to those who live in the West. Increasing population and its
affect on land use, the growing need for electricity, environmental concerns, and
the silent stress of a potential long-term drought hover over the desk of every
resource manager. Understanding and learning from recent climatic and hydro-
logical experiences can and will prove valuable in this new century.

84 APPENDIX D
References
Mote, P., A. Hamlet, M. Clark, and D. Lettenmaier. 2005. Declining mountain snowpack in western
North America. Bulletin of the American Meteorological Society 86(1):39-48.
BPA (Bonneville Power Authority). 2003. Pacific Northwest Loads and Resources Study—2003
White Book. (DOE/BP-3559).
Pagano, T., and D. Garen. 2005. A recent increase in western U.S. streamflow variability and persis-
tence. Journal of Hydrometeorology 6(2):173-179.
Pagano, T., P. Pasteris, M. Dettinger, D. Cayan, and K. Redmond. 2004. Water Year 2004: Western
water managers feel the heat. EOS, Transactions, American Geophysical Union, pp. 385-400.
Stewart, I., D. Cayan, and M. Dettinger. 2004. Changes in snowmelt runoff timing in western North
America under a “business as usual” climate change scenario. Climate Change 62:217-232.
Philip Pasteris is a supervisory physical scientist at the National Water and Cli-
mate Service (NWCS) (in the Natural Resources Conservation Service [NRCS])
in Portland, Oregon, where he is responsible for the production and distribu-
tion of water supply forecasts for the western United States and management of
the NRCS National Climate Program. Mr. Pasteris has also held positions as a
supervisory meteorologist at NWCS and senior hydrologist at the Portland River
Forecast Center for the National Weather Service (NOAA). He received his M.S.
in meteorology from the University of Oklahoma.

APPENDIX D 85
MULTIPLE STRESSES AND CLIMATE VARIABILITY:

WHAT IF THE NEXT KATRINA IS A DROUGHT?
William E. Easterling
Pennsylvania State University
The importance of social and environmental processes as preconditions

for drought is well established. Such processes are often mentioned in standard
definitions of drought. What is missing, however, is a whole human-environment
system approach to defining, monitoring, and measuring the frequency and inten-
sity of droughts that takes into account underlying meteorological, hydrological,
ecological, and sociopolitical processes.
For purposes of discussion it is reasonable to assume that droughts are
ultimately triggered by scarcity of expected natural water supply, whether by
precipitation deficiency, change in timing of snowmelt, inadequate groundwater
recharge, or other forms of water shortage caused by climate variability. Multiple
stressors are environmental and social processes or events that combine to dictate
the limits to which an organism/individual, household, ecosystem, community,
or region can absorb water scarcity before incurring unacceptable environmental
or social cost. Stressors that influence the emergence and intensity of droughts
can usefully be categorized as primary, secondary, and tertiary. Primary stressors
are factors that directly influence water supply or demand across a range of
scales. They may be environmental, as in the case of land cover change that
increases evapotranspiration, or social, as in the case of rapid population growth
or growth of water-intensive industry. Secondary stressors create conditions that
abet vulnerability to water shortage, such as rural depopulation, decreasing agri-
cultural comparative advantage, or high dependence on river-borne transporta-
tion. Tertiary stressors inhibit resiliency or adaptive capacity with respect to water
deficiency. They also may be either environmental, as in crop and ecological
diversity, or social, as in availability of risk management institutions, such as
insurance networks and contingency planning. Adequate assessment of drought
potential must consider not only climatological factors but also the full suite of
interacting primary, secondary, and tertiary stressors.
Although anything but a drought, Hurricane Katrina clearly demonstrates
the importance of multiple stressors interacting with a strong meteorological
event. Katrina was the perfect storm more because of how it combined with a
remarkable set of multiple stressors than its pure thermodynamic energy. It was
a Category 4 (out of 5) hurricane that struck a city that arguably could not have
been constructed to be more in harm’s way. In addition to its elevation below sea
level and the proven inadequacies of its levee and water pumping systems, the
greater New Orleans region had become a critical port of entry for a significant
percentage of the nation’s imports, including natural gas, oil, and subsequent
refinery products. Much of the midwestern grain produced for export passes

86 APPENDIX D
through the port of New Orleans. This guaranteed that the impacts of Katrina
would extend to the nation and the world. The large income divide separating
New Orleans’ wealthy and poor citizens left a large impoverished population
especially vulnerable to the storm and its aftermath simply because it lacked
the means to evacuate. Moreover, the translation of scientific assessments of
hurricane vulnerabilities into practical political decisions did not happen. The list
of multiple stresses at work in New Orleans was large, and their synergism with
themselves and the hurricane surely intensified the loss of life and property and
propagated impacts well beyond the region hit by the storm. An interesting ques-
tion to ask is, What if Katrina had been the drought equivalent of a Category 4
hurricane occurring throughout the Mississippi River Basin—a sort of Dust Bowl
II? What would we want to know about processes of social and environmental
change that exacerbate precipitation deficiency? To try to answer this question
might shed light on what we know and what we do not know about how mul-
tiple stressors might interact with a severe drought and in the process point out
important research gaps.
Precipitation averaged about 20 percent lower and temperatures about 1°C
higher than current during the decade of the 1930s in the central and western
Great Plains. Were such an event to recur today, there are a number of stressors
that likely would amplify the environmental and societal impacts. (In fairness,
there are also improvements in resiliency due to learning from previous droughts
that might provide some protection from a recurrence of the Dust Bowl droughts.)
Some of the key stressors that intensified the destruction and damage of Katrina,
ironically, would intensify the hardship of a long, severe drought. Table 1 lists
a few examples of environmental and social situations and changes that would
almost be certain to intensify the impacts of a superdrought in the Mississippi
River Basin (MRB). As pointed out below, the reliance of the MRB on primary
commodities and their water-borne transport renders the region vulnerable to any
kind of climatic fluctuation that disrupts.
I am not aware of research that has explored how trends in multiple stressors,
such as those listed in Table 1, affect the frequency or intensity of droughts. The
closest vein of research is exemplified by O’Brien et al. (2004), who examined the
vulnerability of Indian agriculture to climate variability and a small set of global
stressors to determine the effects of being “double exposed.” However, common
sense suggests that rapid changes in one or more stressors that outstrip existing
capacity to adapt to water shortage must, ipso facto, increase the frequency of dry
events that become droughts. For example, the volume of barge traffic hauling
corn down the Mississippi River to Louisiana for export increased at an annual
average rate of 3.5 percent during the period 1972-1992. At the same time, the
Pick-Sloan Missouri Basin Program calls for increasing water retention in the
Missouri’s Upper Basin to meet hydroelectric and environmental needs, thus cut-
ting flows to the Lower Basin and Mississippi River. This nonlinear increase in
barge traffic combined with less flow from the Missouri River during dry spells

APPENDIX D 87
TABLE 1 Examples of Primary, Secondary, and Tertiary Stressors of

Mississippi River Basin (MRB) Drought
Primary Stressors
• Basin-wide soil loss and subsequent sedimentation of navigation channels
• Over-pumping of High Plains aquifer leading to groundwater decline
• Increase in urban and municipal water demand caused by urban growth
• Water retention in major tributaries for irrigation, hydroelectric production, and in-
stream environmental values
• Climate change
Secondary Stressors
• Invasive species, especially agricultural pathogens such as soybean rust in Gulf states
• Increasing volume of water-borne shipping of grain commodities
• Increasing debt:asset ratios on farms and crop price volatility combined with loss of
comparative advantage
• Rising energy prices
• Diminished water quality
• Conflicting interstate water use policies (i.e., Pick-Sloan on the Missouri River Basin,
Wyoming-Nebraska disputes over Platte River allocations)
• Aging of rural population
Tertiary Stressors
• Structural dependence on primary commodity industries
• Crop price support policies
• Alternative commodity transport systems
• Inadequate engineering designs of infrastructure, including channelization impacts on
delta subsidence/inundation
suggests greater vulnerability of barge traffic, particularly in the lower MRB to

once-minor low-flow events that now might constitute bona fide drought condi-
tions. A recurrence of the 1930s droughts could be devastating to MRB agricul-
ture, due to the direct impact on crop productivity and the diminished capacity to
move grain to markets cheaply.
Multiple stresses must be accounted for in a comprehensive assessment of
the potential frequency and magnitude of droughts. There are several research
gaps that need to be addressed in this regard:
1. Improved understanding of how multiple stresses interact, both with

themselves and climate variability, to influence vulnerability to water shortage;
2. Improved understanding of how adaptive capacity is influenced by
multiple stresses;
3. Improved understanding of how multiple stresses and coping systems
interact across different levels of scale (in space and time);
4. Using the understanding gained from 1, 2, and 3, whole human-environment
analytical frameworks deployed with integrated assessment models, suites of
leading indicators, and other comprehensive analytical methods are needed—the

88 APPENDIX D
USAID Famine Early Warning System (FEWS) could serve as a reasonable

model;
5. Identification of coping system strengths (i.e., redundant response systems)
and weaknesses (i.e., overburdened response systems); and
6. Stronger stakeholder involvement in whole human-environment system
assessments to minimize exclusion of science-based risk analysis from applied
risk management.
REFERENCE
O’Brien, K. L., R. Leichenko, U. Kelkar, H. Venema, G. Aandahl, H. Tompkins, A. Javed, S.
Bhadwal, S. Barg, L. Nygaard, and J. West. 2004. Mapping vulnerability to multiple stressors:
Climate change and globalization in India. Global Environmental Change: Human and Policy
Dimensions 14(4):303-313.
William E. Easterling is the director of the Penn State Institutes of the Environ-
ment and professor of geography with a courtesy appointment in agronomy at
the Pennsylvania State University. Dr. Easterling’s research interests include the
potential for agriculture in developed and developing countries to adapt to climate
variability and change; the role of scale in understanding the vulnerability of com-
plex systems; how land use change may influence the uptake and release of carbon
in the terrestrial biosphere; the use of experimental long-term climate forecasts
to assist decision making under conditions of uncertainty; and the development of
methodologies for detecting the impacts of observed 20th-century climate change
on natural and managed ecosystems. Dr. Easterling received his Ph.D. in geogra-
phy from the University of North Carolina at Chapel Hill.

APPENDIX D 89
DROUGHT-INDUCED VEGETATION MORTALITY AND

ASSOCIATED ECOSYSTEM RESPONSES:
EXAMPLES FROM SEMIARID
WOODLAND AND FORESTS
David D. Breshears
University of Arizona
Woody plant mosaics are a key attribute of ecosystems. A large portion of

the terrestrial biosphere can be viewed as lying within a continuum of increasing
coverage by woody plants (shrubs and trees), ranging from grasslands with no
woody plants to forests with nearly complete closure and coverage by woody
plants (Breshears and Barnes, 1999; Breshears, 2006). The characteristics of
woody plants determine fundamental descriptors of vegetation types, includ-
ing grassland, shrubland, savanna, woodland, and forest. Because woody plants
fundamentally affect many key aspects of energy, water, and biogeochemical
patterns and processes, changes in woody plant cover are of particular concern
(Breshears, 2006).
Drought can cause rapid changes in vegetation by triggering woody plant

mortality. Assessments of potential global change impacts initially focused on
how vegetation types matched given climatic envelopes. Later focus turned to
how vegetation patterns might migrate with changing climate, focusing on rates
of plant establishment. More recently, the importance of drought-induced die-off
of woody plants was highlighted as a major dynamic response to climate variation
and change. In particular, ecotones have been noted as areas where changes in
vegetation in response to climate ought to be most rapid and responsive, as high-
lighted by a case study of vegetation response to drought during the 1950s (Allen
and Breshears, 1998). In response to a severe drought in the southwestern United
States during the 1950s, ponderosa pine (Pinus pondersosa) trees at lower, drier
sites died, resulting in a shift of the ponderosa pine forest/piñon-juniper woodland
ecotone of more than 2 km in less than five years (Figure 1) and producing a rapid
change in vegetation cover (Figure 2). Similarly, within the distributional range
for piñon pine (Pinus edulis), many trees at lower, drier sites within also died.
Drought can trigger widespread tree mortality across a region. Although

tree mortality almost certainly occurred across much of the Southwest in response
to the 1950s drought (and probably for previous regional-scale droughts as well),
few studies exist that allow scientists assessing impacts of drought to test predic-
tions about the rapidity and extent of vegetation die-off response to drought. A
recent drought beginning around the new millennium impacted the southwestern
United States and was the most severe since that of the 1950s. Mortality of several
species was observed throughout the Southwest. Mortality of piñon pine spanned

90 APPENDIX D
FIGURE 1 An ectone shift. Changes in vegetation cover between 1954 and 1963 in
Bandelier National Monument in northern New Mexico, showing persistent ponderosa
pine forest (365 ha), persistent piñon-juniper woodland (1,527 ha), and the ecotone shift
zone (486 ha) where forest changed to woodland in response to the 1950s drought (from
Allen and Breshears, 1998).
50
40
Forest Area (%)
30
20
10
0
1930 1940 1950 1960 1970 1980
Year
FIGURE 2 A rapid reduction in forest cover. Changes in percent forest cover between
1935 and 1975 in Bandelier National Monument in northern New Mexico. The arrow
indicates the time of historical observations of extensive tree mortality (from Allen and
Breshears, 1998).
Appendix D Breshears
Fig 2
APPENDIX D 91
major portions of the species’ range, with substantial die-off occurring over at
least 12,000 km2 (Breshears et al., 2005; Figure 3). For both droughts, die-off was
related to bark beetle infestations, but the underlying cause of die-off appears to
be water stress associated with the drought.
Drought-induced tree mortality might be exacerbated under higher tem-

peratures. The recent drought in the southwestern United States that triggered
regional-scale die-off of piñon pine across the Southwest was not as dry as the
previous regional drought of the 1950s (Breshears et al., 2005; Figure 4). How-
ever, the recent drought was hotter than the 1950s drought by several metrics,
including mean, maximum, minimum, and summer (June-July) mean tempera-
ture. Tree mortality in response to the recent drought appears to have been more
severe than that of the previous drought. In addition to die-off occurring across
the region, the limited available data suggest that extensive piñon pine mortality
FIGURE 3 Regional drought-induced vegetation changes. Change map for Normalized

Difference Vegetation Index (NDVI) for region encompassing P. edulis distribution within
Arizona, New Mexico, Colorado, and Utah, based on deviation from 2002-2003 relative
to the predrought mean (1989-1999) during the period late May to June (Breshears et al.,
2005). Changes in NDVI were linked to changes in foliar water content and plant water
potential (Stimson et al., 2005) and to tree mortality at an intensively studied site within
the region (from Breshears et al., 2005).

92 APPENDIX D
FIGURE 4 A global-change-type drought. Southwestern North American climate. Annual

mean precipitation (mm) and average of maximum and minimum temperatures ( oC) for
(A) all stations in the four-state region and (B) only stations in or near piñon-juniper
woodlands within that region. Appen
AssociatedD(C) maximum andfig
Breshears (D)4.d
minimum temperatures
for piñon-juniper woodlands (dotted line: long-term mean; dashed line: 10th or 90th
percentile, differentiating driest or hottest years). Shaded bands are the four consecu-
tive driest years of the 1950s drought (1953-1956) and the recent drought (2000-2003).
Compared to the 1950s drought, the recent drought was wetter (P < 0.05) but warmer for
maximum (P < 0.05) and especially minimum temperature (P < 0.001; from Breshears
et al., 2005).

APPENDIX D 93
occurred at upper-elevation wetter sites in response to the recent drought but not
in response to the 1950s drought. Hence, the warmer temperatures associated
with the recent drought may have produced more extensive tree die-off. Because
global change is projected to yield droughts under warmer conditions—referred
to as global-change-type drought—the die-off from the recent drought may be a
harbinger of vegetation response to future global-change-type drought (Breshears
et al., 2005).
Several other changes can accompany die-off of dominant overstory trees. In

addition to the die-off of the dominant overstory tree species, other species under-
went mortality in response to regional drought (Allen and Breshears, 1998; Bres-
hears et al., 2005). These include juniper (Juniperus monosperma), a co-dominant
with piñon pine for much of its range, and blue grama (Bouteloua gracilis), the
dominant herbaceous species for many of these systems. Additionally, reductions
in ground cover may contribute to an increase in erosion rates (Davenport et al.,
1998; Wilcox et al., 2003). In particular, reductions in herbaceous ground cover
might trigger a nonlinear increase in soil erosion once a threshold of herbaceous
cover has been crossed. In addition, reductions in tree canopy cover can dramati-
cally alter the distribution of near-ground energy (Martens et al., 2000). There-
fore, die-off of overstory vegetation affects numerous key ecosystem processes
that are dependent on incoming energy (Breshears, 2006).
Drought-induced fire also triggers rapid canopy change and high soil ero-
sion rates. Drought patterns can also trigger larger-scale fire patterns (Swetnam
and Betancourt, 1998). Crown fire within woodlands and forests also can cause
large reductions in tree canopy cover. Additionally, soil erosion can increase
dramatically following forest wildfire (Johansen et al., 2001). The combined
impacts of fire and drought-induced tree mortality are highlighted by the major
changes in woodland and forest vegetation that have occurred in northern New
Mexico over the past 50 years (Breshears and Allen, 2002; Breshears et al., 2005).
It will be at least several decades following one of these types of disturbances
before reestablishment of similar tree canopy cover in semiarid woodlands and
forests could occur.
Interactions among multiple effects of drought, including potential eco

system cascades, remain major uncertainties requiring future research. Exam-
ples of drought-induced tree die-off in semiarid woodlands and forests highlight
the rapidity and extensiveness with which drought can trigger vegetation change.
Several nonlinear or threshold-like processes may occur and require improved
prediction, including tree mortality, energy and water budget changes, and soil
erosion thresholds. Systems can cascade through multiple states. For example,
a location that had extensive ponderosa pine mortality in the 1950s had little
reestablishment of ponderosa pine over the subsequent 50 years (Allen and

94 APPENDIX D
reshears, 1998) and was within the region exhibiting extensive piñon pine
B
mortality in 2002-2003 (Breshears et al., 2005); rates of soil erosion following
the 1950s drought were and remain high. An ability to predict tree mortality, asso-
ciated ecosystem responses, and effects on the carbon budget and on other eco
system goods and services should be a high priority for future research (Breshears
and Allen, 2002; Millennium Ecosystem Assessment, 2005).
Acknowledgments: I thank my coauthors on previous related publications

for their insights on and contributions to concepts related to those presented here:
C. D. Allen, N. S. Cobb, P. M. Rich, K. P. Price, R. G. Balice, W. H. Romme, J. H.
Kastens, M. Lisa Floyd, J. Belnap, J. J. Anderson, O. B. Myers, C. W. Myers, B.
P. Wilcox, S. N. Martens, F. J. Barnes, S. L. Ustin, H. C. Stimson and D. W. Dav-
enport. Support was provided by Los Alamos National Laboratory and National
Science Foundation (DIREnet:DEB-0443526 and SAHRA:EAR-9876800).
References
Allen, C. D., and D. D. Breshears. 1998. Drought-induced shift of a forest-woodland ecotone: Rapid
landscape response to climate variation. Proceedings of the National Academy of Sciences
U.S.A. 95:14839-14842.
Breshears, D. D. 2006. The grassland-forest continuum: Trends in ecosystem properties for woody
plant mosaics? Frontiers in Ecology and the Environment 4:96-104.
Breshears, D. D., and F. J. Barnes. 1999. Interrelationships between plant functional types and soil
moisture heterogeneity for semiarid landscapes within the grassland/forest continuum: A unified
conceptual model. Landscape Ecology 14:465-478.
Breshears, D. D., and C. D. Allen. 2002. The importance of rapid, disturbance-induced losses in
carbon management and sequestration. Ecological Sounding. Global Ecology and Biogeography
11:1-5.
Breshears, D. D., N. S. Cobb, P. M. Rich, K. P. Price, C. D. Allen, R. G. Balice, W. H. Romme, J. H.
Kastens, M. L. Floyd, J. Belnap, J. J. Anderson, O. B. Myers, and C. W. Meyer. 2005. Regional
vegetation die-off in response to global-change type drought. Proceedings of the National Acad-
emy of Sciences U.S.A. 102:15144-15148.
Collins, W. J., R. G. Derwent, C. E. Johnson, and D. S. Stevenson. 2002. The oxidation of
organic compounds in the troposphere and their global warming potentials. Climatic Change
52:453‑479.
Davenport, D. W., D. D. Breshears, B. P. Wilcox, and C. D. Allen. 1998. Sustainability of piñon-
juniper woodlands—a unifying perspective of soil erosion thresholds. Viewpoint. Journal of
Range Management 51:231-240.
Johansen, M. P., T. E. Hakonson, and D. D. Breshears. 2001. Post-fire runoff and erosion following
rainfall simulation: Contrasting forests with shrublands and grasslands. Hydrological Processes
15:2953-2965.
Millennium Ecosystem Assessment. 2005. Ecosystems and human well-being: Synthesis reports.
Stimson, H. C., D. D. Breshears, S. L. Ustin, and S. C. Kefauver. 2005. Spectral sensing of foliar
water conditions in two co-occurring conifer species: Pinus edulis and Juniperus monosperma.
Remote Sensing of Environment 96:108-118.
Swetnam, T. W., and J. L. Betancourt. 1998. Mesoscale disturbance and ecological response to
decadal climatic variability in the American Southwest. Journal of Climate 11:3128-3147.

APPENDIX D 95
Wilcox, B. P., D. D. Breshears, and C. D. Allen. 2003. Ecohydrology of a resource-conserving semi-

arid woodland: effects of scale and disturbance. Ecological Monographs 73:223-239.
David D. Breshears is a professor at the University of Arizona in the School

of Natural Resources and Ecosystem Sciences Theme Leader for the Institute
for the Study of Planet Earth and has a joint appointment in the Department
of Ecology and Evolutionary Biology. His research focuses on gradients of
vegetation ranging from grassland through forest, vegetation dynamics, includ-
ing drought-induced die-off, ecological-hydrological dynamics (ecohydrology),
wind and water erosion, soil carbon measurement technology, and applications
addressing land use, contaminant risks, and global change. He received his Ph.D.
from Colorado State University.

96 APPENDIX D
UNDERSTANDING AND MANAGING

MULTIPLE STRESSES IN THE CONTEXT OF A
COMPLEX RIVER BASIN:
THE COLORADO RIVER
Roger S. Pulwarty
NOAA/CIRES
Watersheds, Multiple Stresses, and Streams of Thought

Societies are always adapting incrementally and in diverse ways to a variety
of integrated and cumulative changes. There is, however, little understanding
of the long-term and widespread consequences of these adaptations at different
levels of governance.
The “regional scale” has long been advocated as a useful organizational
unit on which to coordinate and evaluate socially relevant research cognizant
of geophysical, cultural, and jurisdictional boundaries. Yet attempts to manage
consistent regional units of analysis, such as the watershed, have not met expecta-
tions. Difficulties arise in managing for particular outcomes given multiple con-
tending perspectives and the uncertainties in variations and interactions between
physical and ecological environments. The western United States offers and has
offered unique opportunities for identifying lessons for strategic learning about
the management of cross-scale environmental risks over time, particularly those
associated with water. For example, droughts have played a major role in the
evolution of western water institutions. Most notably, the droughts of 1865-1872
gave rise to prior appropriation law.
Gilbert White identified the major elements of integrated river basin develop-
ment as follows: (1) multiple-purpose storage reservoirs, (2) basin-wide planning,
and (3) comprehensive regional development. Studies of the first-order impacts
of climate on each of these elements in the western United States indicate that
vulnerability exists in the areas of storage and consumptive depletions versus
renewable supply. Critical factors confronting sustainable resource use across
western basins can be summarized under the following headings: population and
consumption, water quality, environmental water allocation, uncertain reserved
water rights, groundwater overdraft, outmoded institutions, aging urban water
infrastructure, and evolving federal, state, and local relationships. Responses have
included water banking, inter-basin transfers, advanced decision-support/expert
systems, streamflow and demand forecasting, drought management programs,
threshold indicators, and efficiency improvements.
The complications of changes in the spatial and temporal distribution of rain-
fall, soil moisture, runoff, frequency, and magnitudes of droughts and floods have
not been explicitly included in response planning. Systems design, operational

APPENDIX D 97
inflexibility, and legal and institutional constraints also reduce the adaptability
of water systems and confound most recommendations to date on responding
to climate change. Potential water-resource-related focusing events across the
western United States include
(1) extreme and sustained climatic conditions (e.g., drought and floods);
(2) large-scale inter-basin transfers;
(3) quantification of tribal water rights;
(4) an energy crisis;
(5) changing transboundary responsibilities; and
(6) regulatory mandates such as the Endangered Species and Clean Water
acts.
Critical thresholds arise when buffers are diminished and/or response curves
steepen. A conspicuous aspect of water resources management has been the lack
of careful post-audits of the social and economic consequences of previous pro-
grams and projects in the context of background variability and change. Three
kinds of assessment questions may be asked: (1) What is known about the effects
of past development programs on the environment? (2) What are, and how effec-
tive are, present programs (and their associated assumptions) in the context of a
varying environment? (3) What appear to be the principal future effects of alter-
native adjustments? In this presentation we explore the above questions in one
western basin, the Colorado, in which all of the above issues are exemplified.
Case Study: The Colorado River Basin

The Colorado River supplies much of the water needs of seven U.S. states,
two Mexican states, and 34 Native American tribes, representing a population
of 25 million inhabitants with a projection of 38 million by the year 2020. The
Colorado does not discharge a large volume of water. Because of the scale of
impoundments and withdrawals relative to its flow, the Colorado has been called
the most legislated and managed river in the world. It has also been called the
most “cussed” and “discussed” river. As has been well documented, the most
important management agreement (the Colorado River Compact of 1922) was
based on overestimation of the reliable average annual supply of water (estimated
at 16.4 million acre feet) due to a short observational record. Colorado River
streamflow however exhibits strong decadal and longer variations. The Colorado
system also exhibits the characteristics of a heavily over-allocated or “closed
water system.” In such systems, development of mechanisms to allow resource
users to acknowledge interdependence and to engage in negotiations and agree-
ments is not only desirable but also necessary. Climate and weather events form
a variable background on which these agreements and conflicts are played out.
In this context institutional conditions that limit flexibility tend to exacerbate the

98 APPENDIX D
underlying resource issues. This presentation describes how lessons from past
events and new climate information on the Colorado River Basin inform or do
not inform integrated watershed and adaptive management programs intended to
preserve and enhance physical, economic, cultural, and environmental values.
It begins with an overview of the history of Colorado Basin development and
the scales of decision making involved. The decision-making environments are
discussed in terms of critical climate-sensitive issues, including interbasin trans-
fers and transboundary responsibilities, Native American rights, environmental
requirements, and state water issues.
The Colorado system has experienced drought conditions in six of the last
seven years. Until the last few years, the expectation of Colorado River managers
was that significant shortages in the Lower Basin would not occur until after
2030. Events such as the drought expose critically vulnerable conditions and,
though they warn of potential crisis, they also are opportunities for innovation.
Historically, reservoirs and inter-basin transfers have been used to mitigate the
effects of short-term drought in the Colorado Basin. The lessons and impacts
of these adjustment strategies and more recent settlement agreements are still
being gathered. The system’s ability to maintain reliable supply during periods of
severe long-term droughts of >10 years (the timescales of development, project
implementation, and ecosystem management efforts), known to have occurred in
the West over the past 1,000 years, is as yet untested but may be so in the very
near future. While recent modeling studies project up to an 18 percent decrease
in runoff in the basin under climate change scenarios, just the continuation of
drought over the next year will likely induce crisis conditions. Thus the “normal”
situation is critical. In the semiarid Southwest, even relatively small changes
in precipitation can have large impacts on water supplies. Even in areas where
integrated approaches are adopted, cooperation remains mainly crisis driven,
inhibiting iterative, long-term collaboration and learning. While opportunities for
“win-win” situations and rule changes exist, such changes are extremely difficult
to implement. In this context institutional conditions that limit flexibility tend to
exacerbate the underlying resource issues.
Opportunities for Learning and Decision Making

Under Uncertainty
Learning (and the capacity for employing lessons learned) is of strategic
importance in the decades-long process of adapting to global changes, including
climatic variations. Even when physical effects or projections can be established
with fair confidence, there usually exist large uncertainties about biological and
ecological effects and even greater uncertainties with respect to social conse-
quences. Much work and experience has shown that long-term environmental
problems can seldom be dealt with by single discrete actions or policies but
respond only to continuing, sustained efforts at learning, supported by steady

APPENDIX D 99
public attention and visibility. Focusing events provide opportunities for learning.
In the West potential water-resource-related focusing events include:
(1) extreme climatic conditions (e.g., drought and floods);

(2) large-scale inter-basin transfers;
(3) quantification of tribal water rights;
(4) an energy crisis;
(5) changing transboundary responsibilities; and
(6) regulatory mandates such as the Endangered Species and Clean Water
acts.
Crisis conditions can be said to be reached when focusing events occur concur-
rently with awareness of a finite time necessary for response. As mentioned
above, for many basins in the West the normal situation is critical, and relatively
small environmental changes can exceed social thresholds of acceptability and
reliability.
Opportunities for learning also arise from deliberate perturbation (e.g., high
flow releases) of a system to stimulate monitoring and learning. The idea of
“adaptive management” has been widely advocated as a bridge between science
and policy with a specific focus on ecosystems. This presentation explores the
idea in the context of climatic and other uncertainties but grounds the discus-
sion in the implementation of an actual adaptive management program in the
Colorado. Adaptive management has three key tenets: (1) policies are experi-
ments that should be designed to produce usable lessons; (2) it should operate
on scales compatible with natural processes, recognizing social and economic
viability within functioning ecosystems; and (3) it is realized through effective
partnerships among private, local, state, tribal, and federal interests. In a water-
shed setting this can mean balancing hydropower production, habitat manage-
ment, conservation, endangered species recovery, and cultural resources in order
to experiment, learn, incorporate learning, and adapt—a decidedly idealized view.
Each component carries its own type and sources of uncertainty. One goal is to
identify the strengths and weaknesses of an “adaptive management approach” for
mitigating drought risks in the context of changing climatic baselines and early
warning in association with critical thresholds.
Conclusions: Implications for Decision Making

There is increasing awareness that we are engaged in (1) questions about
the nature and role of integrated knowledge and uncertainty in complex settings
and (2) a social process of risk communication and perception, as opposed to the
simple development and dissemination of risk information or even a client-driven
“two-way” process. The experience of development in the Colorado in the face
of environmental uncertainty clearly illustrates that impacts and interventions

100 APPENDIX D
can reverberate through systems in ways that can only be partially traced and
predicted. In addition, adjustments and responses in the short term can increase
vulnerability over the long term. The discussion here is based on the premise that
understanding how effectively society might identify common goals, best use
climatic and other information, and prepare for the consequences of future varia-
tions and surprises requires identification and evaluation of present systematic
efforts (i.e., field-tested alternatives) to experiment, characterize uncertainties,
make decisions, and cope with environmental variability across temporal and
spatial scales. If lessons learned are to be applied, then a large part of the scien-
tific goal should be to inform processes that can decrease impediments to the flow
of information and innovations. This would entail:
(1) clarification of management goals at the human-environment interface;

(2) construction of a cooperative foundation between research and
management;
(3) distillation of lessons from comparative appraisals of current and past
practices;
(4) understanding and assessing adaptive capacity;
(5) characterizing and communicating uncertainties for both minimizing and
managing risk; and
(6) developing effective criteria for validity and acceptability (i.e., robust
information in research as well as practical contexts).
In this light a “seamless suite” of products and services for drought risk
assessment and management, from national through local, may not be optimal in
practice, especially if the goal is improvement of social welfare or at least inform-
ing the implementation of better decisions.
Roger S. Pulwarty is a physical scientist in the NOAA/CIRES/Climate Diagnostics

Center in Boulder, Colorado. Dr. Pulwarty’s interests are in the role of climate
and weather in society-environment interactions and the design of public services
to address associated risks. His work has focused on (1) hydroclimatic variability
and change; (2) social vulnerability and responses to environmental variations;
and (3) the role and use of research-based information in natural resources
policy and decision making in the western United States, Latin America, and the
Caribbean.

APPENDIX D 101
ECOSYSTEM-CLIMATE FEEDBACKS STUDIED WITH

AN INTEGRATED GLOBAL SYSTEM MODEL
Ronald G. Prinn
Massachusetts Institute of Technology
The overall goal of our ecosystem-climate research program is to charac-

terize and quantify the feedback mechanisms between terrestrial ecosystems,
the climatic system, and air pollution involving the cycles of water, energy, and
relevant chemical species. To address this goal we are developing and using
an Integrated Global System Model (IGSM) that includes (1) dynamic and
linked terrestrial hydrology and ecology, including the MBL Terrestrial Eco-
systems Model (TEM); (2) comprehensive coupled physical climate (MIT two-
dimensional atmosphere and three-dimensional ocean); and (3) MIT atmospheric
chemistry (gaseous and aqueous phase chemical processes). The IGSM also
includes a detailed global economics model, including emissions from industrial
and agricultural activity. The integrated models represent the major complex
biological systems on earth and span the scales from local to global.
With our coupled models we have quantified the combined effects of air
pollution (O3), rising CO2, and climate change on the productivity and distribution
of vegetation globally. We have also determined how changes in land ecosystems,
caused by pollution and climate change, can feed back to climate through changes
in carbon storage. We have calculated past and future carbon dioxide and methane
fluxes from northern high latitudes under the joint influence of rising CO 2 levels
and rising Arctic temperatures. We have also computed substantial changes in
soil N2O emissions under the joint influences of changing temperatures, rain-
fall, and soil carbon. As the impacts of multiple stressors acting simultane-
ously on forest production are determined, their roles in amplifying or damping
regional disparities are being elucidated. Toward these ends we have improved
our Terrestrial Ecosystem Model and Atmospheric Chemistry Models to facilitate
their interaction. We have also further developed our 3D Ocean Circulation Model
to incorporate biogeochemical cycles. We also adapt the NCAR MATCH, CCM3,
and CAM3 3D models for selected atmospheric chemistry studies. Our work is
providing significant information for understanding how our future global envi-
ronment will evolve under the joint effects of growing world population, changing
technological and agricultural practices, and economic development. We argue
that uncertainties in most of the relevant feedback processes are large. Therefore,
to understand the above interactions we include comprehensive studies of the
sensitivity of our conclusions to critical input assumptions, and calculations of
the probability distributions of critical output variables. We address this through
the use of multiple (ensemble) simulations, flexible models, and powerful (proba-
bilistic collocation) methods to compute uncertainties. There are significant edu-
cational by-products of this research designed to effectively communicate results

102 APPENDIX D
to students, fellow researchers, journalists, industry, and environmental policy

makers. We are also making contributions to general methodologies to study
and numerically simulate very complex and interactive spatially and temporally
resolved phenomena using distributed memory computers. Much of our research
has already appeared in multiple papers in Journal of Geophysical Research,
Tellus, Geophysical Research Letters, Climatic Change, Global Biogeochemical
Cycles, Journal of Vegetation Science, Journal of Climate, and other journals, and
several papers are under review. Public access to our research, including reports
and journal reprints, is available through our extensive websites: Joint Program
on the Science and Policy of Global Change (http://mit.edu/globalchange/) and
Center for Global Change Science (http://mit.edu/cgcs/).
Here we present the results of three of the projects.
1. INFLUENCE OF CLIMATE AND AIR POLLUTION ON

ECOSYSTEM CARBON FLUXES
Several environmental factors influence carbon sequestration in natural ter-
restrial ecosystems, including climate variability and change, atmospheric carbon
dioxide concentrations, ozone pollution, and atmospheric nitrogen deposition. To
explore the relative importance of these factors on historical carbon sequestra-
tion, we have conducted a series of global simulations with a modified version
of the Terrestrial Ecosystem Model. Model modifications include a more detailed
representation of soil nitrogen pools and fluxes to better account for the influence
of nitrogen deposition, nitrogen fixation, trace N gas emissions, and leaching
losses of nitrate and dissolved organic nitrogen on terrestrial carbon and nitrogen
dynamics. Initial results indicate that natural terrestrial ecosystems accumulated
54.8 Pg C during the 20th century as a result of CO2 fertilization (39.2 Pg C),
atmospheric nitrogen deposition (19.7 Pg C), and climate variability and change
(3.1 Pg C); ozone pollution reduced the potential carbon sequestration benefits
of these other factors by 7.2 Pg C. Over this time period, the rate of carbon accu-
mulation increased from about 1.0 Pg C per year in the 1900s to 1.6 Pg C per
year in the 1990s. Carbon sequestration is not uniformly distributed across the
globe. Preliminary analyses also suggest that carbon losses associated with land
use change over this time period would substantially reduce our estimate of car-
bon sequestration. Looking to the future, lowering of ozone levels resulting from
future air pollution policies is estimated to increase carbon uptake by terrestrial
ecosystems. To better account for the effects of human and natural disturbances,
we are currently developing datasets and revising algorithms to consider the
effects of agriculture, fire, logging, and insect infestations on terrestrial carbon
and nitrogen dynamics.

APPENDIX D 103
2. CLIMATIC INFLUENCES ON METHANE SURFACE FLUXES

We have estimated fluxes by combining observations and models using
(1) methane observations: high-frequency (13: AGAGE, CMDL, etc.) and Flask
(41 comprehensive and 32 more intermittent: CMDL, CSIRO, etc.) monthly
mean observations between 1996 and 2001; and (2) Global 3D MATCH model:
interannually varying transport (NCEP) used with 1.8 deg. × 1.8 deg. resolution
and 28 levels to create the CH4 response of each site to monthly pulses from indi-
vidual regional processes (sensitivity matrix). Using an annually repeating time/
space-varying model OH tuned to AGAGE CH3CCl3 observations, the Kalman
Filter is used to solve for seven seasonally varying processes (three wetland,
three biomass burning, rice) as monthly varying fluxes; and two pseudo-steady
processes (animals and water, coal and gas) as constant fluxes. Deduced interan-
nual variability (monthly anomalies) is large with a 32-33 Tg/yr total emissions
increase in 1998 coinciding with El Niño and global wildfires. Northern/tropical
wetlands and rice region emissions dominate the total variability. Rice areas
(including proximal wetlands) are responsible for 8-17 Tg/yr of this. Wetlands
dominate the remainder, but boreal fires in Siberia may have also contributed to
our deduced strong northern wetlands increase. Compared to previous estimates,
energy-related emissions are smaller (decrease in Russia?) and emissions from
rice-growing regions are larger (proximal forests or wetlands?). The computed
seasonal flux cycles capture the expected seasonal cycles (but rice growing peaks
earlier).
We have also used TEM to study how rates of methane (CH4) emissions and
consumption in high-latitude soils of the Northern Hemisphere have changed
over the past century in response to observed changes in the region’s climate. We
estimate that the net emissions of CH4 (emissions minus consumption) from these
soils have increased by an average .08 Tg CH4 yr–1 during the 20th century. Our
estimate of the annual net emission rate at the end of the century for the region is
51 Tg CH4 yr–1. Russia, Canada, and Alaska are the major CH4 regional sources
to the atmosphere, responsible for 64 percent, 11 percent, and 7 percent of these
net emissions, respectively. Our simulations indicate that large interannual vari-
ability in net CH4 emissions occurred over the last century. Our analyses of the
responses of net CH4 emissions to the past climate change suggest that future
global warming will increase net CH4 emissions from the Pan-Arctic region.
The higher net CH4 emissions may increase atmospheric CH4 concentrations to
provide a major positive feedback to the climate system.
3. CLIMATIC AND NUTRIENT IMPACTS ON

NITROUS OXIDE EMISSIONS
Natural terrestrial fluxes of N2O from soils are important contributors to
the global budget of this greenhouse gas. The IGSM incorporates the global
Natural Emissions Model (NEM) for soil biogenic N2O emissions, which has

104 APPENDIX D
2.5° × 2.5° spatial resolution. It is a process-oriented biogeochemical model

including the processes for decomposition, nitrification, and dentrification. The
model takes into account the spatial and temporal variability of the driving vari-
ables, which include soil texture, vegetation type, total soil organic carbon, and
climate parameters. Climatic influences, particularly temperature and precipita-
tion, determine dynamic soil temperature and moisture profiles and shifts of
aerobic-anaerobic conditions. The major biogeochemical processes included in
the model are decomposition, nitrification, ammonium and nitrate absorption and
leaching, ammonia emission, and denitrification. For present-day climate and soil
datasets, NEM predicts an annual flux of 11.3 Tg-N (17.8 Tg N2O). NEM predicts
large emissions from tropical soils, which is qualitatively consistent with the
observed latitudinal gradient for N2O, and in situ flux measurements. Predicted
emissions of N2O from runs of the NEM through 2100 indicate significant sen-
sitivity to outputs from the climate (temperature, precipitation) and TEM (total
soil carbon) models. Two NEM runs driven by climate outputs only and climate
plus TEM outputs indicate that climate and soil carbon changes contribute about
equally to the predicted very significant increase in N2O emissions. Since soil
carbon and temperature are predicted to change in the future, the importance of
including the feedbacks to climate forcing involving changing natural emissions
of N2O is evident.
Ronald G. Prinn is TEPCO Professor of Atmospheric Chemistry, director of

the Center for Global Change Science, and co-director of the Joint Program
on the Science and Policy of Global Change at the Massachusetts Institute of
Technology. Dr. Prinn’s research interests incorporate the chemistry, dynamics,
and physics of the atmospheres of the earth and other planets, and the chemical
evolution of atmospheres. He is currently involved in a wide range of projects
in atmospheric chemistry and biogeochemistry, climate science, and integrated
assessment of science and policy regarding climate change. He received his Sc.D.
in chemistry from the Massachusetts Institute of Technology.

APPENDIX D 105
THE ROLE OF BIOGEOCHEMISTRY

IN THE CLIMATE SYSTEM:
EARLY EXPERIENCES FROM THE
NCAR COMMUNITY CLIMATE SYSTEM MODEL
Scott C. Doney
Woods Hole Oceanographic Institution
INTRODUCTION
The biogeochemical cycles of carbon, nitrogen, sulfur, and several other
elements are integral components of the climate system that, until recently,
have been neglected to a large degree in traditional physical climate studies.
Perturbations to the planet’s biogeochemical systems affect climate through
changes in atmospheric composition, land surface properties, and ecological
rates, which together in turn alter radiative balance and energy and water cycles.
Several climate modeling groups have begun to include biogeochemical and
ecological components into the coupled 3-D ocean, atmosphere, land climate
models used to assess past, present, and future climate change. Here I discuss
early results with the NCAR Community Climate System Model (CCSM). I focus
on carbon-climate interactions resulting from anthropogenic fossil fuel combus-
tion and climate warming as this example illustrates the complicated nature of
the underlying coupled physical-biological interactions. I conclude with a brief
overview of other biogeochemical processes being incorporated in the CCSM that
may introduce important feedback mechanisms, nonlinearities, and thresholds to
the climate system.
CARBON-CLIMATE EXPERIMENT OVERVIEW

A new three-dimensional global coupled carbon-climate model is presented
in the framework of the Community Climate System Model (CSM 1.4) (Doney et
al., 2006). A 1,000-year control simulation has stable global annual mean surface
temperature and atmospheric CO2 with no flux adjustment in either physics or
biogeochemistry. At low frequencies (timescale > 20 years), the ocean tends to
damp (20-25 percent) slow, natural variations in atmospheric CO 2 generated by
the terrestrial biosphere. Transient experiments (1820-2100) (Fung et al., 2005)
show that carbon sink strengths are inversely related to the rate of fossil fuel
emissions, so that carbon storage capacities of the land and oceans decrease and
climate warming accelerates with faster CO2 emissions (Figure 1). There is a
positive amplification between the carbon and climate systems, so that climate
warming acts to increase the airborne fraction of anthropogenic CO 2 and amplify
the climate change itself. Globally, the amplification is small at the end of the

106 APPENDIX D
0.4
Land Coupled
Land Uncoupled
Ocean Coupled
Ocean Uncoupled
0.35
Cumulative Uptake Fraction
0.3
0.25
0.2
0.15
300 350 400 450 500 550 600 650 700 750 800
Atmospheric CO2 (ppm)
FIGURE 1 Fractional uptake of anthropogenic CO2 into land and ocean sinks as a
function of atmospheric CO2 concentration from transient simulation (1820-2100) using
prescribed historical and SRES A2 CO2 emissions. SOURCE: Fung et al., 2005.
New donney fig 1
21st century in our model because

(theoffile
itswas
lownamed
transient
fig 2) climate response and the
near-cancellation between large regional changes in the hydrological and eco
system responses. notice the Legend text is different in this version
MODEL DESCRIPTION
The physical climate core of the coupled carbon-climate model is a modified
version of National Center for Atmospheric Research CSM1.4, which consists of
atmosphere, land, ocean, and ice components that are coupled via a flux coupler.
Into CSM1.4 are embedded a modified version of the terrestrial biogeochemistry
model CASA, and a modified version of the OCMIP-2 oceanic biogeochemis-
try model. CASA follows the life cycles of plant functional types from carbon
assimilation via photosynthesis, to mortality and decomposition, and the return of
CO2 to the atmosphere via microbial respiration. There are three live vegetation
pools and nine soil pools, and the rates of carbon transfer among them are climate

APPENDIX D 107
sensitive. The carbon cycle is coupled to the water cycle via transpiration and
to the energy cycle via dynamic leaf phenology (and hence albedo). A terrestrial
CO2 fertilization effect is possible in the model because carbon assimilation via
the Rubisco enzyme is limited by internal leaf CO2 concentrations; net primary
productivity (NPP) thus increases with external atmospheric CO2 concentra-
tions, eventually saturating at high CO2 levels. The ocean biogeochemical model
includes in simplified form the main processes for the solubility carbon pump,
organic and inorganic biological carbon pumps, and air-sea CO2 flux. New/export
production is computed prognostically as a function of light, temperature, phos-
phate, and iron concentrations. A fully dynamic iron cycle also has been added,
including atmospheric dust deposition/iron dissolution, biological uptake, vertical
particle transport, and scavenging.
NATURAL VARIABILITY
A sequential spin-up strategy is utilized to minimize the coupling shock
and drifts in land and ocean carbon inventories. In the 1,000-year control, global
annual mean surface temperature is ±0.10 K and atmospheric CO2 is ±1.2 ppm
(1σ) (Figure 2). The control simulation compares reasonably well against obser-
vations for key annual mean and seasonal carbon cycle metrics; regional biases
in coupled model physics, however, propagate clearly into biogeochemical error
patterns. Simulated interannual to centennial variability in atmospheric CO2
is dominated by terrestrial carbon flux variability, ±0.69 Pg C y–1, reflecting
FIGURE 2 Time series of global surface CO2 concentration from CSM1.4 carbon control
simulations. SOURCE: Doney et al., 2006.
Appen D Doney fig 2

108 APPENDIX D
primarily regional changes in net primary production modulated by moisture

stress. Power spectra of global CO2 fluxes are white on timescales beyond a
few years, and thus most of the variance is concentrated at high frequencies
(timescale < 4 years). Model variability in air-sea CO2 fluxes, ±0.10 Pg C y–1
(1σ), is generated by variability in temperature, wind speed, export production,
and mixing/upwelling.
ANTHROPOGENIC TRANSIENTS
Climate change is expected to influence the capacities of the land and oceans
to act as repositories for anthropogenic CO2 and hence provide a feedback to
climate change. A series of experiments with the coupled carbon-climate model
shows that carbon sink strengths are inversely related to the rate of fossil fuel
emissions, so that carbon storage capacities of the land and oceans decrease and
climate warming accelerates with faster CO2 emissions. Furthermore, there is a
positive feedback between the carbon and climate systems, so that climate warm-
ing acts to increase the airborne fraction of anthropogenic CO2 and amplify the
climate change itself. Globally, the amplification is small at the end of the 21st
century in this model because of its low transient climate response and the near-
cancellation between large regional changes in the hydrological and ecosystem
responses. Analysis of our results in the context of comparable models suggests
that destabilization of the tropical land sink is qualitatively robust, though its
degree is uncertain.
THE NEXT STEPS FORWARD

The preliminary treatment of the carbon cycle in CSM 1 is incomplete in
many regards. For example, the terrestrial biogeochemical component neglects
many of the factors thought to govern historical and future carbon sinks, such as
land use changes, disturbance/fire, dynamic vegetation, and nitrogen limitation.
Similarly for the ocean system, a number of other hypotheses have been proposed
on how the planet’s biogeochemical systems can alter climate and ecosystems.
These include variability in ocean carbon storage driven by changes in atmo-
spheric dust (iron limitation) and the impacts of ocean acidification on marine
calcifiers (corals, pteropods, coccolithophores, etc.). All of these processes (along
with reactive atmospheric chemistry) are being incorporated currently into a new
version of the CCSM3. Plans are also underway, though at a very early stage,
for including interactive human decisions, in effect moving CCSM toward a full
earth system modeling capability.

APPENDIX D 109
References
Doney, S. C., K. Lindsay, I. Fung, and J. John. 2006. Natural variability in a stable 1000 year coupled
climate-carbon cycle simulation. Journal of Climate 19(13):3033-3054.
Fung, I., S. C. Doney, K. Lindsay, and J. John, 2005. Evolution of carbon sinks in a changing climate.
Proceedings of the National Academy of Sciences U.S.A. 102:11201-11206.
Scott C. Doney is a senior scientist at the Woods Hole Oceanographic Institution.

Dr. Doney’s research interests include marine biogeochemistry and ecosystem
dynamics, large-scale ocean circulation and tracers, air-sea gas exchange, and
the global carbon cycle. He received his Ph.D. in chemical oceanography from
the MIT/Woods Hole Oceanographic Institution Joint Program.

110 APPENDIX D
UNDERSTANDING
ATMOSPHERE-BIOSPHERE INTERACTIONS:
THE ROLE OF BIOGENIC VOLATILE
ORGANIC COMPOUNDS
Alex Guenther
National Center for Atmospheric Research
Reactive biogenic volatile organic compounds (BVOC) have a substantial

impact on air quality, especially ozone and particles, and the ability of the
atmosphere to remove greenhouse gases, such as methane (Sanderson et al.,
2003; Went, 1960). They are also a component of the carbon cycle that produces
~ 4 Pg of CO2 each year (Guenther, 2002). These atmospheric chemical compo-
sition changes could perturb physical climate and the biosphere. Since BVOC
emission rates are very sensitive to physical, chemical, and biological driving
variables, they may have a significant role in earth system interactions and feed-
backs, including those illustrated in Figure 1. However, these processes are com-
plex, nonlinear, and difficult to predict given our current limited understanding.
Increases in emissions of some VOCs are likely to result in increases of gases that
can increase radiative forcing, but this would be accompanied by an increase in
organic particles, which could decrease radiative forcing.
Although there are hundreds of different BVOC species emitted into the
atmosphere, one compound (isoprene) contributes about half of the total annual
global flux (Guenther et al., 1995). Other BVOC, such as β-caryophyllene, are
particularly important for the production of secondary organic aerosol. There
are ~ 50 BVOC chemical species that have an important role for atmospheric
chemistry, either because they are emitted in large amounts or because they are
particularly important for some atmospheric process (Guenther et al., 2000).
BVOCs that have important ecological roles may differ from those with important
atmospheric roles. For example, some BVOCs used as signaling compounds are
present in the atmosphere at extremely low concentrations and have a negligible
impact on the atmosphere.
RESPONSE TO CHANGES IN
ATMOSPHERIC CHEMICAL COMPOSITION
Isoprene emission rates are sensitive to atmospheric trace gas levels. Isoprene
emission can increase when ozone is increased from background to levels repre-
sentative of a polluted city (e.g., Velikova et al., 2005). However, this may not be
sustained with long-term (months) exposure (e.g., Ennis et al., 1990). In addition,
isoprene emission can decrease in response to an increase in CO2 (e.g., Rosenstiel
et al., 2003). However, this decrease is minimized when plants are grown at less
than optimal soil moisture (Pegoraro et al., 2004).

APPENDIX D 111
FIGURE 1 The role of biogenic VOCs in biosphere-atmosphere interactions.
Very little is known about the potential impact of a changing atmospheric

oxidation capacity on the use of BVOC signaling compounds for defense or
attraction. The oxidation capacity of the atmosphere determines the lifetime of
most BVOCs and so could limit the effective range over which organisms can
use signaling compounds. The ecological implications of this change have not
been quantified.
RESPONSE TO CLIMATE CHANGE

Isoprene emission can be relatively insensitive to drought. Some studies
show that a decrease in soil moisture that results in much lower photosynthesis
and stomatal conductance can cause little or no decrease in isoprene emission
until soil moisture reaches the point where photosynthesis ceases and the leaves
begin to wilt (Pegoraro et al., 2004). There is occasionally even an increase,
possibly due to increasing leaf temperatures associated with decreased transpira-
tion. This may at least partly explain the surprisingly large seasonal variations in
isoprene emissions that have been observed in tropical rainforests, which are the
dominant global source of isoprene emission (Guenther et al., 1999). Isoprene
emissions in the dry season can be a factor of 3 higher than during the wet season
even after accounting for the known response of emissions to changes in tem-

112 APPENDIX D
perature and solar radiation. The unexplained variations appear to be negatively

correlated with soil moisture.
Isoprene emission is very sensitive to temperature and solar radiation. This
is compounded by the ability of plants to adapt to changing temperature and
light and results in even higher emissions associated with extended exposure to
temperature and light (Guenther et al., 1999). As a result, a long-term 3K increase
in temperature could result in an increase in isoprene emission of as much as a
factor of 2.
RESPONSE TO BIOLOGICAL CHANGE

Enclosure studies have shown that biological stresses (e.g., herbivory, fungal
or viral infections, insect pests) can result in large increases in biogenic VOC
emissions (e.g., Wildt et al., 2003). However, this phenomenon has not been
investigated on canopy or landscape scales. Increasing levels of biological stress
could be associated with future global change and would likely lead to elevated
emissions of a variety of BVOCs.
Isoprene and other terpenoid emission rates from different plant species
range across three to four orders of magnitude (Guenther et al., 2000). Landscape
average emissions can vary more than an order of magnitude depending on plant
species composition. On a global scale ~30 percent of all woody plants emit
isoprene (Guenther et al., 1995). Isoprene emitters are found in a wide variety
of landscapes, including tropical, temperate, and boreal ecosystems. There tends
to be a higher abundance of isoprene emitters in disturbed (early successional)
landscapes. In addition, many of the fast-growing tree plantation species (e.g,
poplar, eucalyptus, oil palm, rubber tree) have extremely high isoprene emissions
(Guenther et al., 2000). Large monoculture plantations of these species will likely
have a dramatic impact on local air chemistry.
REFERENCES
Ennis, C. A., A. L. Lazrus, P. R. Zimmerman, and R. K. Monson. 1990. Flux determination and physi-
ological response in exposure of red spruce to gaseous hydrogen peroxide, ozone, and sulfur
dioxide. Tellus Series B-Chemical and Physical Meteorology 42B:183-199.
Guenther, A. 2002. The contribution of reactive carbon emissions from vegetation to the carbon
balance of terrestrial ecosystems. Chemosphere 49(8):837-844.
Guenther, A., C. N. Hewitt, D. Erickson, R. Fall, C. Geron, T. Graedel, P. Harley, L. Klinger, M.
Lerdau, W. A. Mckay, T. Pierce, B. Scholes, R. Steinbrecher, R. Tallamraju, J. Taylor, and P.
Zimmerman. 1995. A global-model of natural volatile organic-compound emissions. Journal of
Geophysical Research-Atmospheres 100(D5):8873-8892.
Guenther, A., B. Baugh, G. Brasseur, J. Greenberg, P. Harley, L. Klinger, D. Serca, and L. Vierling.
1999. Isoprene emission estimates and uncertainties for the Central African EXPRESSO study
domain. Journal of Geophysical Research-Atmospheres 104(D23):30625-30639.
Guenther, A., C. Geron, T. Pierce, B. Lamb, P. Harley, and R. Fall. 2000. Natural emissions of non-
methane volatile organic compounds, carbon monoxide, and oxides of nitrogen from North
America. Atmospheric Environment 34(12-14):2205-2230.

APPENDIX D 113
Pegoraro, E., A. Rey, R. Murthey, E. Bobich, G. Barron-Gafford, K. Grieve, and Y. Malhi. 2004. Effect
of CO2 concentration and vapour pressure deficit on isoprene emission from leaves of Populus
deltoides during drought. Functional Plant Biology 31(12).
Rosenstiel, T. N., M. J. Potosnak, K. L. Griffin, R. Fall, and R. K. Monson. 2003. Increased CO 2
uncouples growth from isoprene emission in an agriforest ecosystem. Nature advance online
publication, January 5 (doi:10.1038/nature 01312).
Sanderson, M. G., C. D. Jones, W. J. Collins, C. E. Johnson, and R. G. Derwent. 2003. Effect of
climate change on isoprene emissions and surface ozone levels. Geophysical Research Letters
30(18):1936, doi:10.1029/2003GL017642.
Velikova, V., P. Pinelli, S. Pasqualini, L. Reale, F. Ferranti, and F. Loreto. 2005 Isoprene decreases
the concentration of nitric oxide in leaves exposed to elevated ozone. New Phytologist
166(2):419‑426.
Went, F. W. 1960. Blue hazes in the atmosphere. Nature 187(4738):641-643.
Wildt, J., K. Kobel, G. Schuh-Thomas, and A. C. Heiden. 2003. Emissions of oxygenated volatile
organic compounds from plants. Part II: Emissions of saturated aldehydes. Journal of Atmo-
spheric Chemistry 45(2):24.
Alex Guenther is a senior scientist, section head, and group leader of the
Biosphere-Atmosphere Interactions Group of the Atmospheric Chemistry Divi-
sion at the National Center for Atmospheric Research. Dr. Guenther’s research
interests include phytogeography and biogeochemistry; biosphere-atmosphere
interactions; developing and applying trace gas and aerosol flux measurement
techniques; understanding biological, chemical, and physical controls of trace
gas fluxes; numerical modeling of chemical exchange between terrestrial ecosys-
tems and the atmosphere; impact of biogenic emissions; and ecosystem uptake
and fire emissions on atmospheric chemistry and sustainability. He received his
Ph.D. from the Department of Civil and Environmental Engineering at Washing-
ton State University.

114 APPENDIX D
Human-Earth System Interactions

J. M. Reilly
Massachusetts Institute of Technology
The societal relevance of understanding complexities of earth system

i nteractions is that human activity both affects these systems and is affected
by changes in them. The goal of our research in this regard is to understand
the complex and dynamic interactions between human and earth systems at the
global scale, but resolved regionally. Our approach is to link a global model
of the world economy, the Emissions Prediction and Policy Analysis (EPPA)
model, with a comprehensive model of terrestrial hydrology and ecology and
the physical climate system as controlled by atmosphere and ocean processes
as they interact with terrestrial systems. At present, the physical and ecological
system is resolved at a level that has come to be known as an earth system
model of intermediate complexity, with less spatial resolution than a full-scale 3-
dimensional Atmosphere-Ocean General Circulation Model (AOGCM) but with
far more detail than the highly parameterized energy-balance-type models that
have been widely used in integrated assessment. We sacrifice further resolution
in the modeling of individual components of the earth system so that the Inte-
grated Global System Model (IGSM) is computationally efficient, allowing us to
evaluate very large ensemble runs (order 102-103). The desire to produce large
ensemble simulations is driven by our interest in understanding the phenomenon
of global environmental change as one of risk management. Research and model
ing of the interaction of physical and ecological earth systems is described by
Prinn (2006) and in greater detail in Sokolov et al. (2005).
The standard approach for evaluating the economics of climate change,
or many policy problems, is to study the policy issue in isolation, focusing on
the efficiency of economic instruments and calculation of an optimal policy. In
contrast, our approach has been to explicitly recognize that economic systems
are complex: multiple economic and environmental problems interact and these
interactions strongly affect the efficiency and effectiveness of policy instruments
and lead to very different conclusions than one would obtain absent this more
realistic representation of economic activity and its relationship to the natural
environment. In this regard, it is useful to consider three broad sets of complex
interactions. Those among: (1) economic policies directed toward different eco-
nomic problems, (2) the economy and the changing set of technological options
as they affect abatement potential and cost, and (3) the economy and the physi-
cal-ecological earth system. And, in fact, there are strong interactions among all
of these categories. Thus, the simplifying assumptions often made in research
to allow separate study of just the economic system, a single country or sector,
or climate in isolation of other economic issues are invalid. Whether these com-
plexities strongly affect conclusions one might reach of viable policy options is

APPENDIX D 115
a major focus of our ongoing research program. Published research appears in

journals such as the Journal of Environmental Economics and Management, The
Energy Journal, Energy Economics, Environmental Modeling and Assessment,
Climatic Change, Science, Nature, Energy Policy, Climate Policy, and others. A
description of the EPPA component and the IGSM and freely available reports
and reprints are on the Joint Program on the Science and Policy of Global Change
website, http://web.mit.edu/globalchange/www/. Here, I would report very briefly
some results from our program on each of the three sets of interactions above,
to illustrate how modeling the complexity of the system leads to different results
than one might otherwise expect. This is not intended to be a comprehensive
review of these topics as in many cases there is a considerable body of accumu-
lated research.
Complex Interactions Among Economic Policies

In an idealized economy the prices of goods reflect the marginal cost of
producing them, and consumers who base decisions on these prices equate their
marginal value of use of the good to these prices. The widely demonstrated result
in neoclassical economics is that the equalization of marginal cost and marginal
value through a price system, in this idealized world, results in an optimal use
of resources. Environmental economics adds to this issue, by observing that an
environmental problem like climate change is an “externality” in that, absent a
specific climate policy, there is no price paid for disposing of greenhouse gases
in the atmosphere, prices of goods and services that emit them thus do not
reflect the damages associated with emissions, and emissions and consumption
of emission-intensive goods are too high. In the idealized economy, one finds a
strong theoretical result that an economic instrument that prices carbon abatement
equally across sectors and regions is “cost effective” in the sense that it achieves
that level of reduction at the least cost, and “optimal” or “efficient” if the carbon
price is also determined to be equal to the marginal benefit in terms of reduced
damage.
There are many widely recognized ways in which real-world economies
depart from the idealized neoclassical economy. This recognition is hardly new;
it was formulated in the 1950s as the “theory of the second best” and also has
had implications in international trade where it was described as the paradox of
“immiserizing growth.” For a review of these issues, see Babiker et al. (2003a).
Violations of the idealized economy are variously described as preexisting dis-
tortions (other taxes that do not reflect a specific externality), externalities (other
than the climate change itself) that are not priced at all or are mispriced, and cases
where actors in the economy are large enough to affect prices and this affects
(or should be considered) in their decisions. The tax system of an economy is
one important distortion, and important interactions of climate policy and tax
policy have led to findings showing a “double dividend” for the U.S. under some

116 APPENDIX D
cases, but we have found that this does not necessarily translate to other countries
because it depends on the specifics of the tax system, so a double dividend was
less likely to exist in many European countries (Babiker et al., 2003b). We have
also shown that divergence from an economy-wide cap and trade system may be
economically superior to a cap and trade system that equalizes the marginal cost
of carbon across sectors (Babiker et al., 2004). This has further implications for
international permit trading, where we find that autarkic compliance with a cap
may be economically superior to an international permit trading system (Metcalf
et al., 2004). This can be traced to the relative level of taxation of fuels (Paltsev
et al., 2004). A further implication is that simple models where the carbon permit
price (marginal abatement cost) is taken as an indicator of the cost of the carbon
policy can be highly misleading. For example, the double dividend finding is one
where it is possible that, by recycling revenue from a carbon tax to offset exist-
ing distortionary capital and labor taxes, the carbon policy has a marginal social
benefit quite apart from any climate damage avoidance (Babiker et al., 2003a). In
other cases interaction of the carbon policy with existing energy taxes can lead
the average social cost of the carbon policy to be on the order of five times higher
than the marginal carbon reduction cost (Paltsev et al., 2004). Agriculture/land
use is an important source of emissions and/or a potential sink particularly in
developing countries (Hyman et al., 2003) and is a sector with extensive policy
intervention (preexisting distortions). Research on the interactions of climate
policy and agricultural policy has been limited or nonexistent to date, but accurate
representation of the agriculture sector and policies is needed to capture the inter-
actions between climate and other policies that affect agriculture. Obviously, agri-
culture is also a sector sensitive to environmental change, and so that interaction
among agricultural policies and the natural environment itself is essential.
Interactions between the economy and

changing technology
A description of the global economy embodied in a computable general
equilibrium model such as our Emissions Prediction and Policy Analysis (EPPA)
model represents explicitly or implicitly the technological options open to the
economy over time (Jacoby et al., 2006; Paltsev et al., 2005a). Much effort
has been invested in the research community to develop endogenous models
of technical change, recognizing that price or other economic signals resulting
from a climate policy would likely change the pace and direction of innovation.
Here, one must realize that the task of endogenously describing technical change
requires the modeler to describe all possible blueprints of relevant technologies
and their ultimate cost and the cost of discovering them. Essentially this demands
the modeler to know the details of technologies ahead of those who are working
to discover them. While this is an impossibility—if we knew it we would not
have to discover it—there are worthwhile avenues of research. Our approach has

APPENDIX D 117
been to first describe the current state of technological options, and examine their
changing potential as resource availabilities and prices changed, as in the case
of carbon sequestration (Jacoby et al., 2006). While at the time this research was
being done, natural gas combined-cycle technology was seen as the dominant
technology and thus the likely future in a carbon-constrained world, we found
that rising gas prices would likely mean that integrated gasification of coal with
sequestration was much more promising in the longer run. Gas prices have since
risen dramatically, and this result would now surprise no one, but the only hope
of escaping whatever the current mindset with regard to prices is to try to rep-
resent underlying fundamentals of demand and resource availability. We have
further explored the value of carbon sequestration in oceans, recognizing that
ultimately the carbon will end up in the ocean anyway, so that it was properly
investigated as “temporary storage” (McFarland et al., 2004). This work found
that there could be no value to temporary storage, and any value depended on the
existence of a backstop that would cap the price of carbon or include a damage
function and optimal carbon price that would likely mean ever-increasing atmo-
spheric carbon levels. Modeling the explicit technological options where there
are diverse technological options such as transportation can be daunting (Schafer
and Jacoby, 2003), with the need to consider the evolution of demand, changing
technological options, and interactions with existing policies such as fuel taxes
(Paltsev el al., 2005b).
Important in the issue of trying to represent technical change is to represent
the resources in the economy that are devoted to innovation and the fact that
allocating these resources to climate change mitigation (or adaptation) means
reallocating them away from other research endeavors (Sue Wing, 2003). In recent
work, we are following up on preliminary investigations (Jacoby et al., 2006) to
unravel the processes at work that may explain why technologies penetrate in
the classic S-shape and exhibit declining costs. Several processes are at work,
including vintaging/irreversibilities in the capital stock for the existing tech
nology, adjustment costs due to rapid scaling up of the capacity to produce the
new technology, monopoly rents associated with at least initially unique skills/
knowledge and possibly enforced through intellectual property rights laws, and
finally the innovation/learning process which may contribute to improvements
in the technology. The technology may also depend on a resource that is varied
in quality and is more or less accessible (e.g., wind or solar) or competes with
other uses (e.g., biomass competition with food for land). Different combina-
tions of these phenomena can lead to S-shaped penetration and/or declining cost.
Much work has focused on learning curves, assuming the declining cost reflects
innovation. Such relatively simplistic analyses would suggest that subsidization
or other stimulation of the market will push the technology cost down a “learn-
ing curve,” but this can be misleading to the extent other processes are at work.
Our preliminary results suggest that subsidization can lead to waste by increasing
adjustment costs if that is the primary explanation, extra profits with no improve-

118 APPENDIX D
ment in the technology if monopoly rents are the primary explanation, or advance
of the technology if learning explains the falling costs.
Key to modeling technical change is to recognize that knowledge is itself a
problem for neoclassical economics because the marginal cost of using knowl-
edge, once discovered, is zero, but pricing knowledge at zero does not compen-
sate for the cost of discovering it, thus the existence of intellectual property rights
protection that tries to balance compensation for innovation through granting
of monopoly rights with economic efficiency of making the technology widely
available. Also, there are typically knowledge spillovers so that even with the
patent protection, developers may never fully capture the returns to investments.
With some advocating a technology policy to solve the climate problem, care-
ful examination of these complex issues is critical. In one study of the Dutch
economy a fully dynamic, forward-looking, multisector general equilibrium set-
ting, including technology spillovers examines this issue (Otto et al., 2006). In
this study, effective technology policy can increase the needed carbon price and
the economic cost of climate policy in absolute terms, albeit the economy is much
larger with effective technology policy than without. We have found a similar
result in a much simpler framework, where we imagine that, exogenously, gas
resources are much larger than any conventional estimate. One might expect this
to lead to substitution away from coal, oil, shale oil, and the like, thus reducing
emissions of CO2. Instead, CO2 emissions increased, again because the growth
effect of lower gas prices dominated the substitution effect. Similarly exogenous
bias toward growth of the service sector—while reducing emissions somewhat
compared with the case of neutral growth in sectors—has a much smaller effect
than one might expect given the low energy intensity of the sector because of
the interindustry demands of the service sector for relatively energy intensive
goods and services (e.g., transportation). While initially surprising or counter-
intuitive, there is an intuition behind these results, and they suggest the need
for consideration of the complex interactions of technical change, growth, and
climate policy.
Interactions between the economy and the

physical-ecological earth system
The grand statement of the problem of interaction of the economy and the
physical-ecological earth system is the formal statement of the global warming
potential index issue, extended beyond the conventional greenhouse gases to
other pollutants and beyond warming to the direct or other effects of substances
of concern (Reilly and Richards, 1993; Reilly et al., 2003). The objective function
in this problem is to minimize the burden on the economy taking into account
the cost of controlling various greenhouse substances and the damage they cause.
This is an exceedingly demanding research agenda in that it requires valuation of
the multiple effects of global warming (crops, health, extreme events, ecological

APPENDIX D 119
disruption), the multiple effects of greenhouse substances (CO2 fertilization, dam-

aging effects on vegetation and health of tropospheric ozone and of aerosols), the
varying costs of abatement of different substances, and the complex interactions
(in the atmosphere due to chemistry, in mitigation cost due to shared generating
processes, and feedbacks such as environment on carbon uptake). We have con-
sidered this starting from a relatively complete description of shared generating
processes and well-articulated model of the complexity of atmospheric chemistry
to show that the 100-year GWPs undervalue methane abatement substantially
(Reilly et al., 1999). We have studied this within the context of a forward-looking
economic model with climate damages explicitly valued to study the implications
of alternative representations of the discount rate, finding that if the declining
discount rate formulation some have proposed for long-term problems is cor-
rect, then concern shifts to the very long-lived substances (Reilly et al., 2001).
And again we used the more fully articulated physical model to show that under
a policy to stabilize CO2 at 550 ppm the effects of undervaluing methane using
GWPs persists for 250 to 300 years (Sarfim et al., 2005). And most recently we
have examined the effects of common air pollutants on climate as already dis-
cussed by Prinn showing countervailing effects at least in terms of global mean
surface temperature changes (Prinn et al., 2006).
A recent focus is on the feedbacks of changing environment on the economy
itself. We have showed that tropospheric ozone could significantly reduce carbon
uptake by vegetation and thus increase the cost of meeting a 550 ppm stabiliza-
tion target by 6 to 21% (Felzer et al., 2005). The surprisingly large cost addi-
tion results from valuing the cost change at the margin. We have also evaluated
multiple environmental changes (climate, CO2, O3, and consequent changes in
soils) on agricultural crops and the economy. This shows significant current
ozone damage in the United States, Europe, and China (4-9% loss of the value
of crop production), and the potential for this loss to increase substantially even
if ozone precursors are controlled (Reilly et al., 2004, 2006). At the same time,
this shows a generally beneficial effect of climate/CO2 largely driven by the
positive CO2 fertilization effect. We find that the agricultural sector “adapts”
almost completely to these changes in terms of the production effect, but that
the economic effect is measured in other sectors as resources shift into or out of
agriculture. So, the cost of adaptation, in terms of % loss of the value of crop
production, is about ½ the direct yield loss even though production changes very
little in response to fairly severe productivity shocks. Notably, international trade
has effects across different regions because of the differential productivity effects.
We have also examined the health effects of air pollution and the resultant effects
on the economy. So far we have focused on the United States and, preliminarily,
China (Yang et al., 2005; Matus et al., 2006; Matus, 2005), finding a substantial
remaining burden of air pollution on the economy, particularly of China, where
the remaining burden is estimated at 10% of macroeconomic consumption. Our
interest in this work is to complete the modeling of feedbacks on the earth sys-

120 APPENDIX D
tem. Effects on agriculture, and changing production and trade, mean changes
in land use in producing regions, with consequent effects on the biogeochemical
cycle. Abandonment of land, or reduced intensity of use, would mean increased
carbon uptake, while expansion of intensified use would likely lead to release
of carbon and other greenhouse gases. We expect important interactions with
mitigation options, in particular biomass energy that competes for land and is
similarly affected by environmental change. Air pollution health effects, through
their effect on the economy, may also affect emissions, but we are also interested
in joint policy solutions whereby a climate policy may affect health via its effect
on air pollution emissions, or conversely air pollution policy driven by the desire
to reduce health effects may lead to changes in climate.
An important goal for us with regard to analysis of impacts is to value dam-
ages in a manner consistent with mitigation costs to make for a more consistent
comparison of benefits and costs. Mitigation cost analysis works on a rich theo-
retical and empirical basis, developed as computable general equilibrium models
that can be simulated dynamically. Damage assessment can be extremely ad hoc,
multiplying a constant wage rate or value of life times an estimate of hours or
lives lost. In this regard, one of the early findings is that even though pollution
levels can fall over time, the absolute damages may rise over time because real
wages and other prices rise. We also find significant improvement of damage
estimates through modeling of the accumulation effect of chronic exposure to
air pollution. The methods we have developed for estimating impacts are based
on the same rich theory and empirical foundation as mitigation costs, and we
have now set in place satellite physical accounts for land, energy resources, and
population (Asadoorian, 2005) so that we can dynamically link this theoretically
based economic model with earth system components.
REFERENCES
Asadoorian, M. O. 2005. Simulating the Spatial Distribution of Population and Emissions to 2100.
MIT Joint Program for the Science and Policy of Global Change Report No. 123, Cambridge,
MA.
Babiker, M., G. Metcalf, and J. Reilly. 2003a. Tax distortions and global climate policy. Journal of
Economic and Environmental Management 46:269-287.
Babiker, M., L. Viguier, J. Reilly, A. D. Ellerman, and P. Criqui. 2003b. The welfare costs of hybrid
carbon policies in the European Union. Environmental Modeling and Assessment 8:187-197.
Babiker, M., J. Reilly, and L. Viguier. 2004. Is emissions trading always beneficial. Energy Journal
25(2):33-56.
Felzer, B., J. Reilly, J. Melillo, D. Kicklighter, M. Sarofim, C. Wang, R. Prinn, and Q. Zhuang. 2005.
Future effects of ozone on carbon sequestration and climate change policy using a global bio-
geochemical model. Climatic Change 73:345-373.
Hyman, R. C., J. M. Reilly, M. H. Babiker, A. De Masin, and H. D. Jacoby. 2003. Modeling non-CO2
greenhouse gas abatement. Environmental Modeling and Assessment 8:175-186.
Jacoby, H. D., J. Reilly, and J. R. McFarland. 2006. Technology and technical change in the MIT
EPPA model. Energy Economics 28(5-6):610-631.

APPENDIX D 121
Matus, K. 2005. Health Impacts from Urban Air Pollution in China: The Burden to the Economy and
the Benefits of Policy. Master’s thesis, MIT.
Matus, K., T. Yang, S. Paltsev, and J. Reilly. 2006. Economic benefits of air pollution regulation in
the USA: An integrated approach. Climatic Change (in press).
McFarland, J., J. Reilly, and H. J. Herzog. 2004. Representing energy technologies in top-down eco-
nomic models using bottom-up information. Energy Economics 26:685-707.
Metcalf, G., M. Babiker, and J. Reilly. 2004. A note on weak double dividends. Topics in Economic
Analysis & Policy 4(1): Article 2.
Otto, V. M., A. Löschel, and J. Reilly. 2006. Directed Technical Change and Climate Policy. MIT Joint
Program for the Science and Policy of Global Change Report No. 134, Cambridge, MA.
Paltsev, S., J. M. Reilly, H. D. Jacoby, and K. H. Tay. 2004. The Cost of Kyoto Protocol Targets: The
Case of Japan. MIT Joint Program for the Science and Policy of Global Change Report No.
112, Cambridge, MA.
Paltsev, S., J. M. Reilly, H. D. Jacoby, R. S. Eckaus, J. McFarland, M. Sarofim, M. Asadoorian,
and M. Babiker. 2005a. The MIT Emissions Prediction and Policy Analysis (EPPA) Model:
Version 4. MIT Joint Program for the Science and Policy of Global Change Report No. 125,
Cambridge, MA.
Paltsev, S., H. Jacoby, J. Reilly, L. Viguier, and M. Babiker. 2005b. Modeling the transport sector: The
role of existing fuel taxes. In: Energy and Environment, R. Loulou, J-P. Waaub, and G. Zaccour,
eds. Springer, New York: 211-238.
Prinn, R. G. 2007. Ecosystem-climate feedbacks studied with an integrated global system model. Pp.
97-99 in Understanding and Responding to Multiple Environmental Stresses. Washington, D.C.:
The National Academies Press.
Prinn, R., J. Reilly, M. Sarofim, C. Wang, and B. Felzer. 2006. Effects of air pollution control on
climate. In: Integrated Assessment of Human-Induced Climate Change. M. Schlesinger, ed.
Cambridge University Press (in press).
Reilly, J., and K. Richards. 1993. An economic interpretation of the trace gas index issue. Environ-
mental and Resource Economics 3:41-61.
Reilly, J., R. Prinn, J. Harnisch, J. Fitzmaurice, H. Jacoby, D. Kicklighter, J. Melillo, P. Stone,
A. Sokolov, and C. Wang. 1999. Multigas assessment of the Kyoto protocol. Nature
401:549‑555.
Reilly, J., M. Babiker, and M. Mayer. 2001. Comparing Greenhouse Gases. MIT Joint Program on
the Science and Policy of Global Change. Report No. 77, Cambridge, MA.
Reilly, J. M., H. D. Jacoby, and R. G. Prinn. 2003. Multi-Gas Contributors to Global Climate Change.
Arlington, VA: Pew Center on Global Climate Change.
Reilly, J., B. Felzer, S. Paltsev, J. Melillo, R. Prinn, C. Wang, A. Sokolov, and X. Wang. 2004. ��The
economic impact of climate, CO2, and tropospheric ozone on crop yields in China, the US, and
Europe. EOS Abstract 7314; B33A-0239.
Reilly, J., S. Paltsev, B. Felzer, X. Wang, D. Kicklighter, J. Melillo, R. Prinn, M. Sarofim, A. Sokolov,
and C. Wang. 2006. ��
Global economic effects of changes in crops, pasture, and forests due to
changing climate, carbon dioxide, and ozone. Energy Policy (in press).
Sarofim, M., C. E. Forest, D. M. Reiner, and J. M. Reilly. 2005. Stabilization and global climate
policy. Global and Planetary Change 47:266-272.
Schafer, A., and H. D. Jacoby. 2003. Technology detail in a multi-sector CGE model: Transport under
climate policy. Energy Economics 27(1):1-24.
Sokolov, A. P., C. A. Schlosser, S. Dutkiewicz, S. Paltsev, D. W. Kicklighter, H. D. Jacoby, R. G.
Prinn, C. E. Forest, J. Reilly, C. Wang, B. Felzer, M. C. Sarofim, J. Scott, P. H. Stone, J. M.
Melillo, and J. Cohen. 2005. The MIT Integrated Global System Model (IGSM) Version 2: Model
Description and Baseline Evaluation. MIT Joint Program for the Science and Policy of Global
Change Report No. 124, Cambridge, MA.

122 APPENDIX D
Sue Wing, I. 2003. Induced Technical Change and the Cost of Climate Policy. MIT Joint Program for
the Science and Policy of Global Change Report No. 102. Cambridge, MA.
Yang, T., J. Reilly, and S. Paltsev. 2005. Air pollution health effects: Toward an integrated assess-
ment. Pp. 267-293 in Coupling Climate and Economic Dynamics, A. Haurie and L. Viguier,
eds. Berlin: Kluwer Publishers.

APPENDIX D 123
CLIMATE CHANGE, MULTIPLE STRESSES,

AND AGRICULTURE:
WHEN ALL ELSE IS NOT EQUAL
William E. Easterling
Pennsylvania State University
Scientific understanding of how agricultural systems respond to rapid social

and environmental change has largely been obtained using research approaches
that focus on individual stresses or forcings (e.g., air pollution, climate variability
and change, pests and pathogens, loss of genetic diversity, desertification, and
land degradation) most often in isolation from one another. Interactions among
such stresses are poorly studied, even within integrated assessment modeling
frameworks. When focusing on a single stressor, assumptions that all else is equal
are the convention.
Agricultural ecosystems currently feed a population of just over 6 billion
people, providing enough food for the average global citizen to consume 2,790
calories per day. One of the great human achievements of the 20th century was the
steady increase in global agricultural production due to the expansion of cropped
land by mechanization plus crop varietal improvements, better plant nutrition,
effective pest management, and other technological advances that increased pro-
ductivity. When asked to feed a growing global population, the world’s farmers
and their supporting institutions responded effectively. Even more remarkable
was that this increase in production occurred in the face of multiple environ-
mental and social stresses on the food system, including widespread climate
variability—prolonged droughts, severe floods, and heat waves—and likely the
early stages of climate change, pest and pathogen outbreaks, loss of genetic
diversity in agroecosystems, desertification and land degradation, water scarcity,
war, epidemics, increasing income divides between rich and poor, government
mismanagement, and global population growth, to name a few. But the news is
not all good as there have been notable regional failures and there are reasons for
concern looking into the future. In this abstract I review recent trends and future
projections of global agricultural production and then consider how multiple
stresses may alter those projections. Thesholds and nonlinearities, where they are
known, are identified, and research gaps and opportunities are mentioned.
One of the greatest stresses on global agriculture has been the historic rise
in population and per capita income, both of which constitute the principal deter-
minants of global food demand. Global food demand governs the scale at which
the world’s farmers must produce in order to maintain healthy and productive
lifestyles of all people. This stress will continue into the future, although there
is reason for optimism. Recent revised United Nations population projections to
2050 anticipate that the deceleration of world population growth may be even
faster than previously thought. The most recent Medium Variant projection for
the world population in 2050 has been revised down to 9.1 billion from its previ-

124 APPENDIX D
ous projection of 9.3 billion. The slowing population growth—1.6 percent p.a.
today versus 0.6 percent p.a. for 2050—combined with a growing percentage of
the world population reaching adequate levels of nutrition (due to increasing per
capita income, especially in developing countries), leads to a gradual slowing
down of growth in world demand for food and, correspondingly, in world produc-
tion required to meet demand.
Recent estimates by the Food and Agriculture Organization project annual
growth in world agricultural production to decline from 1.6 percent in 2000-2015
to 1.3 percent in 2015-2030 and 0.8 percent in 2030-2050 due to slowing growth
in world agricultural demand. This still implies a roughly 55 percent increase in
world production by 2030 compared to current production. These projections
assume that in the developing countries (where almost all global land expansion
takes place, mainly in sub-Saharan Africa and Latin America) another 185 million
ha of arable land (+19 percent) will be brought into production between now and
2050 and another 60 million ha of irrigated land (+30 percent). Average cereal
yields in the developing countries would have to rise from 2.7 tonnes/ha now to
3.8 tonnes/ha by 2050.
No clear picture of how climate change is likely to alter the above assessment
has emerged in the literature. The preponderance of global agricultural studies
suggest that climate change is likely to diminish global agricultural capacity
by only a few percent if at all by 2050 when taking into account regions that
may benefit (i.e., North America, Europe) and regions that may suffer (i.e., the
tropics). Any losses would be on top of substantial gains in world output as noted
above. A small suite of modeling studies predict that world crop (real) prices are
likely to continue to decline through the first two to three degrees C of warming
before rising with additional warming (IPCC TAR, 2001/5)—hence, two to three
degrees of warming appears to be a crucial threshold for crop prices.
While the global situation looks manageable, there are several reasons for
concern at regional levels. Several stresses on agricultural production systems are
impeding the achievement of regional food security, especially in the develop-
ing world. Sub-Saharan Africa is a case in point that is worth a closer look. It is
a region in which stresses on food security occur across several levels of scale.
Table 1 lists some examples of major stresses on sub-Saharan Africa that operate
at global and regional/local levels.
Agricultural stressors in sub-Saharan Africa operating at the global scale
include climate change and widening agricultural trade deficits. Tropical crop-
ping systems have been shown to be vulnerable to even the slightest increase in
temperatures because, for most of the major food crops being grown there, mean
maximum temperatures are already at the high end of effective photosynthetic
temperatures. Modeling studies project tropical crop yields (rice, maize) to fall
markedly with only a one degree warming, even when CO2 fertilization is consid-
ered—such is a critical threshold. In addition, most of the nations of sub-Saharan
Africa became net food importers in the early 1980s and are expected to see their

APPENDIX D 125
TABLE 1 Stressors on Sub-Saharan Africa Food Security

Global
• Climate change
• Widening agricultural trade deficits
Local and Regional
• Rapid population growth
• Emerging infectious diseases, i.e., HIV/AIDS
• Lack of safe drinking water, public health
• Rising incomes leading to dietary changes, i.e., more meat consumption
• Political instability and civil strife
• Long-term drought
• Land degradation
agricultural trade deficit quadruple by 2030. This has inhibited the development
of the agriculture sector in the region and limited agricultural income growth,
which is a major driver of development.
Agricultural stressors in sub-Saharan Africa operating at regional to local
scales include a number of social and environmental challenges. Food production
across most of Africa has not kept up with rapid population growth. The popula-
tion of sub-Saharan Africa is currently growing at a rate of 2.6 percent p.a. and is
expected to grow at 2.2-2.6 percent p.a. out to 2030, which is more than double
the world average. Sub-Saharan Africa is the only region of the world where per
capita food consumption not only remained below acceptable levels (less than
2200 Kcal/day), but has fallen in the last two decades. The effect of emerging
infectious diseases such as HIV/AIDS on population growth will be minimal, but
their effect on agricultural productivity could be catastrophic. The U.S. Census
Bureau estimates that over 40 percent of the reproductive-age population in South
Africa is HIV-positive. FAO projects that 20 percent of South Africa’s agricultural
labor force will have been lost to HIV/AIDS over the period 1985-2020. This
epidemic strikes a population that has generally poor levels of public health and
inadequate access to safe drinking water, which are stresses by themselves. On
top of these demographic and public health stresses, sub-Saharan Africa experi-
ences widespread political instability and civil strife that has particularly inter-
rupted the distribution of food to people who need it most.
Several environmental problems will continue to stress the sub-Saharan food
production system. Southern Africa experienced protracted droughts throughout
the 1990s and is considered to be in a dry period relative to the longer-term
historical record. Again, no clear picture has emerged as to how the frequency
and intensity of droughts may be changed by climate change across sub-Saharan
Africa. The degradation of land resources by agriculture becomes a major feed-
back limiting future agricultural productivity due to deterioration of the land
resource base. This is problematic throughout sub-Saharan Africa. Processes at
work include salinization of irrigated areas, overextraction of groundwater, chem-

126 APPENDIX D
ical depletion of soils, water saturation, leaching of nitrate into water bodies (pol-
lution, eutrophication), off-site deposition of soil erosion sediment, and enhanced
risks of flooding following conversions of wetlands to cropping. There is a great
deal of debate over how much land degradation has affected crop productivity.
Some estimates suggest that land degradation has reduced crop productivity by
25 percent in Africa since World War II.
How the many stresses listed above fit together in determining the vulnerability
and resiliency of sub-Saharan agriculture is not known with any certainty. The
same can be said virtually anywhere else in the world. Environmental stressors
surely interact strongly with social and economic stressors. A drought may sub-
stantially reduce crop yields, but temporary crop price increases brought on by
production disruptions may offset the stress to farmers, although the consumer
pays more. Major gaps exist in understanding how multiple stresses on agriculture
and food security interact. The following list details some potential priorities that
might help address these gaps:
• Compilation of datasets on agriculture and food-related stresses using GIS

technology—data should be scalable and easily assembled in different geographic
units and time scales
• Improvement in basic understanding of how multiple stresses interact to
influence important quantities, such as crop yields, public health, regional and
per capita income, and the like—special attention should be paid to identifying
nonlinearities and thresholds
• Development of modeling methodologies in which interactions between
multiple stressors (social and environmental) are explicit
• Design and implementation of comprehensive food system monitoring capa-
bilities leveraging the success of the U.S. Agency for International Development’s
Famine Early Warning System
• Improvement in basic understanding of the resiliency of agricultural sys-
tems in the face of not one but several stresses occurring simultaneously, espe-
cially the identification of system redundancies and other coping strategies
William E. Easterling is the director of the Penn State Institutes of the Environ-
ment and professor of geography with a courtesy appointment in agronomy at
the Pennsylvania State University. Dr. Easterling’s research interests include the
potential for agriculture in developed and developing countries to adapt to cli-
mate variability and change; the role of scale in understanding the vulnerability
of complex systems; how land use change may influence the uptake and release
of carbon in the terrestrial biosphere; the use of experimental long-term climate
forecasts to assist decision making under conditions of uncertainty; and the
development of methodologies for detecting the impacts of observed 20th-century
climate change on natural and managed ecosystems. Dr. Easterling received his
Ph.D. in geography from the University of North Carolina at Chapel Hill.

APPENDIX D 127
SOCIOECONOMIC IMPACTS OF
MULTIPLE STRESSES ON CITIES
Patricia Romero Lankao
Autonomous Metropolitan University, Xochimilco
Four issues require attention when analyzing the socioeconomic impacts

of multiple stresses on cities when reflecting on possible adaptive strategies:
(1) weight of path dependency, (2) importance of scale, (3) usefulness of exist-
ing analytic tools, and (4) role of institutions. All will be described in terms of
what has been learned, which research gaps still exist, and how understanding
multiple stresses might affect the conduct of future research, assessments, and
decision making.
DEVELOPMENT PATHWAYS: KEY LINK BETWEEN

MULTIPLE STRESSES AND DRIVERS?
Many of the forces behind emissions trajectories and land use changes
induced by urban areas, such as economic growth, technological transformations,
demographic shifts, and governance structures, are similar or even the same to
those underlying diverse pathways of development, explaining in part why
1. industrialized countries and the wealthy in the South account for the
highest share of atmospheric emissions, and have much higher ecological foot-
prints (e.g., land use changes induced by cities’ demands on wood) than poor
countries, regions, and sectors. Rees and Wackernagel (1996) have documented
the influences and effects of world cities, such as Amsterdam in distant places
(“teleconnections”). Another example is the demand for meat in Mexico City,
which led in the 1950s and 1960s to land cover changes in Tabasco 400 km away
(Barkin et al., 1978).
2. some regions and sectors, especially from the developing world, are more
vulnerable to impacts such as climate variability and change than others. The poor
in urban areas for example lack access to climate-controlled shelters; they already
face environmental problems (sanitation, deficiencies in the operation of public
services, location in risk-prone areas); they suffer malnutrition, poor housing
conditions, and low income. All those conditions that have been further worsened
by structural adjustment programs implemented during the last two decades may
produce a highly segregated urban space and contribute to aggravate—especially
for the poor—the negative impacts of changes in biophysical processes in urban
areas.
If the impacts of changes in atmospheric composition go beyond a certain

threshold and/or urban areas are not prepared to deal with that level of impact,

128 APPENDIX D
then cities from both the developed and the developing world may become
vulnerable. Consider the 2003 heat wave in Europe as example of an event a
developed country like France was vulnerable to. This did not necessarily result
from lack of resources. But it rather related not only to health conditions and
health services for the elderly people, but also to social conditions and organi-
zation (i.e., to inadequate climate conditioning in buildings; to the fact that the
elderly people were alone while their families were on vacation). Think about
environmental change’s possible impacts on the pool of resources and ecosystems
an urban area relies on. As some scenarios foresee, climate change will strongly
affect already stressed watersheds in both the developing world (e.g., Lerma
and Cutzamala for Mexico City) and the developed world (Colorado River for
Southern California).
The problem is that most scholars tend to focus either on some drivers of
emissions trajectories (e.g., IPCC’s Work Group III) or on the vulnerability or
impact to multiple stresses (e.g., IPCC’s Work Group II using vulnerability
assessments). And knowledge would advance faster if both groups could explore
and model the complex linkages between drivers of development pathways and
stressors facing urban sectors and localities. Consider another example. Market
forces and the declining role of the state as urban planner are key drivers of urban
growth in risk-prone areas of developing countries and even of industrialized
countries—as New Orleans has recently shown us. The retrenchment of the state
in its role as regulator usually appears as one of the multiple stresses facing urban
dwellers. Wouldn’t it be better to explore how market forces and other drivers
relate or cascade with other and multiple stresses to produce certain vulnerabili-
ties/adaptive capacities? Isn’t it a challenge to develop an integrated perspective
to understand the pathways through which socioeconomic, institutional, and
environmental processes influence livelihoods, economic activities, urban life,
migration, and other realms where adaptive capacity takes place in urban areas?
RELEVANCE OF SCALE
Scale is important in diverse and not yet fully explored ways when assess-
ing the socioeconomic impacts of changes in atmospheric composition and
dynamics.
• First, the referred changes are but a part of a set of multiple stresses
operating at diverse scales in space and through time (e.g., markets, deficiencies
in housing, or in the supply and operation of urban services, infrastructure, sani-
tation, and health). The problem is that research on this issue is in its infancy.
We need to better understand—and if possible model—the global and regional
socioeconomic, geopolitical, and environmental processes affecting the vulner-
ability of urban systems.

APPENDIX D 129
• Second, both the exposure to and the distribution of urban groups and
localities sensitive to the impacts of changes in atmospheric composition and
dynamics vary greatly across scale. The primary social and economic conditions
that influence adaptive capacity (e.g., access to financial resources or to govern-
mental relief) also differ with scale. One could say, for instance, that at a national
scale cities in industrialized countries such as Norway or the United States can
cope with most kinds of gradual environmental changes, but focusing on more
localized differences (between New York and New Orleans for instance) can show
considerable variability in stresses and capacities to adapt. This is another area
where more research is needed.
• Third, temporal scale is a critical determinant of the capacity of urban
systems to adapt to environmental changes. The history of the city—path depen-
dency—will determine the diversity and complexity of its population’s current
and future adaptive capacity. Of course, rapid changes or abrupt changes (e.g.,
flooding, droughts) are more difficult to absorb without painful costs than gradual
change. But gradual change can accumulate until urban areas reach a threshold in
which their adaptive capacity is no longer feasible. Consider how slow changes in
the length and frequency of seasons could affect water supplies in cities, or how
slow changes in temperature and humidity could affect the livelihood of urban
people. These issues have received little attention so far.
ANALYTICAL TOOLS
Rather than using one-dimensional perspectives to understand the responses
of urban areas to the impacts of changes in the composition of the atmosphere,
we need broader multidimensional and multiscale approaches. This is a challenge
indeed. Examples of such tools and concepts that will be described in this section
are the Kaya identity, vulnerability/adaptation assessments, livelihoods, and toler-
able windows approach. According to the IPAT identity, environmental impact is
the product of the level of population combined with affluence (e.g., measured
by income per capita) and the level of technology (e.g., measured by emissions
per unit of income). Numerous articles use this kind of identity to analyze the
drivers of CO2 emissions (Kaneko and Matsouka, 2003), often referring to it as
the “Kaya identity,” according to which:
 gdP   Energy   CO2 

CO2 Emissions = Population� � �
 Population   gdP   Energy 
Note that some of the components can be analyzed for sectors of activity to get greater detail of
emissions sources.

130 APPENDIX D
The Kaya identity has the advantage of being simple and allowing for “some
standardization in the comparison” and analysis of diverse emissions trajectories
(Nakicenovic, 2004). Clearly, it is a very general way of looking at emissions; for
example, it assumes that each variable enters linearly and independently, which is
surely not the case, and it omits other factors (such as institutional ones) as emis-
sions drivers that are left as residual of the equation if looking at past emissions
evolution (IPCC, 2000; Nakicenovic, 2004). The identity is broadly used to con-
struct scenarios; it helps to identify proximate but not ultimate drivers, but it fails
to identify many key drivers—e.g., government regulation, social organization,
and economic organization, as well as public perceptions, which influence pat-
terns of technologies in use affecting trajectories of greenhouse gases and other
atmospheric releases. The question is whether a similar equation in which the
variables do not enter independently could be applied to the analysis of multiple
stresses, whether such an equation could at least allow for some standardization
in the comparison of diverse cases.
Rather than selecting a particular environmental stress of concern and trying
to identify its consequences as impact assessment does, vulnerability/adaptation
assessments:
1. Choose a group or unit of concern (e.g., informal or wealthy settlements,

poor or rich urban dwellers).
2. Try to evaluate the risk of specific adverse outcomes for that unit when
confronted with a set of stresses (e.g., climate variability and change, new patterns
of foreign direct investment and of industrial location, and structural adjustment
programs). Note that all these stresses redefine the economic base of urban areas,
drive new territorial patterns of urban growth and new emissions trajectories
(Romero Lankao et al., 2005); at the same time they define a new physiognomy
and geography of vulnerable groups and localities within the cities.
3. Seek to identify a set of factors (e.g., institutions, assets, social capital)
that may reduce or to the contrary enhance adaptation to those stressors.
Precisely because vulnerability assessments concentrate on exposure units,

they assume that vulnerability is highly dependent on scale and context. This has
two consequences we should address in our workshop. Assessments focusing on
a narrow range of scales will miss most of the importance for societal efforts to
cope with the impacts of changes in the atmosphere. It is therefore essential for
One such case is two institutional changes—privatization of state firms and decreased pub-
lic expenditures—happening during the 1990s in Mexico City. They were aimed at eliminating
inefficient and insolvent enterprises, thereby reducing public expenditure. Those transformations
became one of the drivers of the shift in mode share from Metro and buses to minibuses and low-
capacity modes, by this in increasing GHG emissions by the transportation sector (Romero Lankao
et al., 2005).

APPENDIX D 131
vulnerability assessments to address multiple drivers and stresses that interact

across a variety of scales.
One tool is under development within the IPCC community and intends to
deal with the future: the tolerable windows approach. This tool selects an expo-
sure unit and classifies the potential outcomes either as acceptable or adverse.
The next step is to try to define the dynamic combination of environmental and
socioeconomic and institutional stresses that could result in negative outcomes.
For this it is necessary to develop scenarios of the future that include not only
environmental stresses but also socioeconomic stresses (Toth, 2001).
Livelihood analysis is another multidimensional concept applied to both
rural and urban studies. It pays attention to the way in which a living is obtained
and gives especial consideration to agency. Livelihood analysis comprises “the
capabilities, assets (stores, resources, claims, and access) and activities required
for a means of living” (Ellis, 2000). The tool is useful to study how individuals,
households, and communities in cities adapt to the impacts of changes in the
composition of the atmosphere. It pays attention to the social, economic, insti-
tutional, and environmental dimensions of adaptation strategies by those agents;
it explores how those dimensions interact at different scales. This tool can offer
insights to design policies aimed at both enhancing adaptation to the impacts of
atmospheric changes and reshaping the socioeconomic impacts of existing urban
development pathways.
I would like to close this paper with one citation and some thoughts on the
role of institutions in adapting to the impacts of changes in the composition of
the atmosphere. According to Steffen et al. (2003):
The current states of vulnerability research and vulnerability assessment exhibit

both the potential for substantial synergy in addressing global environmental
risks . . . as well as significant weaknesses which undermine that potential.
A substantial base of fundamental knowledge has been created. However it is
highly fragmentary in nature, with competing paradigms, conflicting theory,
empirical results often idiosyncratic and tied to particular approaches, and a lack
of comparative analysis and findings.
My guess is that one research path for us is to develop meta-analytic tools

to tease out the paradigms behind those approaches and to see whether and,
if so, at which levels it is possible to make those analyses and their findings
comparable.
INSTITUTIONS
Last but not least, institutions play a significant role both as drivers/stressors
(IDGEC, 2005; Romero Lankao et al., 2005) and as enhancers of urban systems’
ability to cope with and adapt to the negative impacts of changes in atmospheric

132 APPENDIX D
composition and land use (IDGEC, 2005). As already mentioned in this paper the
first task we are confronted with is to understand within an integrated perspective
the pathways through which institutions together with other socioeconomic and
environmental processes influence livelihoods, economic activities, urban life,
migration, and other realms where adaptive capacity takes place in cities. The
second and more difficult task is to consider existing research on the science
policy interface, which may help us understand
1. why and how decision makers relate to scientific information and find-
ings on socioeconomic and institutional impacts of changes in land use and in
atmospheric composition;
2. the resources and institutional capacity decision makers have to design
and implement adaptation strategies;
3. the role of other organizations, institutions, and stakeholders in the design
and implementation of adaptation strategies.
REFERENCES
Barkin, D., con la colaboración de Adriana Zavala. 1978. Desarrollo regional y reorganización
Campesina: La Chontalpa como reflejo del problema agropecuraio mexicano. Mexico: Editorial
Nueva Imagen.
Bruchner, T., G. Petschel-Held, F. L. Tóth, H.-M. Füssel, C. Helm, M. Leimbach, and H.-J.
Schellnguber. 1999. Climate change decision-support and the tolerable windows approach.
Environmental Modeling and Assessment 4:217-234.
Ellis, F. 2000. Rural Livelihood and Diversity in Developing Countries. Oxford: Oxford University
Press.
Kaneko T., and Y. Matsouka. 2003. Driving forces behind the stagnancy of China’s energy related
CO2 emissions from 1996-1999: The relative importance of structural change, intensity change
and scale change. Draft.
Nakicenovic, N. 2004. Socioeconomic driving forces of emission scenarios. Pp. 225-239 in The
Global Carbon Cycle: Integrating Humans, Climate and the Natural World, C. B. Field and
M. R. Raupach, eds. Washington, D.C.: Island Press.
Rees, W., and M. Wackernagel. 1996. Urban ecological footprints: Why cities can not be
sustainable—and why they are a key to sustainability. Environmental Impact Assessment
Review 16:223‑248.
Romero Lankao, P., H. López Villafranco, A. Rosas Huerta, G. Günther, and Z. Correa Armenta, eds.
2005. Can Cities Reduce Global Warming? Urban Development and the Carbon Cycle in Latin
America. México, IAI, UAM-X, IHDP, GCP. 92 pp.
Steffen, W., A. Sanderson, P. D. Tyson, J. Jäger, P. A. Matson, B. Moore III, F. Oldfield, K. Richard-
son, J. Schellnhuber, B. L. Turner II, and R. J. Wasson. 2004. Global Change and the Earth
System: A Planet Under Pressure. Berlin, Heidelberg, New York: Springer-Verlag.
Dr. Patricia Romero Lankao is a professor in the Department of Politics and

Culture at the Autonomous Metropolitan University, Campus Xochimilco, in
Mexico City, Mexico. Her general field of expertise and interest is in the interface
of the human dimensions of global environmental change. She has published on

APPENDIX D 133
issues such as the design of Mexican environmental policy, water policy in Mexico
City, environmental perceptions and attitudes toward public environmental strate-
gies and instruments, and vulnerability to climate variability and change among
farmers and water users. Dr. Romero, a sociologist by training, has two doctoral
degrees, one in regional development from the Autonomous Metropolitan Univer-
sity and one in agricultural sciences from the University of Bonn, Germany.

134 APPENDIX D
UNDERSTANDING VULNERABILITY OF ECOSYSTEM

GOODS AND SERVICES AND RESPONSE STRATEGIES
Richard B. Norgaard
University of California, Berkeley
Questions suggested: (1) What has been the economic impact of large-scale
stresses to ecosystems, and how has this impact been characterized? (2) How is
vulnerability of ecosystem goods and services characterized? (3) How best can
society value ecosystems and ecosystem goods and services to effectively prepare
for the impact of multiple stresses?
BROAD RESPONSE
It is important to approach these questions from a dynamic systems perspec-
tive within which environmental and socioeconomic systems are interacting over
time, each affecting the other. Economic impacts are not simply “there” to be
identified by the economist but rather are a product of prior conditions. Thus the
economic impact of large-scale or multiple stresses is very dependent on how well
we are prepared for and how well we respond to the stresses. Hurricane Katrina
laid bare the human and economic costs of being ill prepared and unresponsive.
The critical point here is that a particular stress, or particular combination of
stresses, does not map to a particular economic impact. Between a stress and an
impact there are typically human-modified environments, technological systems,
and socioeconomic institutions and capabilities whose states and hence responses
to stresses depend on both preceding stresses and human decisions to recognize,
develop, and protect environmental qualities, technological capacities, and insti-
tutional capabilities.
Thus the economic implications of an extended drought in the Midwest that
seriously jeopardized corn, soybean, and wheat production depends on the exis-
tence, or not, of the agroecological, technological, and socioeconomic conditions
necessary to raise alternative grains that are quite normally grown, for example, in
arid regions of India. The definition of a drought (i.e., is there enough water?) is
not simply a matter of the difference between rainfall and crop needs, or between
soil storage, rainfall, and evapotransporation. Need, evapotransporation, and soil
storage depend on the crops grown.
As we saw in the case of Katrina, the distribution of income can also criti-
cally affect the impacts observed. The human toll of an extended drought in the
Midwest could be no more than the discomfort of foregoing beef and eating soy
burgers. During the stress of World War II, margarine became a normal food in
the American diet through programs that restricted the availability of butter for
all. But if the rich continue to eat beef through a drought, the price of grains could
quickly become a major hardship on the poor.

APPENDIX D 135
Similarly, there are scalar issues associated with risk that are critical. Histori-
cally, near-subsistence farmers planted a variety of crops in the expectation that
some of them would fit the weather conditions of a particular year. With trade
and specialization, risks and risk reduction strategies become distributed more
globally. Thus, low corn, soybean, and wheat yields in the United States in any
particular year or decade are dampened by global production unless, as is possible
with climate change, much of the globe is affected in the same period. A rapid
transition in climate conditions could also occur globally such that globalizing
risk becomes a less effective strategy.
We should be as concerned about combinations of social and environmental
stresses as about combinations of environmental stresses. It was much more diffi-
cult to be prepared for and to respond to the combination of stresses known as the
“dust bowl” during the Great Depression because the social system was already
highly stressed by a dysfunctional economy. No doubt environmental stresses
during World Wars I and II also resulted in greater hardship because social sys-
tem resources were devoted to the wars. It is also important to keep in mind that
the social conditions, specifically the movement of soldiers and material at the
end of World War I, helped spread the Spanish Flu virus, which ended up killing
10 times as many Americans as the Great War.
We have a tendency to think of fair to good socioeconomic times as normal
and bad as the exception, but surely 20 percent of the years in the United States
during the 20th century were pretty bad. And just as surely, 20 percent of the
world’s population at any one time is engaged in a civil or regional war or having
bad economic times. Yet most of our scenarios for “sustainable development” are
built around fair to good times being an uninterrupted norm.
Acknowledging the dynamic interactions between social and environmental
systems has not yet led to the incorporation of appropriate models to address peri-
odic stresses in environmental assessments, let alone combinations of stresses.
Efforts to date to portray the interactions between social systems and environ-
mental systems in assessments tend to forecast alternative smooth scenarios. For
example, the assessments by the Intergovernmental Panel on Climate Change are
tied to alternative scenarios that differ, for example, by overall rates of growth,
types of technology, and differences between industrialized and less industrial-
ized countries, but these scenarios play out without discontinuities. The scenarios
of the Millennium Ecosystem Assessment are more heuristic and the differences
between them with respect to resilience overall, patchiness globally, and protec-
tion of the poor to stresses are discussed but not actually quantified.
Valuation of ecosystem services as a technique has also presumed a steady
forecast of good times, both environmentally and socially (at least for the rich).
Because a dollar of a poor person tends to be counted the same as the dollar of a
rich person, ecotourism values, for example, can swamp subsistence ecosystem
services important to the poor. Clearly it is these more basic values that are more
important to be studying now that we are more concerned about multiple stresses.

136 APPENDIX D
In any case, we value what we know, and as the uncertainty of the future becomes
more clear to more people, we will value ecosystem services differently. This is
rather a circular problem, for we want the values to inform people of the impor-
tance of ecosystem services. But to the extent that some of the ways economists
derive values flow from an informed public expressing choices, most ecosystem
services, especially under conditions of stress, are highly undervalued because
people are less aware of the importance of the services than they should be.
Richard B. Norgaard is a professor in the Energy and Resources Program at the

University of California, Berkeley. His areas of expertise include environmental
epistemology; ecological, environmental, and resource economics; environment
and development in less developed countries; development as social system and
ecological system coevolution; tropical rain forests; pest management; petroleum
development; and water development. His recent research addresses how environ-
mental problems challenge scientific understanding and the policy process, how
ecologists and economists understand systems differently, and how globalization
affects environmental governance. Dr. Norgaard received his Ph.D. in economics
from the University of Chicago.

APPENDIX D 137
Patterns, Sources, and Impacts of Decadal-to-

Multidecadal Climate Variability in the West
Julio L. Betancourt
U.S. Geological Survey Desert Laboratory
Both instrumental and tree-ring records show that the western U.S. climate
exhibits low-frequency modes in the form of persistent and widespread droughts
that alternate with pluvials. Some notable examples are the dry Medieval period
(A.D. 900-1300) and the ensuing wetter Little Ice Age (A.D. 1400-1850); the
dramatic switch from the megadrought in the late 1500s to the megapluvial in
the early 1600s; and the bracketing of epic droughts in the 1930s and 1950s by
two of the wettest periods (1905-1920 and 1965-1995) in the last millennium.
Such decadal-to-multidecadal (D2M) hydroclimatic variability is assumed to
operate through the oceans and clearly can occur independent from anthropogenic
forcing. Statistical and modeling studies show teleconnections to low-frequency
SST variability in the Pacific, Indian, and North Atlantic Oceans, and there is
mounting interest in both mechanisms and predictability.
D2M precipitation variability in the western United States tends to be spa-
tially coherent and can synchronize physical and biological processes in ways
that are complex and difficult to forecast and monitor. D2M variability can
synchronize fluctuations in surface water availability across major basins and
can thus overextend regional drought relief and interbasin transfer agreements.
Traditionally, both water resource and floodplain management in the West have
been based mostly on stationary assumptions about surface flow—that the mean
and moments of the annual or peak discharge distributions do not change over
time. Federal flood insurance relies on the concept of the 100-year flood, which
is calculated routinely with flood frequency methods that assume stationarity. And
we also assume stationarity in annual operations of critical water resources. For
instance, Article III of the Compact apportions 7.5 million acre feet (MAF) per
year each to the Upper and Lower Colorado River Basins, stipulates that Upper
Basin states cannot deplete the flow at Lees Ferry by more than 75 MAF over a
period of 10 consecutive years and mandates a moving 10-year average release of
8.23 MAF/year from Lake Powell into the Lower Basin. In the case of D2M vari-
ability, 100 years may not be long enough to evaluate stationary assumptions in
water resource planning, and we are forced to rely on tree-ring reconstructions of
precipitation and streamflow. These reconstructions vary in coverage and quality,
but nonetheless are a good first approximation of the history and long-term prob-
ability of D2M precipitation and streamflow regimes. In particular, probability
distribution functions from long reconstructions of hydroclimate (precipitation or
Dr. Betancourt was unable to attend the workshop; his slides and perspectives were integrated into
Jonathan Overpeck’s presentation.

138 APPENDIX D
streamflow) or climatic index series (AMO, PDO, etc.) can be used to calculate
the occurrence and return probabilities of climatic episodes. This information can
then be mapped to decisions about annual water resource operations or facilities
planning.
By synchronizing ecological disturbances and recruitment pulses, D2M vari-
ability also plays a key role in structuring woodland and forest communities in
western watersheds. In the event of longer, hotter growing seasons, D2M vari-
ability will still determine the timing and pace of ecosystem changes. Oscillations
between warm-dry and warm-wet regimes will continue to produce uncommonly
large disturbances followed by accelerated regeneration and succession. Such
large-scale vegetation changes will have complex hydrological effects and will
add further uncertainty to water resource availability. A principal challenge for
land managers in the 21st century will be to manage for disturbance and succes-
sion in purposeful and systematic ways that promote asynchrony and patchiness
at local to regional scales while still preserving goods and services that eco
systems provide.
Julio L. Betancourt is project chief, National Research Program, Water Resources

Division, U.S. Geological Survey and adjunct professor in the Departments
of Geoscience and Geography at the University of Arizona. The focus of his
research is ecosystem and watershed responses to climate variability on different
temporal and spatial scales. Dr. Betancourt received his Ph.D. in geosciences
from the University of Arizona, Tucson.

Appendix E
Committee Biosketches
Rosina M. Bierbaum (Co-chair) is dean of the University of Michigan’s School

of Natural Resources and Environment and professor of natural resources and
environmental policy. Prior to her appointment at Michigan, she served as acting
director of the Office of Science and Technology Policy (OSTP) in the Executive
Office of the President. She joined OSTP in November 1993 as a senior policy
analyst and served as assistant director for environment before being confirmed
by the U.S. Senate as associate director. As the Administration’s senior scientific
advisor on environmental research and development, Dr. Bierbaum provided sci-
entific input and guidance on a wide range of national and international environ-
mental issues, including global change, air and water quality, endangered species,
biodiversity, ecosystem management, endocrine disruptors, environmental moni-
toring, natural hazards, and energy research and development. Dr. Bierbaum
is a fellow of the American Association for the Advancement of Science and
a member of the NRC’s Board on Atmospheric Sciences and Climate. She is a
past member of the Government-University-Industry Research Roundtable of the
National Academy of Sciences. Dr. Bierbaum received her Ph.D. in ecology and
evolution from the State University of New York at Stony Brook.
Mary Anne Carroll (Co-chair) is director of the Program for Research on

Oxidants: PHotochemistry, Emissions and Transport (PROPHET), executive
director of the Biosphere-Atmosphere Research and Training Program (BART),
and a professor of atmospheric, oceanic, and space sciences and chemistry and
geological sciences at the University of Michigan. Her areas of interest include
oxidant photochemistry, distribution, and trends; atmosphere-forest exchange of
139

140 APPENDIX E
reactive nitrogen (a factor in carbon storage); and the impacts of air pollutants
on ecosystem function and emissions. She served as editor for the Journal of
Geophysical Research-Atmospheres from 1997 to 2000, is a past member of the
NRC’s Committee on Geophysical and Environmental Data and the Committee
on Atmospheric Chemistry, and a current member of its Board on Atmospheric
Sciences and Climate. Dr. Carroll received her Sc.D. in atmospheric chemistry
from the Massachusetts Institute of Technology.
Christopher B. Field is director of the Department of Global Ecology at the

Carnegie Institution of Washington and professor by courtesy in the Department
of Biological Sciences at Stanford University. Trained as an ecologist, Dr. Field
has conducted environmental research from tropical rainforests to deserts to
alpine tundra. He developed an evolutionary approach to understanding the spatial
organization of plant canopies and the adaptive significance of leaf aging. These
studies led to work on the role of nitrogen in regulating plant growth and photo-
synthesis and suggested ways that plant physiological responses could be sum-
marized with a few parameters, providing a basis for predicting many aspects of
ecosystem function at very large scales. He has recently emphasized formalizing
approaches for summarizing plant responses into models that simulate ecosystem
exchanges of carbon, water, and energy at the global scale. These models help test
hypotheses and understand the future status of terrestrial ecosystems, especially
responses to and influences on global change factors like increased atmospheric
carbon dioxide or altered climate. Dr. Field is a member of the National Academy
of Sciences and the NRC’s Board on Environmental Studies and Toxicology. He
received his Ph.D. in biology from Stanford University.
Edward L. Miles is the Virginia and Prentice Bloedel Professor of Marine and
Public Affairs in the School of Marine Affairs at the University of Washington
and senior fellow at the Joint Institute for the Study of Atmosphere and Oceans.
Since 1965 Dr. Miles has worked at the interface of the natural and social sci-
ences and law with a focus on outer space, the oceans, and the global and regional
climate systems. Trained originally in political science and international relations,
he has invested close to 30 years in learning about oceanography and fisheries
science/management and 13 years in learning about the planetary climate system.
His research and teaching interests have encompassed international science and
technology policy; the design, creation, and management of international envi-
ronmental regimes; a wide variety of problems in national and international
ocean policy; and the impacts of climate variability and climate change at global
and regional space scales. Dr. Miles is a member of the National Academy of
Sciences and the NRC’s Committee on the Human Dimensions of Global Change
and Policy and Global Affairs Committee. He received his Ph.D. in international
relations/comparative politics from the University of Denver.

APPENDIX E 141
Donald A. Wilhite is founder and director of the National Drought Mitiga-

tion Center and International Drought Information Center and a professor in
the School of Natural Resources at the University of Nebraska-Lincoln, where
he has been on the faculty since 1977. His areas of interest include drought
preparedness, mitigation and policy, climate and drought monitoring, the policy
implications of climate variability and climate change, and the effects of climate
on society. In conjunction with this research, he has worked with many developed
and developing countries, regional organizations, and United Nations agencies on
drought management and policy issues. He has also conducted numerous train-
ing seminars and workshops on drought planning and management. Dr. Wilhite
is chair of the drought discussion group for the U.N.’s Secretariat for the Inter-
national Strategy for Disaster Reduction. He received his Ph.D. in geography,
climatology, and water resources from the University of Nebraska.


Committee On Earth-Atmosphere Interactions: Understanding and Responding To Multiple Environmental Stresses, National Research Council

Uploaded by

Copyright:

Available Formats

Committee On Earth-Atmosphere Interactions: Understanding and Responding To Multiple Environmental Stresses, National Research Council

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Committee On Earth-Atmosphere Interactions: Understanding and Responding To Multiple Environmental Stresses, National Research Council

Uploaded by

Copyright:

Available Formats

Understanding Multiple Environmental Stresses:

Copyright © National Academy of Sciences. Permission is granted for this material to be

Committee on Earth-Atmosphere Interactions:

Board on Atmospheric Sciences and Climate

Division on Earth and Life Studies

The NATIONAL ACADEMIES PRESS

Copyright © National Academy of Sciences. All rights reserved.

International Standard Book Number-13: 978-0-309-10331-2

Copyright 2007 by the National Academy of Sciences. All rights reserved.

Printed in the United States of America

Copyright © National Academy of Sciences. All rights reserved.

The National Academy of Sciences is a private, nonprofit, self-perpetuating society of

Copyright © National Academy of Sciences. All rights reserved.

Copyright © National Academy of Sciences. All rights reserved.

COMMITTEE ON EARTH-ATMOSPHERE INTERACTIONS:

ROSINA M. BIERBAUM (Co-chair), University of Michigan, Ann Arbor

Copyright © National Academy of Sciences. All rights reserved.

BOARD ON ATMOSPHERIC SCIENCES AND CLIMATE

ROBERT J. SERAFIN (Chair), National Center for Atmospheric Research,

Copyright © National Academy of Sciences. All rights reserved.

Periodically the Board on Atmospheric Sciences and Climate (BASC) works

Copyright © National Academy of Sciences. All rights reserved.

This workshop was intended as a step in identifying the types of near-term

Chris Elfring, Director

Copyright © National Academy of Sciences. All rights reserved.

M. Joan Alexander, NorthWest Research Associates/CORA, Boulder, Colorado

Copyright © National Academy of Sciences. All rights reserved.

Oklahoma. Appointed by the National Research Council, he was responsible for

Copyright © National Academy of Sciences. All rights reserved.

Copyright © National Academy of Sciences. All rights reserved.

4 LESSONS LEARNED FROM THE CASE STUDIES 36

Copyright © National Academy of Sciences. All rights reserved.

Copyright © National Academy of Sciences. All rights reserved.

 UNDERSTANDING AND RESPONDING TO MULTIPLE ENVIRONMENTAL STRESSES

1. An integrated regional web of sensors that links existing observations into

Workshop participants identified the Regional Integrated Sciences and

Copyright © National Academy of Sciences. All rights reserved.

1. Building a new U.S. ground-based ecological measurement network to

Copyright © National Academy of Sciences. All rights reserved.

Copyright © National Academy of Sciences. All rights reserved.

DEFINING THE CONCEPT OF “MULTIPLE STRESSES”

Copyright © National Academy of Sciences. All rights reserved.

 UNDERSTANDING AND RESPONDING TO MULTIPLE ENVIRONMENTAL STRESSES

rudimentary understanding of the dynamics of interactions between different

Copyright © National Academy of Sciences. All rights reserved.

THE NATURE OF THE PROBLEM

Copyright © National Academy of Sciences. All rights reserved.

 UNDERSTANDING AND RESPONDING TO MULTIPLE ENVIRONMENTAL STRESSES

Second, although a decade of research on greenhouse gas emissions, ozone

Copyright © National Academy of Sciences. All rights reserved.

WHAT ASSESSMENTS CONCLUDE ABOUT RESEARCH NEEDS

INTRODUCTION TO THE CASE STUDIES

Copyright © National Academy of Sciences. All rights reserved.

10 UNDERSTANDING AND RESPONDING TO MULTIPLE ENVIRONMENTAL STRESSES

NONLINEARITIES, THRESHOLDS, AND

Copyright © National Academy of Sciences. All rights reserved.

cal concepts: nonlinear responses, thresholds, vulnerability, and resilience. The

Copyright © National Academy of Sciences. All rights reserved.

12 UNDERSTANDING AND RESPONDING TO MULTIPLE ENVIRONMENTAL STRESSES

UNDERSTANDING AND RESPONDING TO MULTIPLE ENVIRONMENTAL STRESSES

UNDERSTANDING AND RESPONDING TO MULTIPLE ENVIRONMENTAL STRESSES

UNDERSTANDING AND RESPONDING TO MULTIPLE ENVIRONMENTAL STRESSES

Because of the slow-onset nature of drought, the ability of resource managers