DRAFT EPA/600/R-11/01A
DO NOT CITE OR QUOTE February 2011
External Review Draft
v>EPA
United States
Environmental Protection
Agency
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-stressor Global Change
Vulnerability Assessments
NOTICE
THIS DOCUMENT IS A PRELIMINARY DRAFT. This information is distributed solely for
the purpose of pre-dissemination peer review under applicable Information Quality Guidelines. It
has not been formally disseminated by the U.S. Environmental Protection Agency. It does not
represent and should not be construed to represent any Agency determination or policy.
Global Change Research Program
U.S. Environmental Protection Agency
Office of Research and Development
National Center for Environmental Assessment
1200 Pennsylvania Ave., Washington, DC 20460
-------
DISCLAIMER
This document is an external draft for review purposes only. This information is distributed
solely for the purpose of pre-dissemination peer review under applicable information quality
guidelines. It has not been formally disseminated by the U.S. Environmental Protection Agency.
It does not represent and should not be construed to represent any agency determination or
policy. Mention of trade names or commercial products does not constitute endorsement or
recommendation for use.
Do Not Cite or Quote Page ii
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Authors
Julie Blue The Cadmus Group, Inc.
JeffMaxted
Nupur Hiremath
Matthew Diebel
Charles Hernick
Jonathan Koplos
Chris Weaver EPA/ORD/NCEA
Technical Advisors
David Allan University of Michigan
John Day Louisiana State University
Thomas Meixner University of Arizona
David Gochis National Center for Atmospheric
Kathleen Miller Research
David Yates
EPA Reviewers
Robert Hall EPA/R9
Doug Norton EPA/OW/OWOW
Rachael Novak EPA/OW/OST
Betsy Smith EPA/ORD/NERL
Rick Ziegler EPA/ORD/NCEA
Do Not Cite or Quote Page in
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Preface
This report investigates the issues and challenges associated with identifying, calculating, and
mapping indicators of the relative vulnerability of water quality and aquatic ecosystems, across
the United States, to the potential adverse impacts of external forces such as long-term climate
and land-use change. We do not attempt a direct evaluation of the potential impacts of these
global changes on ecosystems and watersheds. Rather, we begin with the assumption that a
systematic evaluation of the impacts of existing stressors will be a key input to any
comprehensive global change vulnerability assessment, as the impacts of global change will be
expressed via often complex interaction with such stressors: through their potential to reduce
overall resilience, or increase overall sensitivity, to global change. This is an assumption made
by many environmental scientists, but, to date, there has been relatively little exploration of the
practical challenges associated with comprehensively assessing how the resilience of ecosystems
and human systems in the face of global change may vary as a function of existing stresses and
maladaptations. The work described in this report is a preliminary attempt to begin such an
exploration.
To do so we gathered, from the literature, a set of more than 600 indicators of water quality and
aquatic ecosystem condition and changes in condition, along with numerous datasets from EPA,
other federal agencies, and NGOs, and we have used all of this as a testbed for identifying best
practices and challenges for calculating and mapping vulnerability nationally. We investigated
gaps in ideas, methods, data, and tools as well. Specifically, we explored:
• Challenges associated with identifying those indicators that speak specifically to
vulnerability, as opposed to those reflecting simply a state or condition. In this context, we
define vulnerability as adverse impacts accrued over time and associated with external
stresses from, for example, climate or land-use change;
• Challenges associated with calculating and estimating the values of these vulnerability
indicators, including establishing important indicator thresholds that reflect abrupt or large
changes in the vulnerability of water quality or aquatic ecosystems;
• Challenges associated with mapping these vulnerability indicators nationally, including data
availability and spatial aggregation of the data;
• Challenges associated with combining and compositing indicators and developing multi-
indicator indices of vulnerability.
We hope that this report will be a useful building block for future work on multi-stressor global
change vulnerability assessments. Ultimately, we believe the work described here can contribute
to bridging disconnects between the decision support needs of the water quality and aquatic
ecosystem management communities and the priorities and capabilities of the global change
science data and modeling communities. In addition, we hope it will help to synthesize lessons
learned from more detailed, place-based, system-based, or issue-based case studies. Such studies
include those conducted on individual watersheds, on wetlands, and on urban ecosystems. This
synthesis will be used to obtain national-scale insights about impacts and adaptation; and to
prioritize future work in developing adaptation strategies for global change impacts.
Do Not Cite or Quote Page iv
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
I would like to acknowledge the excellent work of the Cadmus Group, Inc. in their collaboration
with NCEA to develop this draft report. In addition, a team of external expert advisors provided
critical insights that have informed all of our work in the project to date: Drs. David Allan,
Kathleen Miller, John Day, David Gochis, David Yates, and Thomas Meixner. Many thanks as
well to Mike Slimak, whose substantial contributions greatly improved this report. Finally, I
would like to thank all the NCEA Global Change Research Program staff for their numerous and
significant inputs to this project.
Chris Weaver
Global Change Research Program
EPA/ORD/NCEA
Do Not Cite or Quote Page v
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Table of Contents
I. Introduction 10
II. Synergies with Other EPA Efforts 13
III. Indicators Considered for this Report 15
A. Literature Search 16
a. Core literature 16
b. Protocol for collecting additional relevant literature 20
B. Creation of a Comprehensive List of Indicators 20
a. Identifying indicators of water quality and aquatic ecosystem condition 20
b. Selection of indicators 21
c. Exclusion of certain indicators and studies 23
d. Deletion of duplicate indicators 23
IV. Challenges Part I: Indicator Classification 24
A. Defining Vulnerability 25
a. Determinants of vulnerability 25
b. Defining a vulnerable situation 26
c. Biophysical and socioeconomic domains 27
d. Predictability and uncertainty 27
B. Classifying Vulnerability Indicators 28
C. How do These Indicators Reflect Vulnerability?. 31
V. Challenges Part II: Determining Relative Vulnerability 45
A. Vulnerability Gradients and Thresholds 45
B. Modifying and Refining Indicators to Incorporate Thresholds 49
VI. Challenges Part III: Mapping Vulnerability 54
A. Assessment of Indicator Data Availability and Mappability at the National Scale 55
a. Identification of data sources for indicators 55
b. Description of major data sources 56
c. Supporting information collected for data sources 62
d. Lack of data and other unresolved data problems 62
1. Data availability issues 62
2. Data sets without national coverage 67
3. Non-uniform spatial distribution of data 68
4. Temporal gaps 68
e. Data problems that could be resolved 69
B. Creation of Example Maps 70
Do Not Cite or Quote Page vi
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
C. Spatial Aggregation 73
a. Local Variation 73
b. Extent of spatial units (HUC Levels) 73
c. Alternate Spatial Frameworks 76
1. Watersheds (and hydrologic units) 76
2. Ecoregions 76
3. Coastal Areas 77
D. Categorical Aggregation 77
VII. Challenges Part IV: Combining Indicators 81
A. Combining Indicators with Other Data 81
B. Composites of Vulnerability Indicators 84
a. Creating a Composite Map 84
b. Characterizing vulnerability profiles 85
VIII. Summary and Recommendations 89
A. Summary of Challenges 89
a. Challenges Part I: Indicator Classification 90
b. Challenges Part II: Determining Relative Vulnerability 91
c. Challenges Part III: Mapping Vulnerability 91
1. Data and mappability 91
2. Spatial Aggregation 92
d. Challenges Part IV: Combining Indicators 92
B. Recommendations for Future Research 94
a. Assessment of non-mappable indicators 94
b. Identifying opportunities to enhance source data 95
c. Development of new indicators from available data sets 96
d. Use of indicators for future studies 97
e. Establishment of stress-response curves, vulnerability thresholds, and baseline
conditions 98
f. Drawing on other established approaches for combining indicators 98
g. Incorporating landscape metrics 98
h. Incorporating metrics of adaptive capacity 98
IX. References 100
X. Appendices In a separate PDF
A. Bibliography
B. Comprehensive List of Indicators
C. Data Sources and Supporting Information
D. Technical Notes
E. Mapping Methodology
Do Not Cite or Quote Page vii
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
F. Example Maps for Indicators of Water Quality and Aquatic Ecosystem Vulnerability by
HUC-4
G. Descriptions of Example Maps by HUC-4
H. Example Maps for Indicators of Water Quality and Aquatic Ecosystem Vulnerability by
Ecoregion
I. Descriptions of Example Maps by Ecoregion
J. Vulnerability Category Matrix
K. Evaluation and Potential Modification of Vulnerability Indicators
L. Research Team Members and Contact Information
Do Not Cite or Quote Page viii
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
List of Tables
Table 1. List of Core Literature 16
Table 2. Indicator Primary and Secondary Categories 23
Table 3. Rationale for Exclusion of Certain Indicators 24
Table 4. List of Vulnerability Indicators 29
Table 5. Indicators with Objective Thresholds and their Vulnerability Categories 49
Table 6. Vulnerability Indicators Categorized in the National Environmental Status and
Trend (NEST) Framework 54
Table 7. Distribution of Data Source 56
Table 8. Indicators Eliminated Due to Lack of Data or Unresolved Data Problems 63
Table 9. Data Gaps 69
Table 10. List of Mapped Vulnerability Indicators 70
Table 11. Principal Components Loadings for the Twenty Four Indicators Included in the
PCA Analysis 86
List of Figures
Figure 1. Flowchart of Methodology Used To Identify and Map Vulnerability Indicators 17
Figure 2. Indicator Definition from EPA's 2008 Report on the Environment 21
Figure 3. Mapping Data Relative to Regulatory Thresholds 50
Figure 4. Modification of Indicator Definitions Using Existing Data 52
Figure 5. Modification of Indicator Definitions Using Existing Data 53
Figure 6. Limitations of Data Sets Containing Self-reported Data 67
Figure 7. Aggregation, Precision, Coverage, and Data Density 75
Figure 8. Data Represented by Different Spatial Frameworks 78
Figure 9. Spatial Framework for Coastal Zone Indicators 79
Figure 10. Different Breaks to Distinguish Data Classes 80
Figure 11. Current and Future Vulnerability to Water Shortages 82
Figure 12. Vulnerability Profile Similarity 88
Figure 13. Indicator Evaluation Process 94
Do Not Cite or Quote Page ix
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
i I. Introduction
2 The U.S. Environmental Protection Agency (EPA) Global Change Research Program (GCRP),
3 located within the Office of Research and Development (ORD), is a national-scale program that
4 supports decision-making about adapting to potential climate change and other global change
5 impacts on air and water quality, aquatic ecosystems, and human health. GCRP collaborates with
6 EPA Program and Regional offices, and state, local, municipal, and tribal natural resource
7 managers, to provide scientific support for these efforts. There is a large body of literature
8 suggesting that improvements to measuring, modeling, and understanding climate changes
9 relevant to the hydrologic cycle, water quality, and aquatic ecosystems are needed (e.g., Bates et
10 al., 2008; Miller and Yates, 2005; Kundzewicz et al., 2007; Lettenmaier et al., 2008; Barsugli et
11 al., 2009; Poff et al., 2002). The management strategies of the past will not necessarily be
12 adequate given increased awareness of stressors such as climate change and land-use change. As
13 emphasized by a number of recent publications, top-down, prediction-based assessments of the
14 interactions between climate change and hydrologic systems, ecosystems, and human
15 communities will likely be of limited usefulness for local decision-making. This is due to current
16 and foreseeable limits on reducing climate uncertainties, and because these kinds of assessments
17 are not necessarily compatible with conclusions from the social sciences about how information
18 is used in decision-making (e.g., see Dessai et al., 2009; Johnson and Weaver, 2009; Moser and
19 Luers, 2008; NRC, 2009; Fischhoff, 1994; Sarewitz et al., 2000).
20
21 Effective decision support will instead start with a deep commitment to understand the systems
22 we manage or aim to protect and a willingness to use what we know now for decision-making,
23 while working to learn more. In general, comparing relative vulnerabilities fits in well with this
24 framework, because direct evaluation of the absolute effects of climate change on water quality
25 and aquatic ecosystems is out of reach given the state of the science for many of our
26 vulnerability indicators. Yet policy decisions must continue to be made in the absence of perfect
27 information. Understanding the current condition and threats posed to our environment now can
28 be the lens through which we view the potential threats posed by global change. This can be
29 achieved through systematic, quantitative planning frameworks that help us to understand and
30 evaluate various management strategies across a wide range of plausible futures. The result of
31 such planning should be the selection of management strategies that alleviate, or at least do not
32 exacerbate, existing and anticipated vulnerabilities of water quality and aquatic ecosystems. In
33 other words, we should seek strategies that are robust with respect to the inherent uncertainties of
34 the problem (e.g., Lempert et al., 2004; Brown et al., 2010).
35
36 Informed by this philosophy, GCRP has developed and is implementing a multi-year research
37 effort designed to improve national-scale understanding of the multiple complex interactions
38 between global change and the nation's waters. Part of this work is a major effort devoted to the
39 development of scenarios of future climate, land-use, and hydrologic change. For example,
40 GCRP is conducting hydrologic modeling in 20 large, U.S. watersheds in an attempt to provide
41 broad, national-scale scenarios of streamflow and nutrient/sediment loading across a wide range
42 of potential climate and land-use changes, to improve our understanding of the plausible range of
43 hydrologic sensitivity to global change. Such scenarios can be used, in principle, to investigate
Do Not Cite or Quote Page 10
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
44 the potential negative water quality and aquatic ecosystem impacts that we must prepare to
45 remedy, nationally, given existing and likely future vulnerabilities of our aquatic ecosystems.
46
47 But what are these existing vulnerabilities? The idea for this report began with a seemingly
48 simple question: How easy would it be to assess, and map, the relative vulnerability of
49 watersheds, across a number of dimensions, for the whole United States in a meaningful, self-
50 consistent way? In this report, we summarize the lessons learned to date in our attempts to
51 answer this question.
52
53 There are two main outcomes that we report on here. First, we have collected, evaluated the
54 quality of, processed, and aggregated a large quantity of data on water quality and aquatic
55 ecosystem indicators across the nation. Second, we have attempted to identify best practices,
56 challenges, and gaps in ideas, methods, data, and tools for calculating and mapping vulnerability
57 nationally. In both contexts, we hope that this report will be a useful building block for future
58 work on multi-stressor global change vulnerability assessments.
59
60 To measure relative vulnerability, we identified indicators that reflect the three components of
61 vulnerability as identified by the IPCC (2007a): sensitivity, exposure, and adaptive capacity.
62 Sensitivity is the extent to which a system responds either positively or negatively to external
63 stimuli; exposure is the degree to which a system is exposed to stressors (and in some cases,
64 specifically climatic variations); and adaptive capacity is the ability of a system to cope with
65 stress. Most vulnerability indicators identified in this report measure the exposure or sensitivity
66 of water quality and aquatic ecosystems to stressors. An understanding of exposure and
67 sensitivity may facilitate the development of adaptive capacity within a system.
68
69 It is important to clarify here that this report does not evaluate impacts of climate change on
70 ecosystems and watersheds. Instead, it deals only with the question of how to estimate the
71 relative effects of other, existing stressors and their potential to reduce overall resilience, or
72 increase overall sensitivity, to climate change. It examines this question by looking at indicators
73 of vulnerability to such stressors. We argue that a systematic evaluation of the impacts of
74 existing stressors is a key input to any comprehensive climate change vulnerability assessment,
75 as the impacts of climate change will be expressed via interaction with such stressors.
76
77 While the idea that existing stressors reduce resilience and increase vulnerability to climate
78 change remains an assumption for many systems, it is an established one, deeply embedded in
79 recent large climate change assessment efforts. For example, the IPCC 4th Assessment Working
80 Group II report states that: "Vulnerability of ecosystems and species is partly a function of the
81 expected rapid rate of climate change relative to the resilience of many such systems. However,
82 multiple stressors are significant in this system, as vulnerability is also a function of human
83 development, which has already substantially reduced the resilience of ecosystems and makes
84 many ecosystems and species more vulnerable to climate change through blocked migration
85 routes, fragmented habitats, reduced populations, introduction of alien species and stresses
86 related to pollution" (IPCC, 2007a). It then goes on to provide examples from terrestrial, marine,
87 and coastal ecosystems.
88
89 Reducing the impact of current stressors is also frequently considered to be a "no regrets"
Do Not Cite or Quote Page 11
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
90 adaptation strategy for enhancing ecosystem resilience to climate change. The U.S. Climate
91 Change Science Program (USCCSP, 2008) reviewed adaptation options for six federally
92 managed programs in the United States: national forests, national parks, national wildlife
93 refuges, national estuaries, marine protected areas, and wild and scenic rivers. Adaptation
94 options were studied by reviewing available literature, data, and models, as well as by assessing
95 the consensus within the scientific community. Decreasing current anthropogenic stresses was
96 the adaptation approach the scientific community believed had the greatest chance of success.
97 Numerous studies confirmed that this approach was likely to be the most successful of those
98 considered.
99 The idea that existing stressors reduce resilience and increase vulnerability to climate change
100 informs both the definition of "vulnerability" that we use, and the selection of individual
101 indicators we examine. It is key to providing the link between what these indicators measure and
102 an understanding of the ecological and watershed impacts of climate change, and we expand
103 upon this idea at other points in this report.
104
105 Returning to our framing question, "How easy would it be to assess, and map, the relative
106 vulnerability of watersheds, across a number of dimensions, for the whole United States in a
107 meaningful, self-consistent way?", our strategy for addressing it was as follows:
108
109 We conducted a literature search and compiled a comprehensive list of broadly defined
110 indicators of the vulnerability of water quality or aquatic ecosystems, including those relating to
111 ambient surface and groundwater quality, drinking water quality, ecosystem structure and
112 function, individual species, and the provision of ecosystem services. This then formed the set of
113 indicators for exploring a number of subsequent challenges. These challenges fall into four broad
114 categories:
115
116 1. Challenges associated with identifying those indicators that speak specifically to
117 vulnerability as opposed to those reflecting simply a state or condition. In this context, we
118 define vulnerability as adverse impacts accrued over time and associated with external stresses
119 from, for example, climate or land-use change;
120 2. Challenges associated with calculating and estimating the values of these vulnerability
121 indicators, including establishing important indicator thresholds that reflect abrupt or
122 large changes in the vulnerability of water quality or aquatic ecosystems;
123 3. Challenges associated with mapping these vulnerability indicators nationally, including
124 data availability and spatial aggregation of the data;
125 4. Challenges associated with combining and compositing indicators and developing multi-
126 indicator indices of vulnerability.
127
128 For this work, we relied on published research and on studies by EPA, other federal agencies,
129 and well-respected institutions like the Heinz Center and the Pew Center, both for indicator
130 definitions and for the data to support the mapping of indicators. While each study reviewed had
131 a slightly different objective, much of the information was relevant to the goals of this project.
132 The intent was to examine what could be accomplished with existing indicators and data sets,
133 and for the most part we did not attempt at this point to conceive of new indicators or collect new
134 data. As part of this work we developed a number of example maps, and we use some of these
135 maps in this report for illustrative purposes. We recognize that approaches other than the one we
Do Not Cite or Quote Page
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
136 took are possible, but the lessons we learned while developing strategies for compiling and
137 mapping national-level indicator data sets under this project would likely be useful for an array
138 of alternative approaches. This project was a starting point and its findings have broad
139 applicability.
140
141 The next section (Section II) briefly describes a number of EPA efforts that informed this work,
142 and with which we could usefully integrate the ideas in this report more closely in the future.
143 Section III describes the compilation and examination of the extensive set of indicators for water
144 quality and aquatic ecosystems that was the starting point for the analyses in this report. Sections
145 IV through VII then discuss the four broad categories of challenges described above. We
146 summarize our findings and propose some recommendations in Section VIII. Finally, several
147 appendices document the following: the literature reviewed (Appendix A); the full set of more
148 than 600 indicators initially evaluated (Appendix B); the data sources and supporting information
149 for the 53 vulnerability indicators that were evaluated for data availability and mapping potential
150 (Appendix C); data limitations and technical notes for those 53 indicators (Appendix D); the
151 methodological details for how the various maps were produced (Appendix E); example maps
152 by HUC-4 watershed (Appendix F) and their descriptions (Appendix G); example maps by
153 ecoregion (Appendix H) and their descriptions (Appendix I); vulnerability categories for each
154 indicator by each HUC (Appendix J); steps for evaluating and modifying vulnerability indicators
155 (Appendix K); and the contact information for selected team members (Appendix L).
156 II. Synergies with Other EPA Efforts
157 There are a number of EPA efforts devoted to indicator-based assessment of environmental
158 condition and impairment. This report draws from these efforts in a number of direct and indirect
159 ways. In addition, greater integration of the work described here with these efforts has the
160 potential for a number of significant benefits. Here, we briefly summarize some of these
161 connections.
162
163 The valued role of environmental indicators in environmental resource assessment and
164 management is evidenced in recent years by several prominent reports from both within the
165 government sector and outside it (e.g., Heinz Center 2008). Notably, EPA tracks roughly 83
166 indicators of environmental and human health for its Report on the Environment (USEPA,
167 2008b). For example, Chapter 3 of the ROE is a report card on trends in the extent and condition
168 of the nation's waters (USEPA, 2008b). The ROE indicators are revisited roughly once every
169 three to four months and subsequently updated online to assess changes over time. They are
170 generally reported as national averages or representative examples, rather than as mapped
171 distributions. The long-term goal for the ROE is to report all indicators as temporal trends. The
172 ROE has its roots in the Environmental Monitoring and Assessment Program (USEPA, 2010a), a
173 research program within EPA's Office of Research and Development that was designed to
174 develop the tools necessary to monitor and assess the status and trends of national ecological
175 resources. EMAP collected field data from 1990 to 2006, and focused on developing the
176 scientific understanding for translating environmental monitoring data from multiple spatial and
177 temporal scales into assessments of current ecological condition and forecasts of future risks to
178 our natural resources. We drew a number of the indicators discussed in this report, as well as
179 general indicator definitions, from the ROE.
Do Not Cite or Quote Page 13
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
180
181 Monitoring of the nation's aquatic resources is now conducted by the EPA Office of Water's
182 National Aquatic Resource Surveys (USEPA, 201 Ob), which publishes a series of studies that
183 report on core indicators of water condition. These studies use standardized field and lab
184 methods that are designed to yield unbiased, statistically-representative estimates of the
185 condition of the whole water resource, such as rivers and streams, lakes, ponds, reservoirs, and
186 wetlands. Products of this program include the National Coastal Condition reports, the National
187 Wetland Condition Assessment, the Wadeable Streams Assessment, and a number of other
188 reports. Again, as with the ROE, we drew a number of indicators from these assessments.
189
190 One of the largest and most important efforts within the agency that has relevance for indicator-
191 based work is the Impaired Waters listing (USEPA, 2010c). Section 303(d) of the Clean Water
192 Act (CWA) requires states, territories, and authorized tribes to assess their waters and identify all
193 water bodies (e.g., streams and rivers) that are impaired. Impaired waters are those that do not
194 meet water quality standards because they are too polluted or otherwise degraded. Waters that do
195 not meet state, territory, or tribal Water Quality Standards due to such impairments are placed on
196 the CWA Section 303(d) list, scheduled for Total Maximum Daily Load (TMDL) development,
197 and eventually restored. EPA maintains responsibility for implementing the 303(d) regulations
198 by ensuring that impaired waters lists are developed. All impaired waters information is then
199 provided to the public via EPAs online data system known as ATTAINS (USEPA, 2010d). For
200 this report, we considered using or developing indicators based on the 303(d) impaired waters
201 lists from each state. Our intent was to use these lists to determine the degree to which waters are
202 impaired for a given unit of spatial aggregation and to frame these identified impairments within
203 a vulnerability context. This link has been previously discussed by EPA during evaluations of
204 how water programs may need to adapt to changes in climate - e.g., EPA's National Water
205 Program Strategy: Response to Climate Change report states that warmer air and water
206 temperatures may lead to "increased pollutant concentrations and lower dissolved oxygen levels
207 will result in additional waterbodies not meeting water quality standards and, therefore, being
208 listed as impaired waters requiring a total maximum daily load (TMDL)" (USEPA, 2008c, p. 9).
209 However, we decided to forego using 303(d)-based indicators because of significant gaps in the
210 impaired waters data. According to the EPA ATTAINS database, only 26.4% of the nation's
211 streams and rivers and 42.2% of the nation's lakes and reservoirs have been assessed for
212 impairments, making it difficult to create national-scale indicators. This is compounded by the
213 variation in assessment programs across states. See section VI.A.d and Figure 6 for additional
214 discussion of these issues.
215
216 EPA's Regional Vulnerability Assessment (ReVA) program (USEPA, 2009a) seeks to
217 characterize vulnerability through investigation of ecosystem dynamics, the connectivity
218 between ecosystems and the broader landscape, and ecosystem interactions with socioeconomic
219 factors. The purpose of the ReVA program is to examine the probability of future problems at a
220 regional scale, even when precise environmental conditions at a given location cannot be
221 predicted. The ReVA program also aims to help decision-makers assess the degree and types of
222 stress posed by human actions on a region's environmental resources. The program's
223 methodology evaluates indicators of vulnerability, aggregates them into indices, and evaluates
224 the likelihood of exacerbation of vulnerability as a result of future stressors. To date, the ReVA
Do Not Cite or Quote Page 14
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
225 program's methodology has been applied to a comprehensive analysis of the Mid-Atlantic region
226 (USEPA, 2000b). EPA plans to conduct similar assessments in other regions.
227
228 The ReVA program is an outstanding source of vulnerability metrics and indicators. The present
229 study complements the ReVA program by building on its extensive work on vulnerability and
230 investigating a similar methodology for national scale investigations of vulnerability focused on
231 climate change. Both the ReVA program and the current study present relative measures of
232 vulnerability and identify future research opportunities that would result in measures of absolute
233 vulnerability. Future efforts may include integration of ReVA tools and data with the indicators
234 presented in the current report.
235
236 EPA's just-released 2010 report, Climate Change Indicators in the United States (USEPA,
237 2010e), is a new effort that is intended to track and interpret a set of 24 indicators, each
238 describing trends related to the causes and effects of climate change. It focuses primarily on the
239 United States, but in some cases also examines global trends. EPA intends to begin using these
240 indicators to monitor the effects and impacts of climate change in the United States, assist
241 decision-makers on how to best use policymaking and program resources to respond to climate
242 change, and assist EPA and its constituents in evaluating the success of their climate change
243 efforts. We did not use these indicators in this report, but we envision integrating them with the
244 methodologies discussed here in future efforts to assess vulnerability of water quality and aquatic
245 ecosystems to climate change.
246
247 Finally, there is a pressing need for objective strategies to prioritize agency efforts by comparing
248 different geographic locations in terms of their expected responses to future conditions and
249 various management options. This can be done with regard, for example, to stream restoration
250 (Norton et al., 2009) and to climate change adaptation (Lin and Morefield, 2010). As Norton et
251 al. (2009) write, "Tens of thousands of 303(d)-listed waters, many with completed TMDLs,
252 represent a restoration workload of many years. State TMDL scheduling and implementation
253 decisions influence the choice of waters and the sequence of restoration. Strategies that compare
254 these waters' recovery potential could optimize the gain of ecological resources by restoring
255 promising sites earlier." Norton et al. (2009) then explore ways that states, tribes, and territories
256 can use measurable metrics of ecological, stressor, and social context to estimate the relative
257 recovery potential of sites, as a key input into decisions that set priorities for the selection and
258 sequence of restoration efforts. Similarly, Lin and Morefield (2010), using the Atlantic and Gulf
259 Coast National Estuaries as their example, propose a framework for assessing and prioritizing
260 management recommendations that might be made in response to communities' vulnerability to
261 climate change and their wishes to develop adaptation strategies. In our view, attention to the
262 issues and challenges discussed in this report is likely to aid in the task of developing objective
263 measures that can inform a broad range of prioritization decisions.
264 III. Indicators Considered for this Report
265 This Section describes the approach used to compile a comprehensive list of potential indicators
266 of water quality and aquatic ecosystem vulnerability from those identified in published sources.
267 Figure 1 outlines the general methodology in the selection of indicators for this study.
268
Do Not Cite or Quote Page 15
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
269 A. Literature Search
We performed an extensive literature search to identify recent studies related to the monitoring
and evaluation of water quality and ecosystem conditions. The types of literature reviewed
included journal articles, studies, and reports. The literature ranged widely in study area, from
local to international. It ranged in technical field from biological, hydrological, and chemical, to
human aspects, and included both primary and secondary literature. The literature sources also
varied, including individual researchers, public institutions, and non-governmental organizations.
Studies reviewed spanned a decade of relevant literature from 1998 through 2008.
The literature reviewed was primarily obtained from the GCRP research team members and
through internet and library database searches conducted by Cadmus. Literature identified by
GCRP as relevant was considered to be "core literature" and was given high priority in the
review process. Thereafter, other references were reviewed to identify additional indicators for
possible inclusion. The citations within the core literature were also useful as sources of
additional relevant literature.
a.
Core literature
As noted above, the GCRP research team identified a short list of studies as core literature that
served as a starting point for identifying vulnerability indicators. These studies are listed in Table
1.
289 Table 1. List of Core Literature
List of Core Literature (see Appendix A for full references)
• Coastal States Organization, 2007 •
. Ebietal., 2007
• Frumhoff etal., 2007 •
• Gilliom et al., 2008 •
• Gleick and Adams, 2000 •
• Hamilton et al., 2004 •
• Heinz Center, 2002 •
• Heinz Center, 2008 •
. Hurd et al., 1998
. Hurd et al., 1999
• Lettenmaier et al., 2008 •
• Millennium Ecosystem Assessment, 2005a
Millennium Ecosystem Assessment, 2005b
Millennium Ecosystem Assessment, 2005c
National Assessment Synthesis Team, 2000a
National Assessment Synthesis Team, 2000b
Poff etal., 2002
USEPA, 2006
USEPA, 2008a
USEPA, 2008b
USGAO, 2005
United States Geologic Survey (USGS), 1999
Zogorski, et al., 2006
Do Not Cite or Quote
Page 16
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
290 Figure 1. Flowchart of Methodology Used To Identify and Map Vulnerability Indicators
Step 1: Conduct literature search.
Extensive literature search conducted.
- —4 86 documents
Step 2: Identify indicators of water quality and
aquatic ecosystem condition.
Literature review conducted and indicators of water
quality and aquatic ecosystem condition identified.
Step 3: Delete duplicate indicators
If identical indicators cited by different literature
sources, a single best indicator selected for further
evaluation. Remaining duplicate indicators (which were
either not defined, poorly defined, or were specific to a
geographic region) deleted.
66 indicators \
eliminated
Step 3: Classify indicators of vulnerability
Indicators of vulnerability identified. State variables
(i.e., those measuring condition at a point in time)
I d
504 indicators \
eliminated
Step 4: Assess data availability.
Data sources identified and some indicators eliminated
because: (a) indicator was conceptual/theoretical in
nature (i.e., no data were available); (b) data
collection was in progress; (c) data were not national,
not recent, or were a projection; (d) combination of
multiple data sets entailed complex methods; (e)
indicator required a complex modeled data set; (f)
data required time-intensive manipulation.
I
** indicators \
eliminated
Step 5: Create example maps.
Data obtained and manipulated to create maps using
CIS software for readily mappable indicators.
FINAL 25
MAPPABLE
INDICATORS
Do Not Cite or Quote
Page 17
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
291 Some studies, typically those that were specifically geared towards identifying indicators of
292 ecosystem change or documenting the results of national environmental monitoring studies,
293 served as a source for many of the indicators in this EPA study. Some key studies in the core
294 literature and how they were used are described below.
295
296 • Hurd et al, 1998 and Hurd et al, 1999
297
298 The report, Water Climate Change: A National Assessment of Regional Vulnerability,
299 prepared for EPA by Hurd et al. (1998), identified key aspects of water supply and quality
300 that could be adversely affected by climate change, developed indicators and criteria useful
301 for assessing the vulnerability of regional water resources to climate change, created a
302 regional database of water-sensitive variables consistent with the vulnerability measures, and
303 applied the criteria in a comparative national study of the vulnerability of U.S. water
304 resources. The result of this study was a series of national-scale maps attempting to
305 demonstrate the vulnerability of different U.S. regions to climate change for each indicator of
306 vulnerability of water supply and quality. An abbreviated version of this study, presenting a
307 few select indicators and outlining the general methodology used in creating national-scale
308 maps for each indicator, was later published in the Journal of the American Water Resources
309 Association (Hurd et al., 1999). The spatial resolution of vulnerability estimates used by
310 Hurd et al. (1998) was a 4-digit Hydrologic Unit Code (HUC) or hydrologic subregion, of
311 which there are 222 nationwide.
312
313 • Heinz Center, 2002 and Heinz Center, 2008
314
315 The State of the Nation's Ecosystems 2008: Measuring the Land, Waters, and Living
316 Resources of the United States prepared by the H. John Heinz Center for Science,
317 Economics, and the Environment (hereafter referred to as the Heinz Center), was the most
318 recent publication in an effort aimed at developing a comprehensive evaluation of the
319 condition of the nation's ecosystems. Aspects of this effort were a model for the
320 methodology used in the present study. We also used an older publication from the same
321 effort (Heinz Center, 2002) to incorporate indicators that were not considered in the Heinz
322 Center 2008 study.
323
324 The indicators in the Heinz Center reports often described the state of ecosystem attributes.
325 Because current state was considered a component of vulnerability, the selection of these
326 indicators typically represented the first screening step in identifying useful vulnerability
327 indicators. The state indicators used by the Heinz Center did not explicitly describe stressors
328 that affected those indicators, although stressors were implied for ecosystem attributes that
329 were in a degraded state.
330
331 The Heinz Center described several indicators for which adequate data were not available.
332 We also adopted the approach of identifying ongoing collection efforts or proposing data
333 collection priorities for indicators of potential importance. The Heinz Center report includes
334 terrestrial ecosystem types; the present study does not. However, the "Coasts and Oceans"
335 and "Fresh Waters" sections of the Heinz Center report included many specific indicators
336 that we used here.
Do Not Cite or Quote Page 18
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
337 • USEPA, 2006
338
339 Wadeable Streams Assessment (WSA): A Collaborative Survey of the Nation's Streams
340 summarizes the results of a collaborative effort led by EPA (2006) to provide a statistically
341 defensible report on the condition of the nation's smaller streams. Standardized methods
342 were used to measure several physical, chemical, and biological attributes at 1,392 sites that
343 represent the small streams in the U.S.
344
345 The database that accompanied WSA was used as a data source for mapping several of the
346 indicators in the present study. As with some indicators from the Heinz reports, the measures
347 reported in EPA's WSA report (2006) reflect the current condition of the wadeable streams,
348 rather than their specific vulnerability to future changes.
349
350 • USEPA, 2008b
351
352 As described in Section II, EPA tracks roughly 83 indicators of environmental and human
353 health, and reported on those indicators in U.S. EPA's 2008 Report on the Environment. The
354 Report on the Environment (ROE) is published less frequently in hardcopy form, but
355 continually updated online (www.epa.gov/roe). Chapter 3 of the ROE is a report card on
356 trends in the extent and condition of the nation's waters. The indicators in this report were
357 generally reported as national averages or representative examples, rather than mapped
358 distributions. Some indicators were reported as temporal trends. Indicator data were derived
359 from multiple sources, and no new data were collected as part of this chapter. The indicators
360 in this report are revisited roughly once every three to four months and subsequently updated
361 online to assess changes over time. The ROE provided several indicators for this report.
362 Some ROE indicators of temporal trends are closely tied to the concept of vulnerability.
363
364 • United State Geologic Survey (USGS), 1999
365
366 The Quality of our Nation's Waters: Nutrients and Pesticides, the first summary report from
367 the USGS' National Water-Quality Assessment (NAWQA) program, reports on the
368 geographic distribution, environmental drivers, and temporal trends of nutrients and
369 pesticides in surface waters. The NAWQA data include several useful summary statistics
370 from the broad range of physical and chemical water quality parameters measured as a part
371 of the NAWQA program.
372
373 Under the NAWQA program, 51 sites are broken up into smaller groups that are sampled in
374 multiple rounds (20 study units in 1991; 16 study units in 1994; and 15 study units in 1997).
375 NAWQA is also considered the best source of information on the occurrence of pesticides in
376 surface and groundwater. However, even with the full complement of study units (including
377 units that were not completed at the time of the present study), the spatial coverage of
378 NAWQA sites is relatively sparse. As with most of the literature used in the present study,
379 NAWQA reports primarily on current condition, rather than vulnerability to future change.
Do Not Cite or Quote Page 19
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
380 b. Protocol for collecting additional relevant literature
381 To develop a comprehensive list of indicators cited in the published literature, an extensive and
382 representative sample of recent studies was needed. We conducted a literature search using
383 publicly available (e.g., Google Scholar) and non-public (e.g., Science Direct) search tools to
384 identify studies with a primary or secondary focus on water quality and aquatic ecosystems. We
385 selected studies based on their likelihood of containing water quality and aquatic ecosystem
386 indicators.
387
388 Along with the core literature, we identified 86 studies that could be used as potential sources of
389 indicators, including:
390
391 • 19 government reports;
392 • 40 peer-reviewed journal articles; and
393 • 27 other reports including those by non-governmental or inter-governmental
394 organizations.
395
396 See Appendix A (Bibliography) for a complete list of the reviewed literature.
397
398 B. '• v<
399 We reviewed the literature collected and identified indicators relevant to the present study. This
400 section describes the guidelines we used to identify relevant indicators, and the details of the
401 choices we made to select only certain indicators from particular studies based on these general
402 guidelines.
403
404 We use the term, "indicator" in this report as it is commonly used in the published literature
405 (Villa and McLeod, 2002; Kurd et al., 1998; Adger et al., 2004), to define a variable or a
406 combination of variables that can be used to measure the change in an environmental attribute.
407 Similar terms, such as "metric" are also widely used in the literature (Norton et al., 2009; Luers,
408 2005), while metric and indicator are used interchangeably in other studies (Adger, 2006;
409 Nicholson and Jennings, 2004). For the purposes of this report, we use the terms metric and
410 indicator interchangeably.
411 a. Identifying indicators of water quality and aquatic ecosystem condition
412 We reviewed all of the studies indentified in the literature search to develop a comprehensive list
413 of indicators. Unlike a typical literature review, we reviewed these studies for indicators of water
414 quality and aquatic ecosystem condition, rather than for their contributions to the body of
415 knowledge on this topic. Therefore, they were reviewed for their explicit or implicit description
416 of indicators that could potentially be used to assess the vulnerability of water quality and
417 aquatic ecosystems to environmental change. We selected indicators following the guidelines for
418 good indicators from EPA's Report on the Environment (ROE) as presented in Figure 2
419 (Indicator Definition from EPA's 2008 Report on the Environment).
420
Do Not Cite or Quote Page 20
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
421 Figure 2. Indicator Definition from EPA's 2008 Report on the Environment
• Useful. It answers (or makes an important contribution to answering) a question in the ROE.
• Objective. It is developed and presented in an accurate, clear, complete, and unbiased manner.
• Data Quality. The underlying data are characterized by sound collection methodologies, data management
systems to protect their integrity, and quality assurance procedures.
• Data Availability. Data are available to describe changes or trends, and the latest available data are timely.
• Representative Data. The data are comparable across time and space and representative of the target
population. Trends depicted in this indicator accurately represent the underlying trends in the target
population.
• Transparent and Reproducible Data. The specific data used and the specific assumptions, analytical
methods, and statistical procedures employed are clearly stated.
422
423 This selection process resulted in a comprehensive list of 623 indicators (presented in Appendix
424 B: Comprehensive List of Indicators). Each indicator was assigned a unique indicator
425 identification number (Indicator ID#) - this was necessary given the large number of indicators
426 and to avoid confusion among indicators with similar names. In subsequent sections of this
427 report, each indicator name is associated with its parenthetical ID# (e.g., Acid Neutralizing
428 Capacity [#!]). These identification numbers also facilitate easier referencing of each indicator in
429 the appendices of this report.
430
431 Most water quality and aquatic ecosystem indicators found in the literature were included in the
432 comprehensive list. However, it is important to discuss why we excluded some indicators from
433 this list and chose not to examine them in subsequent steps of this methodology. We discuss
434 these reasons immediately below.
435 b. Selection of indicators
436 In the interest of thoroughness, we made broad determinations regarding whether or not each
437 indicator, measure, or metric in a particular study could be used to characterize, evaluate, or
438 assess water quality or aquatic ecosystems. On the rare occasions when we excluded indicators
439 from a particular study from the comprehensive list, we documented the reasons for such
440 exclusions - for example, indicators related to air quality were generally not considered relevant
441 to this project, and have been well-studied elsewhere. The wide range of characteristics that
442 describe the comprehensive list of indicators for this project can be summarized as follows:
443
444 • Indicators covered a variety of different disciplines;
445 • Indicators were of varying scales, from local to national;
446 • Indicators had varying amounts of data associated with them;
447 • Indicators were aggregated (made up of smaller input indicators) or disaggregated;
448 • Indicators were drinking water indicators or indicators related to aquatic ecosystems;
449 • Some were indicators related to infrastructure; and,
450 • Indicators were potentially important to decision-makers at a variety of levels, ranging
451 from federal, to regional and local levels.
452
Do Not Cite or Quote Page 21
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
453 Indicators included in the list were vetted in the literature, although to varying extents. Some
454 studies focused solely on identifying robust water quality and ecosystem condition indicators that
455 could be used to observe and explain changes in the natural environment. Other studies merely
456 provided a theoretical rationale for more conceptual indicators.
457
458 In addition to selecting specific indicators, we also reviewed the literature to obtain the following
459 indicator-related information:
460
461 • Indicator definition, as specified in the literature, or written based on supporting text in
462 the literature;
463 • Level at which it is adopted (i.e. local, state, or national);
464 • Whether the indicator is currently in use;
465 • Geographic scope (i.e. local, state, or national);
466 • Spatial resolution;
467 • Target audience (e.g., scientists, policymakers, risk analysts); and
468 • Rationale for the indicator's inclusion on the comprehensive list of indicators (based on
469 information in the literature) to corroborate the indicator's relevance as an indicator of
470 the vulnerability of waterbodies to environmental degradation.
471
472 In addition, a team of technical experts classified the potential application of each indicator to
473 climate change as high, medium, or low. These experts, presented in Appendix L (Research
474 Team Members and Contact Information) represent multi-disciplinary fields related to the
475 impacts of climate change on various aspects of human life and the natural environment.
476
477 In addition to the steps described above, we took two specific actions to ensure the most
478 comprehensive indicator list possible:
479
480 • Creation of Indicator Categories
481
482 Different indicators measure different aspects of potential vulnerability. By grouping like
483 indicators, it was possible to determine which aspects of water quality and aquatic
484 ecosystem condition were reasonably covered by the selected indicators and to identify
485 potential coverage gaps. Therefore, to facilitate reviews of the indicator list, we
486 established indicator categories and sub-categories, as shown in Table 2 (Indicator
487 Primary and Secondary Categories).
488
489 • Review of Indicator List by Technical Experts
490
491 To ensure the most comprehensive indicator list possible, technical advisors reviewed a
492 draft list of indicators and were asked to add indicators where they perceived gaps.
493 Through this process, one indicator (Total Withdrawal Information by Source & Type of
494 Use [#622]) was added to the comprehensive list, and a significant amount of additional
495 detail and new information was added for the indicators already in the comprehensive
496 list.
497
Do Not Cite or Quote Page 22
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
498 Table 2. Indicator Primary and Secondary Categories
Ecological (161)
• Condition of Plant Species
• Distribution of Plants
• Exposure to Contaminants
• Habitat Condition
• Non-Native Species
• Species at Risk
• Species Diversity
• Species Populations
Land Cover/Use (61)
• Agricultural
• Coastal
• Forest
• Freshwater
• Glaciers
• Grasslands/Shrublands
• Natural Cover
• Urban/Suburban
• Wetlands
Other (2)1
Hydrological (104)
• Duration of Natural Events
• Engineered Structures
• Precipitation
• Sea Level Rise
• Temperature
• Water Flow
• Water Levels
• Waves
Socioeconomic (57)
• Housing
• Policy
• Recreation
• Resource Use
Air (19)
• Aerosols
• Ozone
• Temperature
Human Populations (14)
• Population Size
• Susceptible Populations
Chemical (96)
• Carbon
• Chlorophyll a
• Contaminants in Sediment
• Microbes
• Multiple Contaminants
• Nutrients
• Oxygen
• Pesticides
• pH
• Salinity
• Turbidity/Clarity
Extreme Weather Events (16)
• Drought
• Fire
• Flood
• Storm
So/7 (27)
• Composition
• Erosion
• Sediment
500
501
502
503
504
505
506
507
508
509
510
511
499 Note: The "Other" category has no secondary categories.
c.
Exclusion of certain indicators and studies
In some cases, we excluded from the comprehensive list particular indicators, groups of
indicators, or all indicators from a particular study. Table 3 (Rationale for Exclusion of Certain
Indicators) presents the rationale for not selecting some indicators from particular studies.
d. Deletion of duplicate indicators
As indicators for the comprehensive list were identified from various literature sources, some
redundancy was noted in some groups of indicators. When two or more indicators were
identified as being very similar, one was selected to represent the group, and the others were
removed from further consideration for mapping. Selected representative indicators were most
often those that had a clear definition, were relevant at the national level (i.e., not limited to a
small geographic region), could be quantified easily, or were obtained from this study's core
Do Not Cite or Quote
Page 23
-------
512
513
514
515
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
literature sources. Sixty six indicators were deleted because they were redundant with other
indicators in the comprehensive list.
Table 3. Rationale for Exclusion of Certain Indicators
Reasons for Exclusion of
Indicators
Literature Sources
Indicators were modeled
projections, specific to a non-U.S.
location, or were too broadly
defined.
Arnell, 1998
Arnell, 1999
Barnett et al., 2005
Bergstrom et al., 2001
Conway and Hulme, 1996
de Wit and Stankiewicz, 2006
Gleick and Adams, 2000
Kundzewicz et al., 2008
Lettenmaier et al., 2008
Nichollsand Hoozemans, 1996
Palmer etal., 2008
Roderick and Farquhar, 2002
Indicators were of human
adaptive capacity or
socioeconomic indicators, rather
than of aquatic ecosystems or
water quality.
Adger at al., 2004
Brooks et al., 2005
Ebi etal., 2007
Frumhoff et al., 2006
Frumhoff at al., 2007
Gleick and Adams, 2000
Jacobs et al., 2000
Kling et al., 2003
Millennium Ecosystem
Assessment, 2005a
Millennium Ecosystem
Assessment, 2005b
Twilley et al., 2001
Indicators were identical or very
similar to those in another study,
or indicators were better defined
in another study.
Bradbury et al., 2002
Bunn and Arthington, 2002
Chesapeake Bay Program, 2008
Dai et al., 1999
Frumhoff etal.,2007
Grimm etal., 1997
Hamilton et al., 2004
Hayslip et al., 2006
Huntington et al., 2004
Hurd etal., 1998
Kling et al., 2003
Long Island Sound Study, 2008
Ojima etal., 1999
USEPA, 1995
USEPA, 2002
Zogorski et al., 2006
Indicators and their associated
data sources were not adequately
detailed as the study was
primarily a policy/funding-
oriented document.
Coastal States Organization, 2007
Luers et al., 2006
Murdoch etal., 1999
National Assessment Synthesis
Team, 2000b
Poffetal., 2002
USEPA, 2008c
USGAO, 2000
USGAO, 2002
USGAO, 2004
USGAO, 2005
Vincent and Pienitz, 2006
Yamin etal., 2005
Indicators were large aggregates
of smaller indicators.
Gleick and Adams, 2000
USEPA, 2008d
517
518
519
520
521
522
523
516 IV. Challenges Part I: Indicator Classification
This section describes how we evaluated the indicators introduced in the previous section to
determine whether they were suitable, in principle, for assessing relative vulnerability to large-
scale environmental degradation due to external stressors (of which climate change would be one
example). First we considered how to define vulnerability. We then applied that definition to
each of the 623 indicators that resulted from the process described in the previous section,
resulting in a small subset being classified as "vulnerability" indicators.
Do Not Cite or Quote
Page 24
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
524 A.
525 There has been considerable debate in the literature on the meaning of vulnerability in the
526 context of environmental systems and stressors (climate change in particular) and the elements of
527 which it is composed. We summarize some of that discussion here as background.
528
529 It has been argued that the lack of a common definition has hindered interdisciplinary discourse
530 on the topic and the development of a common framework for vulnerability assessments
531 (Brooks, 2003; Fiissel, 2007). Others have argued that the purpose of the analysis should guide
532 the selection of the most effective definition or conceptualization (Kelly and Adger, 2000).
533
534 Some of the purposes for which climate change vulnerability assessments may be performed
535 include: increasing the scientific understanding of climate-sensitive systems under changing
536 climate conditions; informing the specification of targets for the mitigation of climate change;
537 prioritizing political and research efforts to particularly vulnerable sectors and regions; and
538 developing adaptation strategies that reduce climate-sensitive risks independent of their
539 attribution. Each of these purposes has specific information needs and thus might require a
540 targeted approach to provide this information.
541
542 Below is a summary of discussions about the definition of vulnerability in the literature on
543 climate change, including:
544 • Determinants of vulnerability
545 • Defining a vulnerable situation
546 • Biophysical and socioeconomic domains
547 • Predictability and uncertainty
548 a. Determinants of vulnerability
549 The IPCC definition of vulnerability is: "The degree to which a system is susceptible to, or
550 unable to cope with, adverse effects of climate change, including climate variability and
551 extremes. Vulnerability is a function of the character, magnitude, and rate of climate variation to
552 which a system is exposed, its sensitivity, and its adaptive capacity." (IPCC, 2007a, p. 995)
553 (IPCC Def 1). Three terms are defined further in the IPCC report: sensitivity, exposure, and
554 adaptive capacity.
555
556 The IPCC defines sensitivity as "the degree to which a system is affected, either adversely or
557 beneficially, by climate-related stimuli." This definition is generally supported by much of the
558 literature on the topic, but there are two subtly different interpretations. The first considers
559 sensitivity as the probability or likelihood of passing a critical threshold in a variable of interest
560 (e.g., the probability of exhausting water supplies) (Jones, 2001; Fraser, 2003). The second
561 considers sensitivity to be the degree to which outputs or attributes change in response to
562 changes in climate inputs (Moss et al., 2001). This second interpretation incorporates an
563 understanding that some stresses may increase gradually, instead of emphasizing the passing of
564 one critical threshold value as the only kind of important change. In both cases, a system's
565 sensitivity to stress is separate from its exposure to stress.
566
Do Not Cite or Quote Page 25
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
567 Similarly, exposure is "The nature and degree to which a system is exposed to significant
568 climatic variations." A system may be currently exposed (or predicted to be exposed in the
569 future) to significant climatic variations. Because there are multiple factors related to climate and
570 climate change that may cause stress (e.g., temperature, precipitation, winds, changes in spatial
571 and temporal variability and extremes, etc.), the type of exposure ("hazard" in Fiissel's (2007)
572 terminology) should be specified. In this definition, exposure is separate from sensitivity. A
573 system may be exposed to significant climate changes, but if it is not sensitive to those changes,
574 it is not vulnerable. The socioeconomic literature on vulnerability tends to lump these factors
575 together (e.g., "Social vulnerability to climate change is defined as the exposure of groups or
576 individuals to stress as a result of the impacts of climate change" (Adger, 1999)).
577
578 Finally, adaptive capacity is "The ability of a system to adjust to climate change (including
579 climate variability and extremes) to moderate potential damages, to take advantage of
580 opportunities, or to cope with the consequences." In the socioeconomic literature, vulnerability is
581 often defined primarily by adaptive capacity, particularly as it is linked to poverty (e.g., "...the
582 vulnerability of any individual or social grouping to some particular form of natural hazard is
583 determined primarily by their existent state, that is, by their capacity to respond to that hazard,
584 rather than by what may or may not happen in the future." Kelly and Adger, 2000; see also
585 Olmos, 2001; and Tompkins and Adger, 2004). This conceptualization views sensitivity to most
586 hazards as a given, exposure to some hazard(s) as inevitable, and therefore the need for
587 adaptation will arrive sooner or later. Other authors have argued that because adaptive capacity is
588 not necessarily static (i.e., it can be developed), vulnerability assessments should focus on
589 sensitivity and exposure, with the goal of identifying locations to focus the development of
590 adaptive strategies (Kelly and Adger, 2000; O'Brien et al., 2004).
591 b. Defining a vulnerable situation
592 There is general agreement in the literature that the term, "vulnerability," by itself, may not be
593 sufficiently descriptive (Brooks, 2003; Fiissel, 2007; Polsky et al., 2007; Moreno and Becken,
594 2009). Instead, a vulnerable situation should be defined. This definition should include the
595 following components (Fiissel 2007):
596
597 • Temporal reference: the point in time or time period of interest. Specifying a temporal
598 reference is particularly important when the risk to a system is expected to change
599 significantly during the time horizon of a vulnerability assessment, such as for long-term
600 estimates of climate change.
601 • Sphere: Internal (or 'endogenous' or 'in place') vulnerability factors refer to properties
602 of the vulnerable system or community itself, whereas external (or 'exogenous' or
603 'beyond place') vulnerability factors refer to something outside the vulnerable system
604 that adds to the vulnerability of the system.
605 • Knowledge domain: socioeconomic (e.g., poverty) vs. biophysical (e.g., flow regime
606 sustainability).
607 • System: the system of analysis, such as a coupled human-environment system, a
608 population group, an economic sector, a geographical region, or a natural system.
609 • Attribute of concern: the valued attributes of the vulnerable system that are threatened
610 by its exposure to a hazard. Examples of attributes of concern include human lives and
Do Not Cite or Quote Page 26
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
611 health; the existence, income and cultural identity of a community; and the biodiversity,
612 carbon sequestration potential, and timber productivity of a forest ecosystem.
613 • Hazard: a potentially damaging physical event, phenomenon, or human activity that may
614 cause the loss of life or injury, property damage, social and economic disruption, or
615 environmental degradation.
616
617 An example of a fully specified vulnerable situation is: 'vulnerability of the incomes of the
618 residents of a specific watershed to drought'. In practice, only the components of the definition
619 that are not clear from the context (or uniformly applied to multiple situations) need be defined.
620 The advantage of a specific definition of a vulnerable situation is that it is unambiguous. The
621 disadvantage is that it makes it difficult to conduct holistic vulnerability comparisons among
622 locations.
623 c. Biophysical and socioeconomic domains
624 In the climate change literature, the term "vulnerability" has more frequently been applied to
625 socioeconomic situations; the term "risk" has been used to describe biophysical condition
626 situations (e.g., Jones, 2001). Biophysical vulnerability or risk is primarily related to sensitivity
627 and exposure, while socioeconomic vulnerability is more a function of adaptive capacity.
628 Biophysical vulnerability may encompass effects on humans, such as increase in population at
629 risk of flooding due to sea level rise. However, it is related to human exposure to hazard rather
630 than to the ability of people to cope with hazards once they occur (Brooks, 2003). The view of
631 vulnerability as a state (i.e., as a variable describing the internal state of a system) has arisen
632 from studies of the structural factors that make human societies and communities susceptible to
633 damage from external hazards. Social vulnerability encompasses all those properties of a system
634 independent of the hazards to which it is exposed that mediate the outcome of a hazardous event
635 (Brooks, 2003). In theory, this idea could be applied to biophysical systems, inasmuch as
636 previous stress has rendered the system more susceptible to any new hazard.
637
638 Most of what we define as "vulnerability indicators" in this report are biophysical indicators.
639 They therefore primarily encompass sensitivity and exposure to environmental stresses. Adaptive
640 capacity can be developed in locations that are sensitive and exposed to stress. In addition, while
641 much of the literature on ecosystem vulnerability, particularly as it relates to climate change,
642 focuses exclusively on the degradation of ecosystem components that directly serve human needs
643 (Fiissel, 2007), several of the indicators in this report focus on the direct, inherent vulnerability
644 of the aquatic ecosystems themselves, independent of the ecosystem services provided to
645 humans. We also examine other indicators that focus on the vulnerability of drinking water
646 quality, and are thus more obviously and directly related to human needs.
647 d. Predictability and uncertainty
648 The future behavior of socio-ecological systems is difficult, or perhaps impossible, to predict
649 because the components of these systems are constantly adapting to changing conditions. As a
650 result, a system may contain non-linearities, inter-dependencies, and feedback loops that make
651 its overall behavior unpredictable (Holling, 2001, Fraser et al. 2003, Moreno and Becken 2009).
652 A vulnerability assessment itself may reduce future vulnerabilities by helping target the
653 development of adaptive capacity in systems that are sensitive and exposed to external stressors
654 such as climate change.
Do Not Cite or Quote Page 27
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
655
656 For climate change in particular, many of the adverse effects on ecosystems and human systems
657 are expected to occur as a result of stochastic events that may or may not happen, but to which a
658 subjective probability of occurrence could in principle be assigned. Because these probabilities
659 are conditioned on, for example, predictions of future climate and on models of how the system
660 will respond to climate changes (Jones et al., 2001), it may not be possible to constrain them very
661 much given the current limitations of climate prediction, as discussed in the Introduction. This
662 report focuses on the challenges associated with assessing vulnerability across the nation without
663 depending on accurate environmental prediction. That is, for most of the report we evaluate the
664 vulnerability of water quality and aquatic ecosystems in the absence of specific future scenarios
665 of global climate, population, and land use changes. This bottom-up approach of focusing on
666 indicators vetted in the scientific literature, available data, and current vulnerability, can be used
667 in follow-up studies in combination with approaches focused on improving our ability to predict
668 environmental changes.
669
670 B.
671 In the early phases of this project, we held a workshop1 to develop rules of thumb for classifying
672 the comprehensive suite of 623 indicators into two broad categories. The first category is
673 "vulnerability indicators" that, at least in principle, could measure the degree to which the
674 resource being considered (e.g., watershed, ecosystem, human population) is susceptible to, and
675 unable to cope with, adverse effects of externally forced change. Such change could potentially
676 include climate or any other global change stressor. The second category constitutes state
677 variables or indicators of condition that merely measure the current state of a resource without
678 relating it to vulnerability.
679
680 Informed by the literature above, the workshop participants concluded that, in practical terms, to
681 qualify as a measure of "vulnerability," an indicator should inherently include some relative or
682 value judgment. Examples include comparing one watershed to another, comparing the indicator
683 to some objectively defined threshold or possible state, or reporting on the indicator's change
684 over time. Measures of water quality or ecological condition at a point in time without reference
685 to a baseline would not make good vulnerability indicators. Viewed from the perspective of
686 indicator measurement, this can be achieved by such methods as computing a ratio of two
687 quantities, at least one of which is a time rate of change or a measure of variation, or computing
688 the portion of a distribution that lies above or below a defined threshold. Examples abound,
689 including the ratio of the standard deviation of annual streamflow to mean annual streamflow (to
690 measure degree of variability in the stream), the ratio of stream withdrawals of water to mean
691 annual streamflow (to measure the portion of the flow that is being used), the ratio of mean
692 annual baseflow to mean annual total flow (to measure the susceptibility to dry periods), and the
693 average number of days in a year that a metric such as temperature, dissolved oxygen, or salinity
694 in coastal wetlands exceeds a particular threshold.
695
1 The workshop took place at the National Center for Environmental Assessment (NCEA), in Washington, DC, on
December 18, 2008. Participants included members of the Cadmus team, members of the EPA Global Change
Research Program (GCRP) staff from NCEA, and the outside expert consultants acknowledged in this report.
Do Not Cite or Quote Page 28
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
Applying these rules of thumb is straightforward for some of the indicators and less so for others.
Many could arguably fall into either the "vulnerability" or the "state" category. For example,
when assessing vulnerability to flooding, we might examine the total number of people living
within the 100- or 500-year floodplain in a given watershed; when measuring ecosystem health,
we might look at the total number of species in each watershed classified as "at risk." The key
for these examples is that, by embedding an implied threshold in these indicators - i.e., by
choosing the particular flood frequency (e.g., 100-year or 500-year) that we consider to be
damaging, or a particular classification of "at risk" - we have made a judgment about the system
that goes beyond assessing its condition to assessing its susceptibility to harm. Not all
vulnerability indicators incorporate implied thresholds, and those that vary over a gradual
gradient are still of great value and can inform assessments of relative vulnerability, as discussed
in Section V. A.
This classification exercise winnowed the original list of 623 indicators down to 53 indicators
shown in Table 4 (List of Vulnerability Indicators). Examples illustrating these classification
principles include the following:
Vulnerability Indicators:
• Stream Habitat Quality (#284) - compares stream habitat conditions in a given area to
those in a relatively undisturbed habitat in a similar ecosystem;
• Groundwater Depletion (#121) - compares the average groundwater withdrawals to
annual average baseflow, reflecting the extent to which groundwater use rates may be
exceeding recharge.
• Wetland Species At-Risk (#326) - examines the number of threatened and endangered
species inhabiting a particular wetland area.
State Variables:
• Nitrogen and Phosphorus - large rivers (#186) - measurement of nitrogen and phosphorus
in all streams without a reference value.
• In-stream fish habitat (#138) - a measure of in-stream fish concealment features (e.g.,
undercut banks, boulders, large pieces of wood, brush) within a stream and along its
banks, without specifying reference conditions, such as, for example, concealment
features at undisturbed sites.
Table 4. List of Vulnerability Indicators
Indicator
(See Appendix B for definitions)
Acid Neutralizing Capacity (ANC) (#1)
Altered Freshwater Ecosystems (percent miles changed) (#17)
At-Risk Freshwater Plant Communities (#22)
At-Risk Native Freshwater Species (#24)
At-Risk native marine species (relative risk) (#27)
Coastal Vulnerability Index (to sea level rise) - CVI (#51)
Literature Source
(See Appendix A for full
citations)
USEPA, 2006.
Heinz Center, 2008.
Heinz Center, 2008.
Heinz Center, 2008.
Heinz Center, 2008.
Day et al., 2005.
Do Not Cite or Quote
Page 29
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
(See Appendix B for definitions)
Commercially important fish stocks (size) (#55)
Fish and Bottom-Dwelling Animals (comparison to baseline) (#95)
Flood events (frequency) (#100)
Freshwater Rivers and Streams with Low Index of Biological Integrity
(ecosystem condition) (#116)
Groundwater Depletion - Ratio of Withdrawals/ Baseflow (#121)
Groundwater reliance (#125)
Harmful algal blooms (occurrence) (#127)
Invasive species - Coasts affected (area, ecosystem condition) (#145)
Invasive species in estuaries (percent influenced) (#149)
Low flow sensitivity (mean baseflow) (#159)
Meteorological drought indices (#165)
Number of Dry Periods in Grassland/Shrubland Streams and Rivers (Percent
of streams with dry periods over time) (#190)
Ratio of Snow to Precipitation (S/P) (#218)
Ratio of water withdrawals to annual stream flow (level of development)
(#219)
Riparian Condition (Riparian Condition Index) (#231)
Status of Animal Communities in Urban and Suburban Streams (Percent of
urban/suburban sites with undisturbed and disturbed species) (#276)
Stream flow variability (annual) (#279)
Stream habitat quality (#284)
Water Clarity Index (real vs. reference) (#318)
Water Quality Index (5 components) (#319)
Waterborne human disease outbreaks (events) (#322)
Wetland loss (#325)
Wetland and freshwater species at risk (number of species) (#326)
Ratio of water use to safe yield (#328)
Erosion rate (#348)
Instream use/total streamflow (#351)
Total use/total streamflow (#352)
Snowmelt reliance (#361)
Pesticide toxicity index (#364)
Population Susceptible to Flood Risk (#209)
Literature Source
(See Appendix A for full
citations)
Heinz Center, 2008.
Heinz Center, 2008.
Lettenmaier et al., 2008.
Heinz Center, 2008.
Hurd et al., 1998
Hurd et al., 1998
Heinz Center, 2008.
Heinz Center, 2008.
Heinz Center, 2008.
Hurd et al., 1998
Jacobs etal., 2000.
Heinz Center, 2008.
Lettenmaier et al., 2008.
Hurd et al., 1998
Heinz Center, 2008.
Heinz Center, 2008.
Hurd et al., 1998
Heinz Center, 2008.
NEP, 2006.
NEP, 2006.
Heinz Center, 2008.
MEA, 2005.
Hurd et al., 1998
Schmitt et al, 2008.
Murdoch et al., 2000.
Meyer etal., 1999.
Meyer etal., 1999.
IPCC, 2007.
USGS, 2006.
Hurd et al., 1998.
Do Not Cite or Quote
Page 30
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
(See Appendix B for definitions)
Herbicide concentrations in streams (#367)
Insecticide concentrations in streams (#369)
Organochlorines in Bed Sediment (#371)
Herbicides in Groundwater (#373)
Insecticides in Groundwater (#374)
Salinity intrusion (coastal wetlands) (#391)
Heat-Related Illnesses Incidence (#392)
Precipitation Elasticity of Streamflow (#437)
Ratio of reservoir storage to mean annual runoff (#449)
Runoff Variability (#453)
Macroinvertebrate Index of Biotic Condition (#460)
Macroinvertebrate Observed/Expected (O/E) Ratio of Taxa Loss (#461)
Coastal Benthic Communities (#462)
Threatened & Endangered Plant Species (#467)
Vegetation Indices of Biotic Integrity (IBI) (#475)
In-stream Connectivity (#620)
Water Availability: Net Streamflow per capita (#623)
Literature Source
(See Appendix A for full
citations)
USGS, 1999.
USGS, 1999.
USGS, 1999.
USGS, 1999.
USGS, 1999.
Poffetal., 2002
Pew Center, 2007.
Sankarasubramanian et al.,
2001.
Lettenmaier et al., 2008.
Lettenmaier et al., 2008.
USEPA, 2006.
USEPA, 2006.
USEPA, 2008.
USEPA, 2008.
USEPA, 2008.
Heinz Center, 2008.
Hurd et al., 1998
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
All of the indicators listed in Table 4 were further examined for data availability and
mappability, as discussed in detail in Section VI.
C. How do These Indicators Reflect Vulnerability?
All of the 53 vulnerability indicators vary in their responses to environmental stress and in the
degrees to which they reflect vulnerability of water quality and aquatic ecosystems. Here we
discuss, for the subset of 25 vulnerability indicators that were mappable at the national scale,
how the literature characterizes the link between each indicator and the potential vulnerability of
ecosystems or human systems.
Acid Neutralizing Capacity (#1)
The Acid Neutralizing Capacity or ANC (#1) indicator is a measure of the ability of stream water
to buffer acidic inputs (USEPA, 2006). Streams may be naturally acidic due to the presence of
dissolved organic compounds (USEPA, 2006). However, acid deposition arising from
anthropogenic sources may increase the acidity of the stream (USEPA, 2006). Acid mine
drainage, formed by water passing through mines and mine tailings, is the primary source of acid
in surface water, and results in the formation of concentrated sulfuric acid. Acidity is also caused
by acid rain formed by dissolution of industrial and automotive emissions, such as nitrogen oxide
Do Not Cite or Quote
Page 31
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
750 and sulfur dioxide, in rain water (USEPA, 2006). These acidic inputs may lower the pH of a
751 stream with lower ANC, thereby affecting aquatic vegetation and organisms, as well as water
752 quality, particularly in sensitive watersheds. Changes in precipitation due to global climate
753 change may result in increased acid deposition or drainage from acid mines. Areas with a low
754 percentage of streams with suitable buffering capacity could experience disproportionately large
755 adverse effects resulting from increased acid exposure. In contrast, well-buffered streams with
756 higher ANC may not be as sensitive to increased acidity from external sources.
757
758 The ANC indicator is represented by the percent of stream sites that have been deemed to be at
759 risk, i.e., that have ANC values of 100 milliequivalents or less. This indicator is measured
760 relative to a baseline condition of 100 milliequivalents, such that sites with ANC values below
761 this level are considered vulnerable. The data used to map this indicator were collected every
762 five years.
763
764 At-Risk Freshwater Plant Communities (#22)
765 This indicator describes the risk of elimination faced by wetland and riparian plant communities.
766 The condition of these communities is considered important because of the ecosystem services
767 they provide, including habitat for a variety of species, flood storage, water quality
768 improvements, carbon storage, and other benefits (Heinz Center, 2008; NRC, 1992; Johnson et
769 al., 2007). Loss of community types reduces ecological diversity and may eliminate habitat for
770 rare and endangered species. At-risk status is a vulnerability indicator for aquatic ecosystems by
771 definition, identifying communities that may have less resistance to stressors because they are
772 already compromised.
773
774 Identifying which communities are at risk and their degree of endangerment is useful for
775 planning conservation measures (Grossman et al., 1998). The Heinz Center (2008) describes
776 three risk categories: vulnerable (moderate risk), imperiled (high risk), and critically imperiled
777 (very high risk). Factors that were used to assign these risk categories include range, the number
778 of occurrences, whether steep declines have occurred, and other threats.
779
780 A number of environmental changes might alter the risk status of a plant community. Changes in
781 land use and climate-related changes may decrease the range of a given plant community. The
782 ranges of some plants may shift with temperature changes. Drying would reduce the ranges of
783 some plants, but increased precipitation may allow some species to expand their ranges. Sea
784 level rise associated with global climate change or a reduction in the input of freshwater may
785 allow drought-resistant or salt-resistant plants to move into areas once dominated by freshwater
786 plants (Lucier et al., 2006). Many potential effects on at-risk freshwater plant communities are
787 poorly understood, including alterations in biogeochemical cycling and the effects of increased
788 severity of storms.
789
790 At-risk Native Freshwater Species (#24)
791 Similar to the previous entry, this indicator describes the risk of extinction faced by 4,100 native
792 freshwater species, including fish, aquatic mammals, aquatic birds, reptiles and amphibians,
793 mussels, snails; crayfishes, shrimp, and insects (Heinz Center, 2008). Plants are not included.
794 The status of these species is important because of their value both individually (e.g., as food or
795 for other purposes) and as part of aquatic ecosystems. The at-risk status assigned to these species
Do Not Cite or Quote Page 32
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
796 again directly reflects vulnerability, identifying organisms that may have less resistance to
797 stressors because they are already compromised and have experienced a decline; further declines
798 for some may result in extreme rarity or even extinction.
799
800 The Heinz Center (2008) describes four risk categories: vulnerable, imperiled, critically
801 imperiled, and extinct. Assignment to the "vulnerable," "imperiled," and "critically imperiled"
802 categories is based on up to twelve factors, including population size, number of populations,
803 range, steep or widespread decline, or other evidence of risk. A number of external stressors
804 might affect risk category. For example, changes in the hydrologic cycle, whether induced by
805 climate or land-use change, may reduce available habitat and alter the range and number of
806 locations where species occur. Sea level rise may flood freshwater habitats. Degradation of water
807 quality and presence of certain contaminants may affect the health and long-term stability of
808 sensitive species. If habitat is already fragmented by land use, further stress may further
809 endanger freshwater species.
810
811 Various taxa may be sensitive to environmental change, including climate change. Fish are
812 sensitive to temperature, and changes in temperature may shift the ranges of some species,
813 possibly causing local extinctions (Fiske et al., 2005). Changes in water chemistry and limnology
814 may also affect fish. For example, increased temperature reduces dissolved oxygen and increases
815 thermal stratification (Fiske et al., 2005). Some amphibians may experience reproductive issues,
816 such as interference with their life cycles or temperature effects on gender determination (Lind,
817 undated). Climate-related changes in the ranges of pathogens or increases in emerging pathogens
818 may also endanger freshwater species.
819
820 Coastal Vulnerability Index (#51)
821 The Coastal Vulnerability Index, created by Thieler and Hammar-Klose (2000), is intended to be
822 a measure of the relative vulnerability of U.S. coastal areas to the physical changes caused by
823 relative sea-level rise (RSLR) (Thieler and Hammar-Klose, 2000). RSLR, exacerbated by long-
824 term temperature increases, is expected to increase flooding duration as well as salinity stress
825 caused by saltwater intrusion (Mendelssohn and Morris, 2000, as cited in Day et al., 2005).
826 These factors, in turn, will lead to increased RSLR, destroying coastal wetlands which may not
827 be able to accrete upwards at the same rate (Day et al., 2005).
828
829 The CVI at a particular location is calculated based on the values of six variables at that location:
830 geomorphology, coastal slope, rate of RSLR, shoreline erosion and accretion rates, mean tidal
831 range, and mean wave height (Thieler and Hammar-Klose, 2000). Each location on the coastline
832 is assigned a risk value between 1 (low risk) and 6 (high risk) for each data variable. The CVI is
833 then calculated as the square root of the product of the ranked variables divided by the total
834 number of variables: CVI = [(a*b*c*d*e*f*)/6)]Al/2. Thus, a higher value of the CVI indicates a
835 higher vulnerability of coast at that location. The data for each of the six variables used to map
836 this indicator were collected at various frequencies.
837
838 The CVI changes based on changes in the following variables (see Thieler and Hammar-Klose,
839 2000):
840 • Geomorphology, which is a measure of the relative erodibility of different landforms.
841 Landforms may be of the following types, listed in order of increasing vulnerability to
Do Not Cite or Quote Page 33
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
842 erosion or increasing value of CVI: rocky, cliffed coasts, fiords, or fiards; medium cliffs
843 or indented coasts; low cliffs, glacial drifts, or alluvial plains; cobble beaches, estuaries,
844 or lagoons; barrier beaches, sand beaches, salt marshes, mud flats, deltas, mangroves, or
845 coral reefs. For instance, the value of the CVI is relatively higher along the Louisiana
846 coast due to its lower-lying beaches and marshy areas with shallow slopes that are more
847 prone to erosion.
848 • Coastal slope (percentage), which is a measure of the relative risk of inundation and of
849 the rate of shoreline retreat. Shallower slopes are more vulnerable as they retreat faster
850 than steeper ones, and will result in a higher value of the CVI. The lower and upper
851 bounds for the coastal slope are <0.025% and >0.2% for the Atlantic Coast, <0.022% and
852 >0.115% for the Gulf Coast, and <0.6% and >1.9% for the Pacific Coast.
853 • Rate of RSLR (mm/year), which is the change in mean water elevation at the coast.
854 Higher rates of RSLR, resulting in a higher value of the CVI, cause loss of land and
855 destruction of the coastal ecosystem. The lower and upper bounds for RSLR are <1.8
856 mm/yr and >3.16 mm/yr for the Atlantic Coast, <1.8 mm/yr and >3.4 mm/yr for the Gulf
857 Coast, and <-1.21 mm/yr and >1.36 mm/yr for the Pacific Coast. In contrast, the value of
858 CVI is relatively lower along the Eastern Gulf of Mexico coast mostly due to lower rates
859 ofRLSR.
860 • Shoreline erosion and accretion rates (m/year), which is the rate at which the shoreline
861 changes due to erosion or sediment deposition. Positive accretion rates (resulting in lower
862 values of the CVI) lead to more stable shorelines that are less vulnerable to erosion, while
863 positive erosion rates (resulting in higher values of the CVI) lead to loss of coastal land.
864 The lower and upper bounds for shoreline erosion or accretion rates are <-2.0 m/yr
865 (erosion) and >2.0 (accretion) for all U.S. coasts.
866 • Mean tidal range (m), which is the average distance between high tide and low tide.
867 Coastal areas that have higher tidal ranges (resulting in lower CVI values) are less
868 vulnerable to sea-level rise (Kirwan and Guntenspergen, 2010). The lower and upper
869 bounds for mean tidal range are <1.0 m and >6.0 for all U.S. coasts.
870 • Mean wave height (m), which is a measure of the energy of the wave. A higher energy
871 wave (resulting in higher values of CVI) has a greater tendency to mobilize sediments
872 along the coasts, thereby increasing erosion. The lower and upper bounds for mean wave
873 height are <0.55 m and >1.25 for the Atlantic Coast and the Gulf Coast, and <1.1 and
874 >2.60 for the Pacific Coast.
875
876 The CVI is, as noted above, a direct measure of the vulnerability of coastal ecosystems to RSLR
877 induced by climate change, and it also captures a change in the ecological condition of the
878 coastal area with respect to previous conditions (e.g., lower sea-levels).
879
880 Erosion Rate (#348)
881 Erosion rate is a measure of the rate of long-term soil loss due to erosion. Land use patterns, such
882 the use of land for agricultural purposes or deforestation, can also cause erosion (Yang et al.,
883 2002). Increased precipitation and greater storm intensities induced by global climate change
884 may result in increased transport of sediment, leading to higher erosion rates. Soil erosion is a
885 major non-point pollution source of surface water (Yang et al., 2002). Erosion from runoff
886 events may cause higher levels of nutrients, dissolved organic carbon, and sediment loads in
887 surface water sources (Murdoch et al., 2000). The Erosion Rate indicator can, thus, be used to
Do Not Cite or Quote Page 34
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
888 assess differences in the potential vulnerability of surface water sources as a result of erosion
889 effects.
890
891 The Erosion Rate can be estimated using Yang et al.'s (2002) Revised Universal Soil Loss
892 Equation (RUSLE). This estimate is based on four independent variables: rainfall erosivity, soil
893 erodibility, topography, and vegetation. This indicator only takes into account soil erosion
894 caused by rainfall and flowing water, and for a grid cell with coordinates (i, j) it can be
895 calculated as follows (Yang et al., 2002):
896 A(i,j) = R(I,j)xLS(i,j)xK(i,j)xC(i,j)xP(i,j)
897 where R = average rainfall erosivity factor
898 LS = average topographical parameter
899 K = average soil erodibility factor
900 C = average land cover and management factor
901 P = average conservation practice factor
902 These variables affect the Erosion Rate in the following manner:
903 • Average topographical parameter is a measure of the slope length and steepness. Erosion
904 Rate increases with steeper slopes and greater slope length.
905 • Soil erodibility is the average long-term erosive tendency of rainfall and runoff. This, in
906 turn, depends on the texture, proportion of organic matter, soil structure, and
907 permeability. Erosion rate increases with greater erodibility.
908 • Rainfall erosivity represents the erosive force caused by rainfall and runoff. This, in turn,
909 is dependent on the annual precipitation. Greater rainfall erosivity causes a higher rate of
910 soil erosion.
911 • Average land cover and management factor is a measure of land use and is calculated as
912 the average soil-loss ratio weighted by the distribution of annual rainfall.
913 • Average conservation practice factor is a measure of practices that control erosion. For
914 RUSLE, P is assigned a value of 0.5 for agricultural land and 0.8 for mixed agricultural
915 and forest land. Erosion rate decreases with active conservation practices.
916
917 Groundwater Reliance (#125)
918 Groundwater Reliance is a measure of the dependence of a community on available groundwater
919 resources. It is defined as the share of total annual withdrawals from groundwater and calculated
920 as the ratio of withdrawals from groundwater to total annual withdrawals from groundwater and
921 surface water (Kurd et al., 1998).
922
923 This indicator is particularly important as a measure of vulnerability in those regions that depend
924 primarily on groundwater for drinking water, irrigation, and industrial and commercial purposes,
925 because surface water supplies may be limited, contaminated, or expensive to use (Kurd et al.,
926 1998). Long-term changes in the hydrologic cycle, specifically groundwater recharge and surface
927 flows, may make regions with higher groundwater reliance more vulnerable to water shortages.
928 In contrast, regions that today depend primarily on surface water sources, and therefore have not
929 yet had to tap their groundwater reserves, may be less vulnerable in the long-term to scarcity of
930 surface water caused by climate change as they may have available groundwater to meet their
931 water demand (Kurd et al., 1998). The data used to map this indicator were collected every five
932 years.
933
Do Not Cite or Quote Page 35
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
934 Herbicide Concentrations in Streams (#367) and Insecticide Concentrations in Streams (#369)
935 Pesticides are of acknowledged concern for human health as well as the health of aquatic
936 organisms. Their ingestion may lead to a number of health concerns, including kidney problems,
937 reproductive problems, and cancer. These compounds have been studied primarily in laboratory
938 animals, although some information is based on epidemiological data. Pesticides are a primary
939 drinking water quality indicator, with Maximum Contaminant Levels (MCLs) in place for 24
940 pesticides, mostly in the |ig/L range. The data used to map this indicator were collected at
941 various frequencies depending on purpose and collection site.
942
943 Environmental changes that may affect the concentrations of pesticides in streams include
944 alterations to the hydrologic cycle (Noyes et al., 2009). Lower precipitation in the summer may
945 lower streamflow and reduce dilution, leading to higher concentrations, although higher
946 temperatures may offset this by increasing pesticide degradation (Bloomfield et al, 2006). If
947 winter precipitation increases, dilution will tend to increase as well. Climate change may also
948 alter how water moves over the land. For example, increased precipitation, or more extreme wet
949 events, may increase overland flow because the capacity of the soil to infiltrate water will be
950 exceeded. Intense summer storms may promote increased runoff if the antecedent conditions are
951 dry because the soil will be more hydrophobic (Boxall et al., 2009). These effects may promote a
952 greater input of suspended solids into streams, increasing the loading of particle associated
953 pesticides. Climate-induced changes to pest migration or ranges may prompt changes in pesticide
954 usage, which may be reflected in inputs to surface water (Chen and McCarl, 2001). Bloomfield
955 et al. (2006) note, however, that direct climate change effects would be difficult to predict, and
956 that secondary effects from land use changes associated with climate change may be more
957 important as controls on inputs of pesticides to surface water.
958
959 Herbicides in Groundwater (#373) and Insecticides in Groundwater (#374)
960 Because groundwater can contribute herbicides and pesticides to streams, concentrations of these
961 compounds in groundwater need to be considered in evaluations of surface waters and aquatic
962 ecosystems. The presence of these toxics provides an indication of potential contributions of
963 these chemicals to streams. As described in the previous entry, they are also a primary drinking
964 water concern, and EPA has set MCLs for 24 of these compounds. The data used to map this
965 indicator were collected at various frequencies depending on purpose and collection site.
966
967 Changes in precipitation brought on by global climate change may affect groundwater herbicide
968 and insecticide concentrations. Greater winter precipitation would promote the movement of
969 these substances through the soil towards the water table, and large storms in particular may
970 rapidly transport them into groundwater. In addition, during drier summers, less biodegradation
971 occurs in the unsaturated zone, leaving greater amounts of pesticides available to be transported
972 to groundwater. Finally, herbicide and insecticide use may increase if climate change leads to
973 increased prevalence of pests and weeds.
974
975 Instream Use/Total Streamflow (#351)
976 A primary consideration for healthy aquatic ecosystems is having adequate water to maintain
977 fish and wildlife habitat, and competing demands for water can be a significant stressor to these
978 ecosystems (Meyer et al., 1999). This indicator describes the competition by expressing instream
979 water needs for fish and wildlife as a percentage of total available streamflow. The ratio of
Do Not Cite or Quote Page 36
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
980 instream use to total streamflow can be calculated using three variables: total groundwater
981 withdrawals, mean annual runoff, and groundwater recharge. The data for these variables were
982 collected at various frequencies: data on groundwater withdrawals were collected every 5 years,
983 data on mean annual runoff were collected as a one-time effort in 1975, and groundwater
984 recharge data were collected as a one-time effort between 1951 and 1980.
985
986 Changes in water withdrawals due to population change can decrease the streamflow available
987 for instream use. Alterations in the hydrologic cycle due to climate change might also decrease
988 streamflow in some areas. This would cause the instream use/total streamflow ratio to increase.
989 A WRC (1978) report notes that a ratio > 100 (based on 1975 data) indicates that withdrawals of
990 water are having a deleterious effect on the instream environment. DeWalle et al. (2000),
991 however, discuss the scenario of concurrent urbanization and climate change. They note that
992 urbanization can significantly increase mean annual streamflow and may offset reductions in
993 flow caused by climate change. This indicator serves as a good vulnerability indicator because
994 regions with greater competition between instream flow uses and consumptive uses are more
995 vulnerable to decreases in streamflow resulting from climate change.
996 Macroinvertebrate Index ofBiotic Condition (#460)
997 The Macroinvertebrate Index ofBiotic Condition indicator (#460) is a composite measure of the
998 condition of macroinvertebrates in streams. Assessing the condition is these macroinvertebrate
999 species is a good measure of the overall condition of the aquatic ecosystem as they often serve as
1000 the basic food for aquatic vertebrates and are, therefore, essential to aquatic ecosystems with
1001 vertebrate species (USEPA, 2004; USEPA, 2006; USEPA, 201 Of). Furthermore, the structure
1002 and function of macroinvertebrate assemblages is a reflection of their exposure to various
1003 stressors over time, as these organisms have long life-cycles over which they change in response
1004 to stress (USEPA, 2004). Stable ecosystems are likely to contain a variety of species, some of
1005 which are sensitive to environmental conditions. These sensitive taxa are most likely to be
1006 subject to local extirpations when exposed to climate-induced changes in temperature or flow
1007 conditions. Similarly, these species may not tolerate increases in precipitation or temperature
1008 variation, which subsequently increase the frequency of disturbance events.
1009
1010 This indicator allows qualitative measurements of macroinvertebrate condition to be represented
1011 as a numerical value. It can be considered a good indicator of relative vulnerability as it
1012 compares macroinvertebrate condition at study sites with those at undisturbed reference sites
1013 located in similar ecoregions (USEPA, 2006). Furthermore, this indicator may be tracked over
1014 time to determine temporal changes in vulnerability relative to a baseline (USEPA, 201 Ob).
1015
1016 The Macroinvertebrate Index indicator is represented by the average Macroinvertebrate Index
1017 value in a given area. It depends on field observations of six variables: taxonomic richness,
1018 taxonomic composition, taxonomic diversity, feeding groups, habits, and pollution tolerance
1019 (USEPA, 2006). Each variable is assessed using the benthic macroinvertebrate protocol in which
1020 stream samples are collected and the characteristics of macroinvertebrates in them are assessed
1021 (USEPA, 2004). Each variable is assigned a score based on field observations and individual
1022 scores are summed to obtained the value of the Macroinvertebrate Index, ranging from 0 to 100
1023 (USEPA, 2006). The data used to map this indicator were collected every five years.
1024
Do Not Cite or Quote Page 3 7
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
1025 The Macroinvertebrate Index changes based on the following variables:
1026 • Taxonomic richness, which is the number of distinct taxa or groups of organisms. A
1027 stream with more taxa, which indicates a wider variety of habitats and food requirements,
1028 will be less vulnerable to stress.
1029 • Taxonomic composition, which is a measure of the relative abundance of ecologically
1030 important organisms to those from other taxonomic groups. For example, a polluted
1031 stream will likely have a higher abundance of organisms that are resilient to pollution
1032 with lower representation from other taxa and will be more vulnerable to stress.
1033 • Taxonomic diversity, which is a measure of the distribution of organisms in a stream
1034 amongst various taxonomic groups. Higher taxonomic diversity represents a healthier
1035 stream that is less vulnerable to stress.
1036 • Feeding groups, which is a measure of the diversity of food sources that
1037 macroinvertebrates depend on. A more diverse food chain is representative of a more
1038 stable aquatic environment that is less vulnerable to stress.
1039 • Habits, which is measure of the characteristics of different organisms and their
1040 preferences for different habitats. A stream environment with more diverse habitats (e.g.,
1041 streambed sediment, rocks, woody tree roots, debris) supports a wider variety of
1042 macroinvertebrates and will be less vulnerable to stress.
1043 • Pollution tolerance, which is a measure of the degree of resilience to pollution of
1044 macroinvertebrate species in a stream. Highly sensitive organisms will be more
1045 vulnerable to contamination in streams, compared to pollution-resistant ones.
1046
1047 Macroinvertebrate Observed/Expected (O/E) Ratio of Taxa Loss (#461)
1048 Stable ecosystems are likely to contain a variety of species, some of which are sensitive to
1049 environmental conditions. These sensitive taxa are most likely to be subject to local extirpations
1050 when exposed to climate-induced changes in temperature or flow conditions. Similarly, these
1051 species may not tolerate increases in precipitation or temperature variation, which subsequently
1052 increase the frequency of disturbance events. A measure of the loss of sensitive species may thus
1053 serve as an important indicator of vulnerability to climate change and other stressors.
1054
1055 The Macroinvertebrate Observed/Expected (O/E) Ratio of Taxa Loss (#461) indicator is a
1056 measure of the biodiversity loss in a stream (USEPA, 2006). This indicator (also known as O/E
1057 Taxa Loss) is represented by the ratio of the taxa observed at a site to the ratio of the taxa
1058 expected to be present at that site as predicted by a region-specific model (EPA, 2006). Observed
1059 taxa are assessed using the benthic macroinvertebrate protocol in which stream samples are
1060 collected and the characteristics of macroinvertebrates present in them are assessed (USEPA,
1061 2004). Expected taxa are predicted by models developed from data collected at undisturbed or
1062 least disturbed reference sites within a region, for each of three major U.S. regions - Eastern
1063 Highlands, Plains and Lowlands, and the West (USEPA, 2006). O/E Taxa Loss ratios are
1064 represented as a percentage of the expected taxa present, and they range from 0% (i.e., none of
1065 the expected taxa are present) to greater than 100% (i.e., more taxa than expected are present)
1066 (USEPA, 2006). The data used to map this indicator were collected every five years. The O/E
1067 Taxa Loss directly reflects the vulnerability of an ecosystem based on its loss of biodiversity
1068 (USEPA, 2006). It also reflects a change in ecological condition relative to undisturbed reference
1069 sites (USEPA, 2006).
1070
Do Not Cite or Quote Page 38
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
1071 Meteorological Drought Indices (#165)
1072 Meteorological Drought Indices provide a representation of the intensity of drought episodes
1073 brought on by a lack of precipitation (Heim, 2002). For example, the Palmer Drought Severity
1074 Index (PDSI) takes into account precipitation and soil moisture data from a water balance model
1075 as well as a comparison of meteorological and hydrological drought (Heim, 2002). The PDSI can
1076 be used as a proxy for surface moisture conditions and streamflow (Dai et al., 2004). The data
1077 used to map this indicator were collected monthly. PDSI trends are also linked to climate
1078 patterns such as the El Nino-Southern Oscillation (Dai et al., 1998). Because drought is a well
1079 recognized stressor for natural and human systems, indicators of the spatial and temporal
1080 distribution of drought severity are relevant to vulnerability to additional external stressors. This
1081 is particularly true for climate change, as drought is directly linked to changes in meteorology
1082 that themselves are likely to be affected by climate change.
1083
1084 Organochlorines in Bed Sediment (#371)
1085 As part of its National Water Quality Assessment (NAWQA) program, the U.S. Geological
1086 Survey has analyzed organochlorines in bed sediment (USGS, 1999). Although they have not
1087 been used for decades, organochlorine insecticides linger in sediments, posing a potential threat
1088 to humans and aquatic organisms. For example, any increase of organochlorines in shellfish may
1089 find its way into the human food chain. As a vulnerability indicator, organochlorines in sediment
1090 are deleterious compounds that can cause ecological condition to deviate from what would be
1091 expected in an undisturbed system. The data used to map this indicator were collected at various
1092 frequencies depending on purpose and collection site.
1093
1094 Any environmental factor that disturbs bed sediment or affects its transport may affect the
1095 exposure of humans or aquatic organisms to organochlorines. Dredging of rivers and harbors
1096 may resuspend sediments, increasing contact with aquatic organisms. More intense storms may
1097 also resuspend sediment. On the other hand, climate-related increase of sediment input to larger
1098 water bodies may provide some "burial" of contaminated sediments, especially if the new
1099 sediment is uncontaminated.
1100
1101 Pesticide Toxicity Index (#364)
1102 This indicator combines pesticide concentrations for a stream water sample with toxicity
1103 estimates to produce a number (the Pesticide Toxicity Index or PTI value) that indicates the
1104 sample's relative toxicity to aquatic life. This method, developed by Munn and Gilliom (2001),
1105 allows data for multiple pesticides to be linked to the health of an aquatic ecosystem, and it
1106 allows streams to be rank ordered by their PTI values (Gilliom et al., 2006). It is a suitable
1107 vulnerability indicator in that it attempts to estimate the potential damage to an ecosystem's
1108 resilience as a result of pesticides. The data used to map this indicator were collected at various
1109 frequencies depending on purpose and collection site.
1110
1111 The PTI value for a stream increases as pesticide concentrations increase. Concentrations may
1112 change due to environmental factors such as urbanization, whereby increased streamflow may
1113 decrease concentrations due to greater dilution or produce greater pesticide inputs through
1114 increased sediment input. Potential climate-related effects include decreased streamflow, which
1115 may increase concentrations through reduced dilution, or increased precipitation, leading to
1116 increased streamflow and hence sediment inputs. Conversely, increased temperature may
Do Not Cite or Quote Page 39
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
1117 accelerate pesticide degradation, leading to lower concentrations. However Noyes et al. (2009)
1118 note that if water temperature increases, pesticides can become more toxic to aquatic organisms.
1119 It is not known if this effect would apply to humans. Determining the toxicity of mixtures of
1120 pesticides to humans is extremely challenging; exploring toxicity changes as a result of climate
1121 change is an important direction for future research.
1122
1123 Precipitation Elasticity of Stream flow (#437)
1124 The Precipitation Elasticity of Streamflow indicator is designed to assess the sensitivity of
1125 streamflow to changes in precipitation patterns. It measures the sensitivity of Streamflow to
1126 climate change and is useful in assessing the vulnerability of regions where maintaining
1127 relatively constant streamflow is critical (Sankarasubramanian et al., 2001).
1128
1129 The Precipitation Elasticity of Streamflow (Ep) is defined as a change in streamflow caused by a
1130 proportional change in precipitation. It can be calculated as follows:
1131
1132 EP(P. Q)= dQ P
1133 dP Q
1134
1135 where P = precipitation and Q = streamflow
1136
1137 An indicator value greater than 1 indicates that a large change in precipitation is accompanied by
1138 a relatively smaller change in streamflow, and thus, streamflow is elastic or less sensitive to
1139 precipitation changes. An indicator value of less than 1 indicates that a small change in the
1140 precipitation is accompanied by a relatively larger change in the streamflow, and thus streamflow
1141 is inelastic or more sensitive to precipitation changes. The data for these variables were collected
1142 at various frequencies: data on streamflow were collected annually, and data on precipitation
1143 were collected monthly.
1144
1145 Streams do not respond uniformly to increased precipitation due to underlying differences in
1146 geology, terrain, and other factors. Precipitation elasticity can be used to predict how increased
1147 precipitation brought on by global climate change might affect streams in a given region.
1148 Increases in precipitation and storm intensity could result in disproportionately large adverse
1149 effects, such as flooding, in areas with high precipitation elasticity. Climate change, as well as
1150 anticipated increased urbanization, both contribute to the expected increase in the intensity of
1151 storms in some areas, leading to more flooding and severe erosion in flashier stream systems.
1152
1153 Ratio of Reservoir Storage to Mean Annual Runoff (#449)
1154 The Ratio of Reservoir Storage to Mean Annual Runoff indicator is a measure of the storage
1155 capacity of reservoirs relative to runoff within the basin (Graf, 1999). Dams can be used to
1156 manage water resources to ensure reliable supply of water to regions that depend on surface
1157 water (Lettenmaier et al., 2008). On the other hand, dams can also alter riparian ecosystems and
1158 hydrologic processes, causing unnatural variability in streamflow when water released,
1159 fragmenting aquatic ecosystems, and leading to erosion and sedimentation (Graf, 1999). The
1160 ability to store a large portion of water from land runoff indicates that a community already has
1161 the capacity to harness more surface water, if needed, and may, therefore, be less vulnerable to
1162 changes in hydrologic processes. Arid or semi-arid regions, where water is scarce, tend to have
Do Not Cite or Quote Page 40
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
1163 larger reservoirs, some of which may be able store up to three or four times the volume of annual
1164 runoff (Graf, 1999). Climate change may introduce increased inter- and intra- annual variation in
1165 runoff. Areas with relatively low reservoir storage compared to the availability of runoff may be
1166 more vulnerable to intense and prolonged droughts or changes in the seasonal timing of runoff.
1167
1168 The Ratio of Reservoir Storage to Mean Annual Runoff is determined by the magnitude of its
1169 individual components. The storage capacity of reservoirs in a given region is determined by the
1170 size of the dam, and the mean annual runoff is determined largely by precipitation and snowmelt.
1171 The data used to map this indicator include runoff data that were collected as a one-time effort
1172 between 1951 and 1980, and dam inventory data for which the collection frequency is unknown.
1173 This indicator is a good indicator of the vulnerability of water supply; however, it may have a
1174 limited ability to predict the vulnerability of water quality and aquatic ecosystems as dams tend
1175 to adversely affect both these variables, while they benefit water supply or availability.
1176
1177 Ratio of Snow to Precipitation (#218)
1178 The Ratio of Snow to Precipitation is the ratio of the amount of snowfall to the amount of total
1179 precipitation. It can also be described as the percentage of precipitation falling as snow. As such,
1180 a decreasing ratio can indicate either a relative decrease in snowfall or relative increase in
1181 rainfall, although annual trends in the Ratio of Snow to Precipitation primarily reflect the former
1182 (Huntington et al., 2004). The data used to map this indicator were collected annually. Changes
1183 in the Ratio of Snow to Precipitation are driven by temperature variations (Karl et al., 1993).
1184 Thus, the ratio will be affected by temperature changes associated with global climate change.
1185 Trends in the Ratio of Snow to Precipitation can lead to changes in runoff and streamflow
1186 patterns, because of the effect on the timing and amount of spring snowmelt (Huntington et al.,
1187 2004; Knowles et al., 2006). Because of this, areas with decreasing ratios can be more vulnerable
1188 to summer droughts (Feng and Hu, 2007).
1189
1190 Ratio of Water Withdrawals to Annual Streamflow (#219)
1191 The Ratio of Water Withdrawals to Annual Streamflow indicator is a measure of a region's
1192 water demand relative to the potential of the watershed to supply water. This indicator is defined
1193 as the share of total annual water withdrawals (from surface water and groundwater) to the
1194 unregulated mean annual streamflow (Kurd et al., 1998). The ratio of water withdrawals to
1195 annual streamflow can be calculated using three variables: mean annual precipitation, mean daily
1196 maximum temperature, and water-use data. The data for these variables were collected at various
1197 frequencies: mean annual precipitation data were collected monthly, mean daily maximum
1198 temperature data were collected monthly, and water-use data were collected every five years.
1199
1200 Streamflow is important for the sustenance of surface water supply as well as for riparian
1201 ecosystems. It is also important for aquifers that are fed by streamflow. Regions with higher
1202 water demand will withdraw higher amounts of water from streamflow both for immediate use as
1203 well as for storage in reservoirs. These regions also rely on institutional management to maintain
1204 the critical flow in rivers and streams (Kurd et al., 1998). In the long-term, such regions are
1205 likely to be more vulnerable to climate changes which lead to large changes in streamflow,
1206 whereas regions where water demand is a smaller proportion of the unregulated streamflow are
1207 likely to be less vulnerable to climate-induced changes in streamflow, as there is greater
1208 available supply to draw from without affecting the critical flow (Kurd et al., 1998).
Do Not Cite or Quote Page 41
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
1209 Runoff Variability (#453)
1210 Runoff Variability is defined as the coefficient of variation of annual runoff. This indicator
1211 largely reflects the variation of annual precipitation (Lettenmaier et al., 2008; Maurer et al.,
1212 2004). Small or moderate changes in precipitation can lead to larger changes in runoff amounts,
1213 increasing runoff variability (Burlando and Rosso, 2002; Karl and Riebsame, 1989). Runoff is
1214 also linked to and affected by other factors, such as temperature, evapotranspiration, snowmelt,
1215 and soil moisture, and is a critical component of the annual water-balance (Maurer et al., 2004;
1216 Gedney et al., 2006; Karl and Riebsame, 1989; Wolock and McCabe, 1999).
1217
1218 Understanding inter-annual variation in runoff is important for future scenarios in which climate
1219 change will affect both precipitation and temperature, both of which affect runoff (Maurer et al.,
1220 2004). The spatial and temporal variability of runoff is also essential for predicting droughts and
1221 floods (Maurer et al., 2004). The data used to map this indicator were collected every three
1222 hours. Moreover, it is easier to measure runoff than it is to measure other variables in the water-
1223 balance, such as precipitation and evapotranspiration, thus making it a more reliable indicator
1224 (Wolock and McCabe, 1999).
1225
1226 Stream Habitat Quality (#284)
1227 The Stream Habitat Quality (#284) indicator is used to assess the condition in and around
1228 streams. Physical features such as in-stream vegetation, sediment, and bank vegetation create
1229 diverse riparian habitats that can support many plant and animal species (Heinz Center, 2008).
1230 Streams degraded by human use, characterized by decreased streambed stability, increased
1231 erosion of stream banks, loss of in-stream vegetation, are marginal habitats for most species
1232 (Heinz Center, 2008), and hence may be particularly vulnerable to additional stresses. Stream
1233 habitat can be altered quickly due to stochastic events such as major flooding, or slowly over
1234 time due to subtle changes in flow regime. Climate-induced changes in storm intensity, runoff
1235 seasonality, average flows, or flow variation could result in disproportionately large negative
1236 effects on high quality stream habitats.
1237
1238 The Stream Habitat Quality indicator is represented by the Rapid Bioassessment Protocol score,
1239 an index that can be used to assess the condition of underwater and bank habitats. The Rapid
1240 Bioassessment Protocol is a methodology developed by EPA to assess habitat conditions based
1241 on field observations often variables: epifaunal substrate/ available cover, embeddedness (for
1242 riffles) or pool substrate characterization (for pools), velocity and depth regimes (for riffles) or
1243 pool variability (for pools), sediment deposition, channel flow status, channel alteration,
1244 frequency of riffles or bends (for riffles) or channel sinuosity (for pools), bank stability, bank
1245 vegetative protection, and riparian vegetated zone width (USEPA, 2004). Each of these variables
1246 is observed and assigned a qualitative category and score: Poor (0-5), Marginal (6-10), Sub-
1247 optimal (11-15), or Optimal (16-20) (USEPA, 2004). The scores for all the parameters are
1248 summed to obtain the Rapid Bioassessment Protocol score for that stream (USEPA, 2004). A
1249 higher Rapid Bioassessment Protocol score indicates higher Stream Habitat Quality, while a
1250 lower Rapid Bioassessment Protocol score indicates a degraded stream.
1251
1252 Stream Habitat Quality changes based on changes in the following variables (USEPA, 2004):
1253 • Epifaunal substrate or available cover, which measures the relative quantity and variety
1254 of natural structures in the stream, such as cobble (riffles), large rocks, fallen trees, logs
Do Not Cite or Quote Page 42
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
1255 and branches, and undercut banks, available as refugia, feeding, or sites for spawning and
1256 nursery functions of aquatic macrofauna. The abundance of these structures in the stream
1257 creates niches for animals and insects, and allows for a diversity of species to thrive in
1258 the same habitat.
1259 • Embeddedness in riffles, which measures the extent to which rocks (gravel, cobble, and
1260 boulders) and snags are buried in the silt or sand at the bottom of the stream. Fewer
1261 embedded features increase the surface area available to macroinvertebrates and fish for
1262 shelter, spawning, and egg incubation. Similarly, pool substrate characterization is a
1263 measure of the type and condition of bottom sediment in pools. Firmer sediment, such as
1264 gravel, and rooted aquatic vegetation support more organisms.
1265 • Velocity and depth regimes for riffles measure the variety of habitats caused by different
1266 rates of flow and stream depth, such as slow-deep, slow-shallow, fast-deep, and fast-
1267 shallow. The ideal stream habitat will exhibit four patterns which represent the stream's
1268 ability to maintain a stable environment. Pool variability is a measure of the different
1269 pool types, such as large-shallow, large-deep, small-shallow, and small-deep. The more
1270 diverse the pool types, the greater the diversity of the habitat that can be supported by the
1271 stream.
1272 • Sediment deposition is a measure of the amount of sediment accumulation in streams.
1273 More sediment deposition is indicative of unstable streambeds which are an unfavorable
1274 environment for aquatic organisms.
1275 • Channel flow status is the extent to which the stream channel is filled with water. Low
1276 channel flow may not cover the streambed and vegetation leaving them exposed, thereby
1277 reducing available habitat for organisms. Optimal channel flow covers the streambed
1278 creating more available habitat for organisms to thrive in.
1279 • Channel alteration is a measure of the significant changes, typically human-induced, in
1280 the shape of the stream channel, such as straightening, deepening, diversions, or
1281 conversion to concrete. Altered channels are often degraded and limit the natural habitat
1282 available to organisms.
1283 • Frequency of riffles is a measure of the number of riffles in a stream. Riffles provide
1284 diverse habitats in which many organisms can thrive. Similarly, channel sinuosity in
1285 pools is a measure of the degree to which the stream meanders. More sinuous streams
1286 allow for diverse natural habitats and can also adapt to fluctuations in water volumes,
1287 thereby providing a more stable environment for aquatic organisms.
1288 • Bank condition is a measure of the extent to which banks are eroded. Eroded banks
1289 indicate moving sediments and unstable stream habitat for aquatic animals and plants.
1290 • Bank vegetative protection refers is a measure of the vegetative cover of the stream bank
1291 and near stream areas. Banks with dense plant growth prevent erosion, control nutrients
1292 in the stream, and provide shade, thus maintaining a healthier riparian ecosystem. In
1293 contrast, banks that are covered with concrete in urban areas or experience high grazing
1294 pressure from livestock in agricultural areas prevent vegetative growth along the stream,
1295 thereby creating a poorer aquatic environment.
1296 • Riparian vegetated zone width is a measure of the extent of the vegetative zone from the
1297 edge of the stream bank through to the outer edge of the riparian zone. The riparian
1298 vegetated zone buffers the riparian environment from surrounding areas, minimizes
1299 runoff, controls erosion, and shades the riparian habitat.
1300
Do Not Cite or Quote Page 43
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
1301 The Stream Habitat Quality indicator allows qualitative measurements of habitat condition to be
1302 represented as a numerical value. However, most measurements of independent variables that
1303 affect the score are "visual-based", that is they are dependent on the visual assessment of the
1304 field team that will score the study sites for each variable (USEPA, 2004). Despite this, Stream
1305 Habitat Quality can be considered a good indicator of relative vulnerability for our purposes as it
1306 compares stream conditions at study sites with those at undisturbed reference sites located in
1307 similar regions (USEPA, 2006; Heinz Center, 2008). Furthermore, this indicator may be tracked
1308 over time to determine temporal changes in relative vulnerability, thus allowing one to assess the
1309 impacts of future stressors in relation to present ones. The data used to map this indicator were
1310 collected every five years.
1311
1312 Total Use/Total Streamflow (#352)
1313 This is the second indicator expressing the competition between water needs and water
1314 availability in streamflow. According to WRC (1978), the ratio of total use to total Streamflow is
1315 a measure of the water available for "conflict-free development of offstream uses." It is similar
1316 to Indicator #351 (Instream Use/Total Streamflow), except that the numerator includes the needs
1317 for both instream and offstream use. The ratio of total use to total stremflow can be calculated
1318 using three variables: mean annual runoff, groundwater recharge, and water use. The data for
1319 these variables were collected at various frequencies: mean annual runoff data were collected as
1320 a one-time effort from 1951-1980, groundwater recharge data were collected as a one-time effort
1321 in 1975, and water-use data were collected every five years. It is a good vulnerability indicator
1322 because regions that have high offstream needs may be less able to withstand decreases in
1323 streamflow that may occur due to climate change.
1324
1325 Meyer et al. (1999) note that climate-induced changes in water availability will occur in a
1326 context in which human-induced changes in water demand are also occurring. A reduction in
1327 streamflow (e.g., due to changes in climate) or an increase in offstream use (due to greater
1328 withdrawals for consumptive use) will increase this ratio. According to WRC (1978), a ratio >
1329 100% indicates a conflict between offstream uses and instream flow needs. As with instream
1330 use/total streamflow, total streamflow may be increased by urbanization. This is presumably due
1331 to increased impervious area. This may offset any flow reductions due to climate change in areas
1332 undergoing population expansion.
1333
1334 Wetland and Freshwater Species at Risk (#326)
1335 The Wetland and Freshwater Species at Risk is a measure of the level of stress that a watershed
1336 is experiencing based on the number of water-dependent species "at risk" (Kurd et al., 1998).
1337 Watersheds may be stressed due to changes in the hydrological cycle related to global climate
1338 change and encroachment or other disturbances from human activities (Kurd et al., 1998). This
1339 may cause populations dependent on affected niches to diminish, and may even lead to
1340 extinction of species in some cases (Kurd et al., 1998).
1341
1342 The Wetland and Freshwater Species at Risk indicator is defined as the number of aquatic and
1343 wetland species that are classified as vulnerable, imperiled, or critically imperiled by
1344 NatureServe, a non-profit conservation organization that maintains biological inventories for
1345 animal and plant species in the U.S. A watershed with a higher value of this indicator might be
1346 considered to be more vulnerable than a watershed with the lower value of this indicator.
Do Not Cite or Quote Page 44
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
1347
1348 Assessing the condition of species in watershed can be a good indication of the health of the
1349 watershed. However, indicator is not necessarily a very strong indicator of the vulnerability of
1350 aquatic ecosystems, as it only looks at the absolute number of at-risk species, regardless of the
1351 total number of species that occupy that habitat (Kurd et al., 1998). Furthermore, this indicator
1352 does not account for the inherent diversity in the watershed; watersheds with historically more
1353 species may be less vulnerable to species loss (Kurd et al., 1998).
1354
1355 Water Availability: Net Streamflow per Capita (#623)
1356 Water availability is a measure of the availability of freshwater resources per capita to meet
1357 water demand for various human consumptive uses (Kurd et al., 1998). It is defined as the net
1358 streamflow per capita and can be calculated as follows:
1359
1360 Water Availability = (Unregulated annual streamflow - Annual water withdrawals)
1361 Population
1362
1363 This indicator depends on three variables: mean annual runoff, groundwater recharge, and water
1364 use. The data for these variables were collected at various frequencies: mean annual precipitation
1365 data were collected monthly, mean daily maximum temperature data were collected monthly,
1366 and water-use data were collected every five years. We might reasonably assume that regions
1367 with abundant per capita water availability are less vulnerable to long-term changes in the
1368 hydrologic cycle brought on by climate change as well as to population growth, and, conversely,
1369 regions with lower per capita water availability are more vulnerable.
1370
1371 V. Challenges Part II: Determining Relative Vulnerability
1372 A,
1373 A variety of approaches are available to water quality and natural resource managers who must
1374 interpret indicator values and indicator-based vulnerability assessments. These approaches vary
1375 depending on the state of available knowledge for a given indicator. In many cases, research
1376 suggests that responses of water quality or ecosystem condition to external stressors are linear,
1377 meaning that changes in condition (or in indicators of condition) occur over a gradual gradient
1378 rather than abruptly. Thus, management decisions can be made based on the value of the
1379 indicator along the gradient. In other cases, the response may be non-linear, but the thresholds
1380 that distinguish acceptable from unacceptable conditions are not yet fully understood. Given this
1381 state of knowledge, management decisions to prevent ecosystem degradation or a risk to human
1382 health may be based on the relative value of an indicator along the gradient of known values. For
1383 example, managers may act out of an abundance of caution when the value of an indicator
1384 increases following a long period of stability, even if the risks associated with inaction are
1385 unclear. Managers may also choose to act if an indicator value appears to be significantly
1386 different from values in other, more pristine locations.
1387
1388 Another approach is the use of known thresholds to facilitate indicator interpretation by
1389 indicating points at which management action is required to prevent adverse impacts to human
1390 health and the environment (Kurtz et al. 2001). Vulnerability thresholds reflect abrupt or large
Do Not Cite or Quote Page 45
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
1391 changes in the vulnerability of water quality or aquatic ecosystems. EPA's Office of Research
1392 and Development (ORD) Evaluation Guidelines, which describes key concepts in environmental
1393 indicator development, describes the role that thresholds can play in interpreting the values of
1394 indicators of ecological condition:
1395
1396 "To facilitate interpretation of indicator results by the user community, threshold values
1397 or ranges of values should be proposed that delineate acceptable from unacceptable
1398 ecological condition. Justification can be based on documented thresholds, regulatory
1399 criteria, historical records, experimental studies, or observed responses at reference sites
1400 along a condition gradient. Thresholds may also include safety margins or risk
1401 considerations." (USEPA, 2000a)
1402
1403 In this study, we attempted to divide the range of values calculated for appropriate indicators into
1404 different classes based on evidence in the literature of abrupt or large changes in vulnerability
1405 associated with certain values of the indicator. These functional break points (i.e., objective
1406 thresholds that distinguish between acceptable and unacceptable conditions) can be highly useful
1407 to decision makers. The literature reviewed for this study, however, most often presented
1408 arbitrary cutoffs based on round numbers or frequency distributions. It is not surprising that
1409 functional break points do not currently exist for many indicators. Groffman et al. (2006) point
1410 out that determining such break points can be challenging due to the non-linear response of many
1411 indicators and the multiple factors that can affect the value of functionally relevant indicator
1412 break points. For example, natural variation in water chemistry and ecosystem types across the
1413 nation leads to spatial variation in critical thresholds for dissolved oxygen (DO). Persistently low
1414 DO levels in any one ecosystem can yield a community of flora and fauna that are unaffected by
1415 DO levels that would be detrimental to another ecosystem.
1416
1417 In some cases, objective break points in non-linear system responses may be characterized
1418 through additional research, either through meta-analysis of previous research efforts or through
1419 new data collection and analysis. In either case, collection of indicator values associated with a
1420 range of ecological responses is required to establish functionally relevant break points. There
1421 are several statistical approaches for identifying thresholds in non-linear relationships, including
1422 regression tree analysis (Breiman et al., 1984) and two-dimensional Kolmogorov-Smirnov
1423 techniques (Garvey et al., 1998). Future research may yield additional insights into how these
1424 break points vary spatially (Link, 2005).
1425
1426 In general, we considered three different types of thresholds for the suite of indicators evaluated
1427 in this project.
1428
1429 Human health-based thresholds, such as Maximum Contaminant Level Goals (MCLGs) or
1430 Health Advisories (HAs), which are set based on scientific studies can potentially be used as
1431 thresholds for water quality indicators. EPA establishes MCLGs for contaminants detected in
1432 drinking water based on an extensive review of available data on the health effects of these
1433 contaminants.
1434
1435 The MCLG is the maximum concentration of a contaminant in drinking water which has no
1436 known or anticipated adverse health effect on the population consuming this water, (USEPA,
Do Not Cite or Quote Page 46
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
1437 2010g; USEPA, 2009b). MCLGs for carcinogens are set to zero, based on any evidence of
1438 carcinogenicity, as these effects typically manifest over a lifetime of exposure. MCLGs for non-
1439 carcinogens are often based on a Reference Dose (RfD), which is the amount of contaminant that
1440 a person can be exposed to daily without experiencing adverse health effects over a lifetime
1441 (expressed in units of mg of substance/kg body weight/day). MCLGs are non-enforceable and
1442 are based purely on the risk posed by a contaminant to human health (USEPA, 2010c; USEPA,
1443 2009a). The MCLG is, thus, a threshold based on scientific data (as opposed to a Maximum
1444 Contaminant Level [MCL] that takes other factors into account ).
1445
1446 Similarly, HAs are estimates of acceptable concentrations of drinking water contaminants that
1447 are developed by EPA as guidelines to help Federal, State, and local entities better protect their
1448 drinking water quality (USEPA, 2009a). Like MCLGs, HAs are not enforceable, but are
1449 determined solely based on health effects data, such as exposure and toxicity. Unlike MCLGs,
1450 HAs are revised from year to year as new data become available.
1451
1452 Other parameters could also be used to assess the toxicity of a drinking water contaminant
1453 (USEPA, 2009c):
1454 • Median Lethal Dose (LD50), which is the oral dose of a contaminant that will cause 50
1455 percent of the population it is administered to die (expressed in mg per kg of body weight);
1456 • Cancer Potency (for carcinogens), which is the concentration of a contaminant in drinking
1457 water that poses a risk of cancer equivalent to 1 in 10,000 individuals or 10"4;
1458 • No Observed Adverse Effect Level (NOAEL), which is the highest dose at which no adverse
1459 health effects are observed; and
1460 • Lowest Observed Adverse Effect (LOAEL) associated with the RfD, which is the lowest
1461 dose at which adverse health effects are observed.
1462 These parameters are considered preliminary or less developed thresholds than an RfD value but
1463 could still, potentially, be used as thresholds for drinking water indicators.
1464
1465 Ecological thresholds are central to the ecological theory of "alternate stable states" (Lewontin,
1466 1969; Rolling, 1973; Sutherland, 1974; May, 1977; Scheffer et al., 2001), where the biotic and
1467 abiotic conditions within an ecosystem can reach multiple equilibria. It is believed that the
1468 transition between stable states occurs when a significant perturbation results in the breaching of
1469 one or more ecological thresholds. The "ball-in-cup" model is commonly used to illustrate this
1470 concept (Beisner et al., 2003). A stable ecosystem can be thought of as a ball that resides at the
1471 bottom of a cup. There may be many adjoining cups (i.e., the alternate stable states) that the ball
1472 could reside in. Small perturbations may push the ball up the side of the current cup, but the ball
1473 will eventually return to the bottom - this steep slope illustrates the concept of resilience. If the
1474 perturbation is large enough, the ball may be pushed across the lip of the cup (i.e., the ecological
1475 threshold) and eventually settle into the bottom of a different cup.
2 In contrast MCLs are National Primary Drinking Water Regulations (NPDWRs) established by EPA as legally
enforceable standards that can be applied to public water systems to ensure safe drinking water supply to the public
(USEPA, 2010c). An MCL is defined as the "highest level of a contaminant that is allowed in drinking water"
(USEPA, 2009a). While the MCL is set such that it is as close to the MCLG as possible, it is typically higher than
the MCLG as it is determined based not only on health considerations, but also on the sensitivity of analytical
techniques available to detect the contaminant as well as on the availability of treatment technologies and the extent
to which they can remove the contaminant from drinking water (USEPA, 2009a).
Do Not Cite or Quote Page 47
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
1476 Identifying precise ecological thresholds is widely considered to be a difficult task. Ecosystems
1477 can be, and often are, a complex mix of biotic and abiotic elements that are difficult to evaluate.
1478 Aside from the complex logistics of examining multiple variables simultaneously over
1479 ecologically-relevant timescales, ecosystem evaluations can be complicated by the influence of
1480 exogenous factors (e.g., climate, human interference) that introduce uncertainty into
1481 observations. Furthermore, it is reasonable to believe that many ecosystems are truly unique,
1482 meaning that even if ecological thresholds are well understood, they are not widely applicable
1483 for the purposes of understanding vulnerability at broad scales. Finally, in many cases, ecological
1484 thresholds are difficult to observe unless breached, and the alternate stable state may not be
1485 desirable for social, environmental, or economic reasons. Thus, experiments designed to observe
1486 ecological thresholds through artificial induction of an alternate stable state are not commonly
1487 implemented.
1488
1489 As the science of alternate stable states advances, it may be possible to define objective
1490 thresholds for some of the aquatic ecosystem vulnerability indicators in this study. In the
1491 meantime, relative comparisons of indicator values can be made, and the range of values may or
1492 may not extend across thresholds that could be used to distinguish between vulnerable and less
1493 vulnerable areas.
1494
1495 Sustainability thresholds differentiate between sustainable and unsustainable conditions. In the
1496 context of this study, Sustainability thresholds are most useful in determining where a water
1497 resource may currently be being used unsustainably. The construction of indicators that use
1498 Sustainability thresholds differs somewhat from other indicators. Instead of directly measuring an
1499 environmental condition, they frequently use ratios that attempt to identify whether or not a
1500 system is in balance. These ratios may help answer basic questions for a given area, such as "Do
1501 groundwater withdrawals exceed groundwater recharge?" Or "Do surface water discharges equal
1502 surface water withdrawals?"
1503
1504 The critical value for many ratios centered on these questions is one. For example, for a
1505 theoretical indicator evaluating the balance between groundwater withdrawals and groundwater
1506 recharge, the indicator values may be calculated as Recharge / Withdrawals. Areas where the
1507 value of this ratio is greater than one have more groundwater available than is currently be used
1508 and could be considered sustainable (i.e., providing a "safe yield"). These areas could also be
1509 considered less vulnerable to additional exposure to stresses that reduce groundwater availability.
1510 Conversely, values less than one indicate areas where groundwater withdrawals exceed recharge
1511 - a potentially unsustainable condition. These areas would be more vulnerable to further
1512 exposure to climate-related stresses that reduce recharge.
1513
1514 We calculated values and produced maps for the 25 indicators described in Section IV.C, and
1515 included in Appendix F. When available, we applied objective threshold values identified in the
1516 literature, as shown in Table 5. In these cases, data were divided into two or more categories as
1517 specified in the literature. In cases where objective thresholds were not available and
1518 visualization of changes in indicator values along a gradual gradient was more appropriate, we
1519 produced maps using a continuous grayscale color ramp.
Do Not Cite or Quote Page 48
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
1520 Table 5. Indicators with Objective Thresholds and their Vulnerability Categories
Indicator
Literature
Source
Vulnerability Categories and Thresholds
Instream
Use/Total
Stream flow
(#351)
Meyer et al.,
1999.
No thresholds were provided in Meyer et al. (1999). However, the original
data source (WRC, 1978) used a threshold of one to indicate regions where
water exports are already adversely affecting the instream environment. We
displayed this indicator in Appendix F with the following categories: <1.00
(sustainable) and >1.00 (unsustainable).
Precipitation
Elasticity of
Stream flow
(#437)
Sankarasubr
amanian et
al., 2001.
Sankarasubramanian (2001) identified a value of one as a breakpoint
between elastic and non-elastic responses in streamflow to precipitation.
We displayed this indicator in Appendix F with the following categories: <1
(inelastic) and >1 (elastic).
Total Use / Total
Streamflow
(#352)
Meyer et al.,
1999
No thresholds were provided in Meyer et al. (1999). However, the original
data source (WRC, 1978) used a threshold of one to indicate a potential
conflict between offstream uses and the estimated instream flow needs. We
displayed this indicator in Appendix F with the following categories: <1.00
(sustainable) and >1.00 (unsustainable).
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
B. Modifying and Refining Indicators to Incorporate Thresholds
A major strength of the approach pursued in this study is the use of readily available data, much
of which has been vetted by other researchers, agencies, or institutions. Few indicators, however,
directly incorporate objective thresholds. Such thresholds, as noted above, can be highly useful
to decision makers, especially when they distinguish between acceptable and unacceptable
conditions. In some cases, slight modification of an indicator definition can facilitate the
identification of objective thresholds. For example, the pesticide indicators (#367, #369, #371,
#373, and #374) do not incorporate regulatory or human health thresholds because these
indicators are calculated as aggregates of multiple pesticides, some of which are unregulated, and
whose health effects are less well understood. As an alternative, a predictive model (Larson et
al., 2004) is used to map the average probability of exceeding the human health threshold
(maximum contaminant level (MCL)) for atrazine, which is the most commonly used herbicide
(Figure 3). The predictive modeling approach is currently being expanded by USGS to other
pesticides (R. Gilliom, personal communication). Because these models are built from variables
that may be affected by climate change, they may be particularly well-suited to assessing
changes in vulnerability across different scenarios of climate and land-use change.
In addition, new indicators may be developed by integrating multiple existing data sets. For
example, methylmercury production potential could be a useful indicator of vulnerability of
aquatic animals to anthropogenic waste. Currently, there is no existing data source that describes
methylmercury potential across the entire U.S. However, a new analysis could be conducted
using data for wet soils, temperature, and methylmercury deposition, to assess exposure of
aquatic life to this contaminant. Existing data sets could be used for the variables in such an
analysis, such as wet soils data from the United States Department of Agriculture-Natural
Resources Conservation Service (USDA-NRCS, http://soils.usda.gov/); temperature data from
Do Not Cite or Quote
Page 49
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
1547 Figure 3. Mapping Data Relative to Regulatory Thresholds
1548 This map displays the probability of predicted concentrations ofatrazine, a pesticide, exceeding its regulatory threshold (i.e., its
1549 Maximum Contaminant Level or MCL). The resulting map places pollutant concentrations into a human health context.
1550
Average Probability of Exceeding
the Atrazine MCL (3 ug/L)
| | 0.00 - 0.03
j ^\ 0.04 - 0.06
^^ 0.07-0.09
^B 3.10-G.12
^H '3-13-0.14
II States
100 200 300 400 500 Miles
I I I I I
Do Not Cite or Quote
Page 50
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
1551 NOAA's NCDC (http://www.ncdc.noaa.gov/oa/ncdc.html); and atmospheric deposition data
1552 from the University of Illinois Urbana-Champaign's National Atmospheric Deposition Program
1553 (NADP; http://nadp.sws.uiuc.edu/). Development of such aggregate indicators using easily
1554 available existing data sets may yield additional useful indicators that are critical for assessing
1555 regional vulnerability.
1556
1557 An alternative approach would be to define ideal water quality and aquatic ecosystem
1558 vulnerability indicators, and then appropriately transform existing data or collect new data to
1559 assess vulnerability. Development of indicators that more directly compare the sensitivity and
1560 exposure components of vulnerability would facilitate a quantitative comparison of their relative
1561 importance. For instance, in an effort to understand the relative importance of temperature and
1562 population changes on groundwater availability, water use indicators may have to be scaled
1563 relative to water availability or per capita demand. As an example, groundwater availability per
1564 capita could accommodate adjustments from these diverse influences: precipitation effects on
1565 recharge, temperature effects on evaporation, and population effects on demand. The hydrologic
1566 component of this evaluation would require a model whose drivers include climate variables,
1567 scenarios of whose future values can be developed. Creating primary indicators of ecological
1568 function would allow for similar evaluations. Although an approach that defines ideal indicators
1569 may yield objective thresholds/breakpoints and clear connections to the three aspects of
1570 vulnerability, it is likely that difficulties in collecting all requisite data would limit the number of
1571 indicators that could be constructed. However, Figure 4 and Figure 5 represent examples of two
1572 indicators that can be developed using existing data. Figure 4 depicts total water use efficiency, a
1573 modification of the industrial water use efficiency indicator cited in Kurd et al., 1998. Figure 5
1574 depicts total water demand for human uses. Both indicator maps were created using the USGS
1575 National Water-Use Dataset to provide a complete picture of U.S. water use.
1576
1577 The National Environmental Status and Trend (NEST) Indicator Project used another approach
1578 to assemble a suite of indicators. The process used in that project included the distillation of
1579 many perspectives on water into five categorical questions (Table 6) that guided the search and
1580 development of indicators. All of the questions are addressed to some extent by the indicators
1581 mapped during this project, although some key subcategories do not have representative
1582 indicators. Some of these indicator classes could be filled by further examination of existing
1583 data, but others would require additional data collection efforts. Several published examples of
1584 these indicator classes were included in the comprehensive list of indicators first assembled for
1585 this project, but were subsequently eliminated based on a lack of data, data gaps, or unreliable
1586 quality of the available data sets, or inadequate or incomplete data collection efforts. Data
1587 collection or manipulation efforts geared specifically towards informing these indicators, such as
1588 those discussed below, might provide the necessary data for creating national-scale maps.
1589
Do Not Cite or Quote Page 51
-------
1590
1591
1592
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Figure 4. Modification of Indicator Definitions Using Existing Data
This map of 1995 Water Use Efficiency is a refinement of indicator #135. This example demonstrates how minor refinements using
existing data sets may result in indicators that more directly assess vulnerability.
1593
% of Withdrawals Released
(Withdrawals - [Conveyance Loss +• Consumption]) / Withdrawals
| | 94%-98%
| | 82% - 93%
^B 46%-81%
I 29% - 45%
0 100 2CO 30C 400 5DO Mites
I I I I I I
States
Do Not Cite or Quote
Page 52
-------
1594
1595
1596
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Figure 5. Modification of Indicator Definitions Using Existing Data
This map of 1995 Water Demand was developed using data sets that were also used to develop indicators #125 and #135. Many of the
available data sets used to develop the indicator maps can be used to develop additional indicators of vulnerability.
1597
Total Water Withdrawals
MGal/Day
| | 2.1-330
i~^| 340-910
11 920-1,700
^B 1.800-3,200
^B 3-300-11,000
I ~~| States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
Page 53
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
1598
1599
1600
1601
1602
Table 6. Vulnerability Indicators Categorized in the National Environmental Status and Trend
(NEST) Framework
Vulnerability indicators from this project categorized according to the question framework from
the National Environmental Status and Trend (NEST) Indicator Project. Indicators numbers
associated with subcategories are discussed in Table 13.
NEST Question
How much water do
we have?
How much water do
we use?
What is the condition
of aquatic ecological
communities?
What is the physical
and chemical quality of
our water?
Is the water we have
suitable for human use
and contact?
No clear fit to above
questions
Example Indicators
• Meteorological Drought Indices (#165)
• Ratio of Snow to Precipitation (S/P) (#218)
• Precipitation Elasticity of Streamflow (#437)
• Ratio of Reservoir Storage to Mean Annual
Runoff (#449)
• Runoff Variability (#453)
• Groundwater Reliance (#125)
• At-Risk Freshwater Plant Communities (#22)
• At-Risk Native Freshwater Species (#24)
• Stream Habitat Quality (#284)
• Wetland and Freshwater Species at Risk (#326)
• Macroinvertebrate Index of Biotic Condition
(#460)
• Macroinvertebrate Observed/Expected Ratio of
Taxa Loss (#461)
• Acid Neutralizing Capacity (ANC) (#1)
• Herbicide Concentrations in Streams (#367)
• Insecticide Concentrations in Streams (#369)
• Organochlorines in Bed Sediment (#371)
• Herbicides in Groundwater (#373)
• Insecticides in Groundwater (#374)
• Coastal Vulnerability Index (#51)
Subcategories Not Represented
• Flooding (e.g., Population
Susceptible to Flood Risk
[#209])
• Groundwater availability (e.g.,
Groundwater Depletion
[#121])
• Total water use (e.g., Ratio of
Water Use to Safe Yield
[#328])
• Habitat Fragmentation (e.g.,
In-Stream Connectivity
[#620])
• Nutrients (e.g., Water Quality
Index [#319])
• Recreational water quality
• Waterborne pathogens (e.g.,
Waterborne Human Disease
Outbreaks [#322])
1603
1604
1605
1606
1607
1608
1609
1610
1611
VI. Challenges Part III: Mapping Vulnerability
Producing a single map to represent numerical data from disparate sources in an accurate and
unbiased manner is a classic cartographic challenge. This challenge is rooted in the fact that "a
single map is but one of an indefinitely large number of maps that might be produced.. .from the
same data" (Monmonier, 1991). The choices made with regard to the metrics calculated, the
categories used to generalize those metrics, the spatial units used to aggregate localized data, and
the symbols used to display map features can all lead to substantially different maps.
Furthermore, these choices can be used to emphasize or minimize spatial trends and patterns.
Do Not Cite or Quote
Page 54
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
1612 The effort to produce indicator maps for this study was met with these same cartographic
1613 challenges. The following sections discuss these challenges in greater detail and provide example
1614 maps, using the indicators discussed above, to illustrate how these challenges can affect use of
1615 indicators for assessments of vulnerability across the nation.
1616
1617 Mapping the above-described indicators at the national scale requires the compilation of multiple
1618 reliable data sets that provide consistent sample density at this scale. In recent years, agencies
1619 such as EPA, USGS, and NOAA have invested considerable resources to develop such data sets.
1620 These are immensely informative and were used to develop many of the maps contained in this
1621 report.
1622
1623 A. at the
1624
1625 We examined the 53 vulnerability indicators (see Table 4 and Figure 1) for data availability and
1626 mappability, in the process identifying existing, available data that could potentially be used for
1627 creating national maps for each of these indicators.
1628 a. Identification of data sources for indicators
1629 We determined data availability for each indicator by re-examining the literature in which the
1630 indicator was cited. In most cases, the study that cited the indicator also cited a data set, either
1631 one that was collected and assembled during the study itself or a publicly available data set
1632 containing data compiled by the authors of the study or by one or more private or public entities.
1633 If no specific data set was cited in the original literature, data sets recommended by team
1634 members or technical advisors were used. If a data set was not available or could not be
1635 recommended, the indicator was marked as having no associated data and was not evaluated for
1636 mapping.
1637
1638 Data availability was the most serious limitation in evaluating whether or not we could produce
1639 maps for the 53 vulnerability indicators. Of these, only 32 indicators were initially assessed as
1640 having adequate data (using data sources identified in the literature) for nationwide mapping.
1641 Furthermore, not all of these 32 indicators could be mapped, as the data sources referenced in the
1642 literature were not always tailored specifically to the indicator. This was frequently the case with
1643 indicators that were identified by one entity and whose data were collected by another entity. In
1644 contrast, several indicators identified in USGS' The Quality of Our Nation's Waters report (e.g.,
1645 Herbicide Concentrations in Streams [#367]; Insecticide in Groundwater [#374];
1646 Organochlorines in Bed Sediment [#371]) are based on NAWQA data that are also collected by
1647 USGS.
1648
1649 For indicators that met minimum criteria for availability and for which we identified data sets,
1650 nationwide mappability at the level of 4-digit HUC watersheds (as a minimum screening
1651 criterion) was assessed simultaneously with data availability. This was because we found that it
1652 was not possible to establish mappability without beginning the process of manipulating and
1653 mapping the data to determine what obstacles there may be to mapping.
Do Not Cite or Quote Page 55
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
1654 b. Description of major data sources
1655 The data sets identified for these 53 indicators varied in size, level of detail, quality, and
1656 relevance to the indicator. Some data sets were collected specifically with the concerned
1657 indicator in mind; in other cases, the indicator was designed with a specific data source in mind.
1658 From an initial assessment of data sources, it was evident that major national organizations, such
1659 as EPA, USGS, NOAA, and NatureServe, were key players in national-scale data collection
1660 efforts for indicators of water quality and aquatic ecosystems. For some indicators, we used data
1661 sets produced by other organizations or published in peer-reviewed literature.
1662
1663 A distribution of how often we used data sources from these organizations and other entities for
1664 assessing indicator mappability is shown in Table 7 (Distribution of Data Sources). The
1665 following 14 indicators (out of 53) had no data available and are, therefore, not included in the
1666 39 indicators in the table: Flood Events (#100), At-Risk Native Marine Species (#27),
1667 Freshwater Rivers and Streams with Low Index of Biological Integrity (#116), Harmful Algal
1668 Blooms (#127), Invasive Species-Coasts Affected (#145), Invasive Species in Estuaries (#149),
1669 Riparian Condition (#231), Status of Animal Communities in Urban and Suburban Streams
1670 (#276), Streamflow Variability (#279), Snowmelt Reliance (#361), Salinity Intrusion (#391),
1671 Threatened and Endangered Plant Species (#467), Vegetation Indices of Biotic Integrity (#475),
1672 and In-stream Connectivity (#620). See Appendix C for a complete and more detailed listing of
1673 data sources for each of the 39 indicators in Table 7.
1674
1675 Table 7. Distribution of Data Source
Indicator
Acid Neutralizing
Capacity (ANC)
(ffl)
Altered
Freshwater
Ecosystems (ff!7)
At-Risk
Freshwater Plant
Communities
(#22)
At-Risk Native
Freshwater
Species (ff24)
Data Source Organization
EPA
X-
Wadeable
Streams
Assessment
X-
National
Land Cover
data set
(NLCD)
USGS
X- National
Hydrography
data set (NHD)
NOAA
NatureServe
X-
Customized
data set
X-
Customized
data set
Other
X- U.S. Fish & Wildlife
Service's (USFWS)
National Wetlands
Inventory (NWI)
Do Not Cite or Quote
Page 56
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
Coastal Benthic
Communities
(#462)
Coastal
Vulnerability
Index -CVI(ffSl)
Commercially
Important Fish
Stocks (ff55)
Erosion Rate
(#348)
Fish and Bottom-
Dwelling Animals
(#95)
Groundwater
Depletion (#121)
Groundwater
Reliance (#125)
Heat-Related
Illnesses
Incidence (#392)
Herbicide
Concentrations in
Streams (#367)
Herbicides in
Groundwater
(#373)
Data Source Organization
EPA
X-
Sampling
data in
National
Coastal
Assessment
(NCA)
database
X-
Wadeable
Streams
Assessment
(WSA)
USGS
X- National
Water-Use
Dataset
X- National
Water-Use data
set
X- NAWQA
X - NAWQA
NOAA
X-Annual
Commercial
Landing
Statistics
NatureServe
Other
X- Carbon Dioxide
Information Analysis
Center's (CDIAC) Coastal
Hazards Database
X-Yang, D. W.,S.
Kanae, T. Oki, T. Koike,
and K. Musiake. 2003.
Global Potential Soil
Erosion with Reference
to Land Use and Climate
Changes. Hydrological
Processes 17:2913-2928.
X - National Center for
Health Statistics
(NCHS)'s Mortality data
Do Not Cite or Quote
Page 57
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
Insecticide
Concentrations in
Streams (ff369)
Insecticides in
Groundwater
(#374)
Instream
Use/Total
Stream/low
(ff351)
Low Flow
Sensitivity (#159)
Macroinvertebrat
e Index ofBiotic
Condition (#460)
Macroinvertebrat
e Observed/
Expected (O/E)
Ratio of Taxa
Loss (#461)
Meteorological
Drought Indices
(#165)
Number of Dry
Periods in
Grassland/
Shrubland
Streams and
Rivers (#190)
Organochlorines
in Bed Sediment
(#371)
Pesticide Toxicity
Index (#364)
Population
Susceptible to
Flood Risk (#209)
Data Source Organization
EPA
X-
Wadeable
Streams
Assessment
X-
Wadeable
Streams
Assessment
USGS
X - NAWQA
X - NAWQA
X- National
Water-Use
Dataset
X- Hydro
Climatic Data
Network
(HDCN)&
Stream Gauge
Data
X - NAWQA
X - NAWQA
NOAA
X- Divisional
Data on the
Palmer
Drought
Severity Index
(PSDI)
NatureServe
Other
X - Water Resources
Council. 1978. The
Nation's Water
Resources: The Second
National Water
Assessment, 1975-2000.
Volume 2.
X- EPA's ECOTOX
database
X-FEMA'sQ3 Flood
Data & ESRI ArcUSA's US
Census tract data
Do Not Cite or Quote
Page 58
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
Precipitation
Elasticity of
Stream/low
(#437)
Ratio of
Reservoir Storage
to Mean Annual
Runoff (#449)
Ratio of Snow to
Total
Precipitation
(#218)
Ratio of Water
Use to Safe Yield
(#328)
Ratio of Water
Withdrawals to
Annual
Stream flow
(#219)
Runoff Variability
(#453)
Stream Habitat
Quality (#284)
Total Use/Total
Streamflow
(#352)
Data Source Organization
EPA
X-
Wadeable
Streams
Assessment
uses
X- HDCN
X-Mean
Annual Runoff
Data
X- National
Water-Use
Dataset
NOAA
X- Monthly
Climate Data
NatureServe
Other
X - Oregon State
University's PRISM
Climate Modeling
System
X- USAGE'S National
Inventory of Dams (NID)
X-Schmitt, C. V.,
Webster, K. E.,
Peckenham, J. M.,
Tollman, A. L, and J. L.
McNelly. 2008.
Vulnerability of Surface
Water Supplies in Maine
to the 2001 Drought.
Journal of the New
England Water Works
Association. 122 (2):
104-116.
X - Oregon State
University's PRISM
Climate Modeling
System
X- University of
Washington's Variable
Infiltration Capacity
(VIC) Land Surface Data
Set
X - Water Resources
Council. 1978. The
Nation's Water
Resources: The Second
National Water
Assessment, 1975-2000.
Volume 2.
Do Not Cite or Quote
Page 59
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
Water
Availability: Net
Stream/low per
Capita (#623)
Water Clarity
Index (ff318)
Water Quality
Index (ff319)
Waterborne
Human Disease
Outbreaks (#322)
Wetland and
Freshwater
Species at Risk
(#326)
Wetland Loss
(#325)
Data Source Organization
EPA
X-NCA
X-NCA
uses
X- National
Water-Use
Dataset
NOAA
NatureServe
X-
Customized
data set
Other
X - Oregon State
University's PRISM
Climate Modeling
System
X - Centers for Disease
Control and Prevention
(CDC)'s Waterborne
Disease and Outbreak
Surveillance System
(WBDOSS)
X-USFWS National
Wetlands Inventory
(NWI)
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
As can be seen in Table 7, some data sources furnished data for multiple indicators. These major
data sources are discussed in greater depth below.
• EPA 's Wadeable Streams Assessment (WSA)
EPA's WSA was designed to be the first statistically defensible summary of the condition of
the nation's streams and small rivers. Chemical, physical, and biological data were collected
at 1,392 wadeable perennial stream locations in the coterminous United States. Data were
collected by field crews during summer index periods between 2000 and 2004. Sample sites
were selected using a probability-based sample design; rules for site selection included
weighting based on the 1st- through 5th-order stream size classes and controlled spatial
distribution. Due to this sampling system, the sampling effort for the WSA varies across
HUC-4 units. Because a probability-based sampling design was used, the WSA data set may
have avoided the bias that may occur with ad hoc data sets. However, it is still less than ideal
for mapping average conditions in 4-digit HUCs because lakes, reservoirs, and large rivers
were not sampled, and because some HUCs had few or no sampling sites.
• USGS 's National Water-Quality Assessment (NA WQA) Program
USGS's NAWQA Program collects chemical, biological, and physical water quality data.
From 1991 to 2001, the NAWQA program collected data from 51 study units (basins) across
the United States; after 2001, data collection continued at 42 of the study units. Although the
program spanned 10 years, not all 51 sites were sampled every year, but were, instead,
Do Not Cite or Quote
Page 60
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
1699 broken up into smaller temporal frames (20 study units in 1991; 16 study units in 1994; and
1700 15 study units in 1997).
1701
1702 The NAWQA data warehouse currently contains sampling information from 7,600 surface
1703 water sites (including 2,700 reach segments for biological studies) and 8,800 wells. The
1704 NAWQA sampling design uses a rotational sampling scheme; therefore, sampling intensity
1705 varies year to year at the different sites. In general, about one-third of the study units are
1706 intensively investigated at any given time for 3-4 years, followed by low-intensity
1707 monitoring. Due to this sampling scheme, the sampling effort for the NAWQA Program
1708 varies across HUC-4 units.
1709
1710 • USGS' National Water- Use Dataset
nn
1712 USGS's National Water-Use Dataset contains water-use estimates for each county in the
1713 United States, the District of Columbia, Puerto Rico, and the U.S. Virgin Islands. USGS
1714 publishes reports every five years (starting in 1985) that present water-use information
1715 aggregated at the county, state, and national levels. USGS study chiefs from each state are
1716 responsible for collecting and analyzing information, as well as making estimates of missing
1717 data and preparing documentation of data sources and methods used to collect those data.
1718 The study chiefs are also responsible for determining the most reliable sources of information
1719 available for estimating water use for each state. Because of this, data sources and quality
1720 may vary by location.
1721
1722 • NOAA 's Monthly Climate Data
1723
1724 NOAA's National Climatic Data Center (NCDC) is the world's largest active archive of
1725 weather data. NCDC's Monthly Climate Data Set contains information collected for 18,116
1726 sites across the United States from 1867 to the present. The data set includes an assortment of
1727 parameters such as measurements of rain, snow, evaporation, temperature, and degree days.
1728 NCDC Monthly Climate data are primarily intended for the study of climate variability and
1729 change. NOAA reports that, whenever possible, NCDC observations have been adjusted to
1730 account for effects from factors such as instrument changes, station relocations, observer
1731 practice changes, and urbanization.
1732
1733 • NatureServe Data Set Customized for EPA
1734
1735 NatureServe collects and manages detailed local information on plants, animals, and
1736 ecosystems though natural heritage programs and conservation data centers operating in all
1737 50 U.S. states, Canada, Latin America, and the Caribbean. The data sets were originally
1738 customized for the Heinz Center for publication in the 2008 State of the Nation's Ecosystems
1739 report. We obtained updated state-level data on At-Risk Native Freshwater Species (#24) and
1740 on At-Risk Freshwater Plant Communities (#22) to produce the maps for these indicators in
1741 this study. These data sets were provided in Excel format by NatureServe on July 29, 2009.
1742 Data on freshwater species were updated from those presented in the Heinz Center, 2008
1743 report, and included counts of at-risk (GX-G3) and total native freshwater animal species by
1744 state for the U.S. Due to incomplete state distribution, the data set did not include giant
Do Not Cite or Quote Page 61
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
1745 silkworm moths, royal moths, sphinx moths, or grasshoppers. NatureServe did not update
1746 data on plant communities as they determined that plant community data have not changed
1747 significantly since the original analysis for the Heinz Center.
1748 c. Supporting information collected for data sources
1749 To assess data availability, we isolated information about the underlying data on which the
1750 indicators were based. This information is also presented in Appendix C (Data Sources and
1751 Supporting Information). Information considered when assessing the mappability of data
1752 included:
1753
1754 • Data sets used and the organizations or individuals who published or own the data;
1755 • How to obtain the data (download online or contact a specific person/organization) and
1756 whether or not payment was necessary to obtain the data set;
1757 • Spatial resolution of data (e.g., state, study sites, HUC level, ecoregion);
1758 • Temporal resolution of data (i.e., frequency of data points and duration of data
1759 collection);
1760 • Extent of coverage of data (e.g., national, regional, state, local);
1761 • Type of data source (e.g., survey, census, database, modeled data set);
1762 • Format of data (e.g., Excel tables, GIS shapefiles); and,
1763 • Relevant metadata (either as a website or a supporting document).
1764
1765 In many cases, the supporting documentation accompanying the data did not provide all of the
1766 abovementioned details. However, the available information has proven useful for prioritizing
1767 indicators for further investigation into their mappability.
1768 d. Lack of data and other unresolved data problems
1769 1. Data availability issues
1770 To streamline the process of determining indicator mappability, we identified issues with
1771 data availability and how data was presented as early in the process as possible. We
1772 encountered problems both in the effort to locate, access, and download indicator data and in
1773 the effort to manipulate, transform, or modify the data so that they could be mapped using
1774 GIS software at the appropriate scale. Based on our assessment of data availability, 28
1775 indicators were determined to be non-mappable. Although data sets were available for a few
1776 of these indicators, the problems with the data sets could not be reconciled, even with greater
1777 time and effort spent on data manipulation and mapping, and, therefore, these indicators were
1778 considered non-mappable. These 28 indicators presented one or more of the problems listed
1779 in Table 8 (Indicators Eliminated Due to Lack of Data or Unresolved Data Problems).
1780
Do Not Cite or Quote Page 62
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
1781 Table 8. Indicators Eliminated Due to Lack of Data or Unresolved Data Problems
Data
Availability
Problem
Description of the
Problem
Example
Indicators
Specific Data Availability Problem
Data
reported by
individual
states
Reporting, sampling, and
assessment methods vary
between states. These
indicators are likely to
reflect programmatic
differences instead of
differences in
vulnerability.
Fish and Bottom-
Dwelling Animals
(#95)
The indicator is derived from STORET, a
database that relies substantially on self-
reported data.
Waterborne
Human Disease
Outbreaks (#322)
The WBDOSS datasets relies on voluntary
reporting from public health departments
within the United States.
303(d) Impaired
Waters3
The ATTAINS database relies on data reported
by individual states.
Multiple
Data Sets
Complete data set could
only be obtained by
combining more than one
data set, as specified in
the literature. The effort
necessary to combine the
data ranges widely.
Population
Susceptible to
Flood Risk (#209)
This would require combining digital flood
data from FEMA (unavailable at time of
inquiry) and Census Bureau demographic
data.
Water Quality
Index (#319)
Five data sets combined into an index.
Wetland Loss data
(#325)
USFWS' National Wetlands Inventory data are
at different scales at different locations.
Coastal Benthic
Communities
(#462)
Benthic indices vary by region and it is unclear
whether regional indices are comparable.
Data set
derived from
extensive
modeling
Complete data set needed
to be recreated with
extensive modeling using
raw data.
Groundwater
Depletion (#121)
Indicator based on a modeled base-flow data
set developed by Vogel et al., 1999 and
presented in Hurd et al. (1998).
Low Flow
Sensitivity (#159)
Indicator based on a modeled base-flow data
set developed by Vogel et al., 1999 and
presented in Hurd et al. (1998).
Stream Flow
Variability (#279)
Indicator based on a model developed by
Vogel et al., 1999 and presented in Hurd et al.
(1998).
Data
collection in
progress
Data are unavailable
because collection efforts
are in progress.
In-stream
connectivity
(#620)
USGS is currently collecting data on indicator
as a part of its National Hydrography Dataset.
This indicator was not assigned an indicator ID# because it was not derived from the scientific literature.
indicator was added to incorporate EPA's extensive water quality assessment database.
The
Do Not Cite or Quote
Page 63
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Data
Availability
Problem
Description of the
Problem
Example
Indicators
Specific Data Availability Problem
Not national,
recent, or
current
Data are unavailable
nationally, or are not
recent enough (cutoff date
varies with the indicator),
or are based on future
projections.
Number of Dry
Periods in
Grassland /
Shrubland
Streams and
Rivers (#190)
The data set identified by the Heinz Center
contained an analysis of grassland and
shrubland watershed areas for Western
ecoregions only.
Water Clarity
Index (#318)
Data are only available for certain US coastal
regions.
Waterborne
Human Disease
Outbreaks (#322)
Most recent data are from 2006, and
according to the Heinz Center (2008), data are
no longer reported.
Heat-Related
Illnesses Incidence
(#392)
Data comprised of projections for the years
2020 and 2050.
Invasive Species-
Coasts Affected
(#145)
This indicator evaluates invasive species
within the context of local land use, a scale
that is relatively uncommon. No national
datasets have been identified that
simultaneously evaluate local land
management and the presence of invasive
species.
Ratio of Water
Use to Safe Yield
(#328)
Data set identified by the source only contains
data for the state of Maine.
Salinity Intrusion
(#391)
Data sources cited in the information source,
(Poff et al., 2002) are local studies with limited
(and non-comparable) data sets. No
comprehensive national data sets are known
to exist.
Conceptual
indicator
without
existing data
set
Indicator is conceptual or
theoretical in nature. Data
for the indicator are
unavailable or have been
identified by the original
investigator as a data
need.
At-Risk Native
Marine Species
(#27)
The Heinz Center 2008 study, which is the
source of this indicator, identifies NatureServe
as a potential source of information relevant
to this indicator, but acknowledges that data
availability is limited to a small set of species.
Flood Event
Frequency (#100)
No data source was identified in this study
that could be used to map this indicator at a
national scale.
Freshwater Rivers
and Streams with
Low Index of
Biological
Integrity (#116)
There are currently no regional or national
data bases that assemble this information for
a broad range of taxa.
Do Not Cite or Quote
Page 64
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Data
Availability
Problem
Duplicate
Indicator
Description of the
Problem
Data are available, but the
indicator was a duplicate
of another indicator.
Example
Indicators
Harmful Algal
Blooms (#127)
Invasive Species in
Estuaries (#149)
Status of Animal
Communities in
Urban and
Suburban Streams
(#276)
Riparian Condition
Index (#231)
Snowmelt
Reliance (#361)
Threatened &
Endangered Plant
Species (#467)
Vegetation Indices
of Biotic Integrity
(#475)
Altered
Freshwater
Ecosystems
(percent miles
changed) (#17)
Commercially
important fish
stocks (#55)
Fish and Bottom
Dwelling Animals
(#95)
Specific Data Availability Problem
Currently, there are no nationwide monitoring
or reporting programs for harmful algal
events.
Currently, there are no national monitoring
programs for invasive species in estuaries and
no agreed-upon methods for combining
information on the number of species and the
area they occupy into a single index.
The Heinz Center 2008 study, which is the
source of this indicator, states that currently
available data are not adequate for national
reporting.
The Heinz Center 2008 study, which is the
source of this indicator, identifies four
literature sources that outline various ways to
create such an index, but acknowledges that
no raw data are currently available.
The information source (IPCC, 2007) only has
theoretical discussion of indicator. No specific
data source is cited.
This indicator was provided as an example
EPA's National Wetland Condition
Assessment. This report does not identify a
specific data source for this indicator.
This indicator was provided as an example
EPA's National Wetland Condition
Assessment. This report does not identify a
specific data source for this indicator.
A national database with the number of
impounded river miles does not exist. Data
from three sources need to be integrated, one
of which currently does not provide data in
electronic form.
Data for change in fish stock size over time are
not currently available. The change in a fish
stock size over time would need to be
calculated for each area where fish stock data
are available.
1782
Do Not Cite or Quote
Page 65
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
1783 This table highlights two challenges to the adoption and use of indicators at a national scale.
1784 First, it draws attention to the issue of measurability. In many cases, a measurable indicator
1785 requires a substantial effort to calculate the value at a single location. This may be due to the
1786 need for prolonged observation periods, complex sampling protocols, or other factors. For
1787 example, Vegetation Indices of Biotic Integrity (#475) uses the relationships between
1788 anthropogenic disturbances and observations of plant species, plant communities, plant guilds,
1789 vegetation structure, etc. to describe wetland condition. Typically, the highest IBI values
1790 represent reference standards or least-disturbed ecological conditions. To collect the data
1791 required to calculate an IBI, a trained observer must record multiple parameters in the field for
1792 each local IBI score. Though the indicator is measurable and highly useful in the locations where
1793 data exist, the effort required to collect data for this indicator at a national scale may be
1794 prohibitive.
1795 Second, Table 6 highlights how data sources that may otherwise be excellent may be problematic
1796 for the purposes outlined in this study. We will discuss the issue of self-reported data in further
1797 detail as an example. Data sets that rely on individual state reports are problematic for three
1798 reasons. First, the monitoring activities and subsequent reporting may be limited by the
1799 availability of the state's resources. This can result in data gaps stemming from varying levels of
1800 reporting activity across states. Second, state-based assessments that require sampling from a
1801 population (e.g. stream assessments) may not rely on statistically rigorous sampling methods,
1802 resulting in sampling that may not be representative. Third, assessment methods may vary from
1803 state to state. For example, the assessment and classification methods used by states during the
1804 development of the 303(d) impaired waters lists vary substantially among states. Together, these
1805 inconsistencies in reporting, sampling, and assessment result in maps that may reflect
1806 programmatic differences instead of actual differences in vulnerability. For these reasons,
1807 indicators based on national data sets that had national coverage but rely on individual entities to
1808 voluntarily report data, (e.g., EPA's Storage and Retrieval (STORET) database for water quality
1809 data, CDC's Waterborne Disease and Outbreak Surveillance System (WBDOSS), and EPA's
1810 Assessment, TMDL Tracking and ImplementatioN System (ATTAINS) database), were not used
1811 in the present study.
1812
1813 Figure 6 shows a national map that relies on one such national data set, the ATTAINS database.
1814 Panel A shows a map that relies on the total stream-miles designated as 303(d) impaired waters.
1815 This first map is problematic because it does not account for large differences in assessment rates
1816 across states, or for the fact that overall assessment rates are low. According to the EPA
1817 ATTAINS database, only 26.4% of the nation's streams and rivers and 42.2% of the nation's
1818 lakes and reservoirs have been assessed for impairments, making it difficult to create national -
1819 scale indicators. Panel B attempts to account for differences in assessment rates by showing the
1820 percentage of assessed stream-miles that are designated as 303(d) impaired waters. Though this
1821 second map is an improvement over the first because it normalizes the assessment effort, the
1822 programmatic differences still result in areas that may not appear to be vulnerable simply
1823 because sampling and assessment methods vary substantially between states. Conversely, areas
1824 that appear to be the most vulnerable may attract restoration efforts in the near term, leading to a
1825 restored condition and enhanced resilience.
Do Not Cite or Quote Page 66
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
Figure 6. Limitations of Data Sets Containing Self-reported Data
The following maps display number (panel A) and per cent (panel B) of stream-miles designated
as 303(d) impaired waters using data from EPA 's Assessment, TMDL Tracking and
ImplementatioNSystem (ATTAINS) database.
A.
Stream-miles designated as impaired
^0-699
|TOO- 1.590
^^1,600-20,900
I[states
B.
Percent of assessed stream-miles designated as impaired
^| 0% - 20.7%
^| 20.8% - 44.6%
^^ >44.6%
I States
0 200 400 600 800 1,000 Miles
I I I I I I
2. Data sets without national coverage
In some cases, the data required to calculate indicator metrics were incomplete in terms of
national coverage. Indicators based on a particular ecosystem or land cover type (e.g., grassland
or shrubland) may not extend to all parts of the country. For example, few, if any, streams in
Eastern ecoregions are grassland or shrubland streams. Other national coverage data gaps
stemmed from data availability. For example, although 500 year flood plains can be identified
for all parts of the country, GIS-compatible digital flood plain data from FEMA are only
available for certain parts of the country where paper maps have been digitized.
Do Not Cite or Quote
Page 67
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
1840 Other data gaps were the result of incomplete data collection. For example, for the indicator
1841 Commercially Important Fish Stocks (#55), the Heinz Center (2008) study evaluated only about
1842 21% of the commercially important fish landings found in U.S. waters. Similarly, for the
1843 indicator Number of Dry Periods in Grassland/Shrubland Streams and Rivers (#190), the data set
1844 provided by the Heinz Center contained an analysis of grassland and shrubland watershed areas
1845 for Western ecoregions only. Although the reasons for mapping Western ecoregions only are
1846 unclear, it is likely that few, if any, sites in Eastern ecoregions satisfied the definition of a
1847 "grassland" or "shrubland" watershed used in the 2001 National Land Cover Dataset.
1848
1849 In some cases national coverage was unavailable because data collection efforts are still in
1850 progress. For the indicator Wetland Loss (#325), wetlands in 13 states are either unmapped or
1851 are recorded only on hardcopy maps. Similarly, data for the indicator Coastal Benthic
1852 Communities (#462) (from EPA's National Coastal Assessment (NCA)) and digital flood data
1853 for the indicator Population Susceptible to Flood Risk (#209) (from the Federal Emergency
1854 Management Administration (FEMA)) were not available at the time of this study for several
1855 areas within the U.S.
1856 3. Non-uniform spatial distribution of data
1857 In some cases, the national-scale data required to calculate a vulnerability metric are available,
1858 however the data are not distributed homogeneously across the country. As a result, varying
1859 amounts of data are available within each of the HUC-4 units. This variation can be substantial,
1860 and in cases where only few sample points are available within a HUC-4 boundary, individual
1861 sites may exert a large influence on the calculated metric value.
1862
1863 The indicator Acid Neutralizing Capacity (#1), for example, is calculated using data from 1,601
1864 stream sites across the country that were sampled as part of EPA's Wadeable Streams
1865 Assessment. The number of sites sampled within each of the 204 HUC-4 units varies from 0 to
1866 93, with a median value of 5 sample sites. The calculated vulnerability metrics for HUC-4 units
1867 containing the median number of samples (or fewer) are particularly sensitive to measurements
1868 at individual sites. A change in the status of a single site from "not at risk" to "at risk" changes
1869 the calculated metric (percentage of "at risk" sites) by 20%. This could result in the entire HUC-
1870 4 unit being placed in a different category of vulnerability as a result of a single measurement. A
1871 mapping challenge emerges when vulnerability metrics calculated from a small pool of data are
1872 mixed with those calculated from a larger pool. It is difficult, and sometimes impossible, to
1873 illustrate on a single map where low density would be most likely to result in an erroneous
1874 vulnerability classification.
1875 4. Temporal gaps
1876 Many indicators are derived by comparing data contained in two separate data sets, or by
1877 comparing data from one data set collected over two distinct time periods. In the first case, it is
1878 important to consider the time period in which the data are collected, especially if the
1879 information collected may change over time. Temporal gaps between data sets may result in
1880 erroneous vulnerability assessments and inaccurate maps. For example, Net Streamflow
1881 Availability per Capita (#623) depends on time-sensitive information from a range of data sets.
1882 Evaluating streamflow, withdrawals, and population figures from different time periods may
1883 provide a different assessment of vulnerability when compared to data collected from the same
Do Not Cite or Quote Page i
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
year. In the second case, indicators based on comparisons to a historical condition are dependent
on the existence of historical data. For some indicators considered during the course of this
project, this historical information was not available. The Wetland Loss (#325) indicator
provides an example of such a case. Information regarding wetland extent is not available at the
national scale in a format suitable for mapping with a GIS.
Another issue related to temporal gaps pertains to future data collection. One objective of this
project is to identify indicators that can be updated over time to track changes in vulnerability. In
cases where data collection and reporting have been discontinued, the indicator no longer meets
this key objective. The Waterborne Human Disease Outbreaks (#322) and Runoff Variability
(#453) indicators fall into this category. If future data collection efforts are proposed, these
indicators may become more useful for national level assessments.
e. Data problems that could be resolved
Of the 53 indicators that were examined for data availability, twenty-five indicators were
mapped. Data sources and supporting information for 32 indicators that had some form of data
available that could be examined for mapping are presented in Appendix C (Data Sources and
Supporting Information for Indicators Evaluated for Mapping).
We identified various types of data gaps in the search for data to represent our vulnerability
indicators at the national scale. In some cases, additional assessment of an indicator suggested
that there were too many obstacles to nationwide mapping at the present time. Because one rule
of thumb for this project was to identify those vulnerability indicators that could be readily
mapped, we did not consider indicators that appeared to be mappable but only with extensive
data processing efforts. The extent of the data gaps that affected the production of maps differed
from one indicator to another, and prohibited production of maps for some indicators. In other
cases the problems were minor and maps could be produced (with a few accompanying caveats).
The data gaps for this project could typically be placed into one of the three categories shown in
Table 9.
Table 9. Data Gaps
Data
Availability
Problem
Data Sets
Without
National
Coverage
Description of the
Problem
National data collection
is incomplete or
indicator is location-
specific.
Example Indicators
Population
Susceptible to Flood
Risk (#209)
Number of Dry
Periods in
Grassland/Shrubland
Streams and Rivers
(#190)
Specific Data Availability Problem
At time of inquiry, GIS-compatible digital
flood plain data from FEMA were only
available for certain parts of the country.
Heinz Center data identifies grassland and
shrubland watershed areas for Western
ecoregions only.
Do Not Cite or Quote
Page 69
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Data
Availability
Problem
Non-uniform
Spatial
Distribution
of Data
Temporal
Gaps
Description of the
Problem
Data are not distributed
homogeneously across
the country (therefore,
number of data points
within each HUC varies).
Lack of historical data
(which are needed as a
baseline) or time-
sensitive data which
must be updated
frequently.
Example Indicators
Acid Neutralizing
Capacity (#1)
Wetland Loss (#325)
Water Availability:
Net Streamflow
Availability per Capita
(#623)
Specific Data Availability Problem
EPA Wadeable Streams Assessment data
were collected at 1,601 sites. However, the
number of sites within HUC-4 units ranged
between 0 and 93 sites.
Historical data on the extent of wetlands is
not available.
Variables that this indicator depends on
(streamflow, water withdrawals, and
population) are all time-sensitive. Indicator
maps are not useful if recent data are not
available.
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
Mapped indicators typically used nationally recognized data sets or data sets created by national
agencies, such as EPA, USGS, and NOAA. While these data sets are comprehensive in nature
and cover the entire country, they still have data gaps as well as data quality issues. Nevertheless,
the data issues associated with the mapped indicators were either resolved or considered minor
enough that a map would still provide useful information for a vulnerability assessment. Minor
data issues were carefully documented for the mapped indicators.
B. Creation of Example Maps
We evaluated for mapping purposes 32 indicators for which national data had been collected.
Twenty-five indicators were considered to be mappable (Table 10). Six of the remaining
indicators were not mapped for this project due to challenges with acquiring data or representing
the source data spatially. One of these indicators was mappable, but had substantial gaps in
coverage that limited our ability to assess relative vulnerability at a national scale.
Table 10. List of Mapped Vulnerability Indicators
Indicator
(See Appendix Bfor definitions)
Acid Neutralizing Capacity (ANC) (#1)
At-Risk Freshwater Plant Communities (#22) 1
At-Risk Native Freshwater Species (#24) 1
Coastal Vulnerability Index (CVI) (#51) 2
Erosion Rate (#348)
Groundwater Reliance (#125)
Herbicide Concentrations in Streams (#367) l' 3
Herbicides in Groundwater (#373) *' 3
Insecticide Concentrations in Streams (#369) l' 3
Literature Source
(See Appendix A for full citations)
USEPA, 2006b.
Heinz Center, 2008.
Heinz Center, 2008.
Day et al., 2005.
Murdoch etal., 2000.
Hurdetal., 1998.
USGS, 1999.
USGS, 1999.
USGS, 1999.
Do Not Cite or Quote
Page 70
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
(See Appendix Bfor definitions)
Insecticides in Groundwater (#374) l' 3
Instream Use/Total Streamflow (#351)
Macroinvertebrate Index of Bi otic Condition (#460)1
Macroinvertebrate Observed/Expected (O/E) Ratio ofTaxa Loss
(#461)
Meteorological drought indices (#165) 2
Organochlorines in Bed Sediment (#371)*' 3
Pesticide Toxicity Index (#364)
Precipitation Elasticity of Streamflow (#437)
Ratio of Reservoir Storage to Mean Annual Runoff (#449)*' 3
Ratio of Snow to Total Precipitation (#218) 2
Ratio of Water Withdrawals to Annual Streamflow (#219) 3
Runoff Variability (#453)
Stream Habitat Quality (#284) 1
Total Use / Total Streamflow (#352)
Water Availability: Net Streamflow per Capita (#623) 1'3
Wetland and freshwater species at risk (number of species) (#326) 1
Literature Source
(See Appendix A for full citations)
USGS, 1999.
Meyer etal., 1999.
USEPA, 2006b.
USEPA, 2006b.
National Assessment Synthesis Team, 2000a.
USGS, 1999.
Gilliom etal., 2006.
Sankarasubramanian et al., 2001.
Lettenmaier et al., 2008.
Lettenmaier et al., 2008.
Hurd etal., 1998.
Lettenmaier et al., 2008.
Heinz Center, 2008.
Meyer et al., 1999
Hurd etal., 1998.
Hurd etal., 1998.
1930 Indicator definition changed based on available data.
1931 2 Indicator not defined in information source. Definition obtained from primary literature cited in the information
1932 source or new definition created based on available data.
1933 3 Indicator name changed to more appropriately match its definition or the available data.
1934
1935 The software we used for creating the maps for the 25 indicators was ArcMap 9.2 (© 1999-2006
1936 ESRI). For most indicators, data were available either in a GIS format, such as shapefiles, or in
1937 tabular form. In some cases, we processed tabular data in Microsoft Excel 2002 or Microsoft
1938 Access 2002 prior to importing into ArcMap. In other cases, we manipulated these data and
1939 calculated summary statistics directly in ArcMap. We used ArcMap to overlay different data
1940 sets, and we ultimately overlaid all data sets with HUC-4 boundaries. The data layer for such
1941 boundaries was obtained from the USGS.
1942
1943 For illustrative purposes, we had to choose a spatial unit of analysis. We chose to use USGS
1944 hydrologic units at the 4-digit scale here, for three practical reasons. First, USGS hydrologic
1945 units provide complete, continuous coverage of the continental U.S., which we established early
1946 on as a requirement of this project. Second, hydrologic units are usually synonymous with
1947 watersheds. Using a spatial unit with an inherent link to existing hydrography seems appropriate
1948 for a project that is evaluating indicators of vulnerability for drinking water and aquatic
1949 ecosystems. HUCs are frequently used by USGS and other agencies to monitor water-related
1950 phenomena across the country. Finally, 4-digit HUCs were chosen because they balance the need
1951 to convey interpretable regional patterns with the objective of providing detailed local
Do Not Cite or Quote
Page 71
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
1952 information. In other words, in our judgment, they do not over-generalize regional patterns and
1953 they do not over-extend the underlying data by providing more local resolution than is
1954 warranted. However, we reiterate that the maps we show are to illustrate the various issues we
1955 discuss, and we are not advocating any particular spatial aggregation as a matter of best practice.
1956 Alternative spatial frameworks or resolutions of course exist, and we discuss the implications for
1957 mapping of using such alternatives in more detail in sub-section E (Spatial Aggregation) below.
1958
1959 We aggregated or dis-aggregated the data, depending on their native scale (e.g., state-level data
1960 [where there is one data value provided for each state] vs. point data), to obtain a single value of
1961 the indicator for each HUC-4 watershed. Using Symbology, we assigned different colors or gray
1962 shades to represent the HUC-4 watersheds in different vulnerability categories on each indicator
1963 map. The detailed step-by-step methodology for each indicator is documented in Appendix E
1964 (Mapping Methodology).
1965
1966 We produced 25 complete example maps by HUC-4 watershed (see Appendix F). In addition, we
1967 produced an incomplete map for one indicator for which data suitable for mapping were
1968 available for portions of the country. However, substantial gaps in national coverage limit the
1969 ability to assess the relative vulnerability of ecosystems to environmental change at a national
1970 scale using this indicator. The remaining five indicators were not mapped for this project due to
1971 challenges with acquiring data or representing the source data spatially. These issues are
1972 discussed in detail below.
1973
1974 The mapped indicators fall into five categories established during the evaluation of the literature
1975 (see Section III). The categories (with number of indicators mapped shown in parentheses) are:
1976 chemical (7); ecological (6); hydrological (8); soil (1); socioeconomic (3). The indicators we
1977 mapped are not distributed evenly across these categories. For example, we mapped few
1978 socioeconomic and soil indicators.
1979
1980 Assuming that vulnerability can be inferred from metric values that were at the high (or low,
1981 depending on the indicator) end of the range of mapped values, regional differences in relative
1982 vulnerability were apparent for some of the mapped indicators. For example, the map for the
1983 indicator Meteorological Drought Indices (#165) displays high vulnerability in the Western
1984 United States, an area that has historically been exposed to prolonged drought. The map also
1985 shows high vulnerability for the Southeastern U.S., an area that has experienced a severe drought
1986 in recent years.
1987
1988 In some cases, there are no strong regional patterns. For example, the map for Stream Habitat
1989 Quality (#284) displays a spatially heterogeneous pattern, with no particular portion of the
1990 country strongly distinguished from any other.
1991
1992 Regions for which a single indicator might suggest greater vulnerability may not appear as
1993 vulnerable across a full suite of indicators. An examination of the full set of maps by HUC-4
1994 watershed in Appendix F suggests determining overall water quality- and aquatic ecosystem -
1995 related vulnerability across all of these dimensions may be complicated. Appendix G contains
1996 detailed descriptions of each of the 25 maps created for the mappable indicators. We return to the
1997 issue of combining indicators in more detail in Section VII below.
Do Not Cite or Quote Page 72
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
1998 C
1999 To create a national map illustrating an indicator of vulnerability, it is necessary to aggregate
2000 data collected at discrete locations and calculate summary statistics that describe conditions
2001 across a larger area. Examples of such statistics may include the mean value of an indicator or
2002 the percentage of sites that exceed a threshold value. In many cases, this aggregation process
2003 results in a slightly different metric. For example, Acid Neutralizing Capacity is reported in
2004 milliequivalents/L at the site scale. However, an aggregate statistic that can be calculated, and is
2005 both referred to in EPA's Wadeable Streams Assessment report and mapped for this report, is the
2006 percentage of sites with ANC less than 100 milliequivalents/L. When developing maps using
2007 aggregated metrics, it is important for both the producers and consumers of maps to understand
2008 how the underlying data and the aggregation methods may affect the validity of objective
2009 thresholds and the patterns illustrated in the final map. In the above example, the threshold of
2010 100 milliequivalents/L is a relevant threshold at the scale of an individual site. However, no
2011 objective thresholds are defined for the range of aggregated percentage values calculated for
2012 each HUC. Appendix K includes an evaluation of the effects of aggregation on the validity of
2013 theoretical breakpoints for each of the mapped indicators. These issues of aggregation
2014 underscore the concept that a single set of data can be used to produce many different maps. The
2015 following sections discuss additional factors to be considered when aggregating data.
2016 a. Local Variation
2017 Measurements at individual sample sites are affected by local factors such as land use, the
2018 presence of an industrial facility, an urban center, a protected region (e.g. a National Park), or
2019 other features that exist in a heterogeneous landscape. Within a large area (like a HUC-4 unit)
2020 that contains a wide variety of these local factors, measurements collected at individual sites may
2021 vary substantially. When a group of values within such an area are aggregated into a single
2022 value, local variation can be masked. Understanding the degree of local variation is an important
2023 component of interpreting vulnerability. For this reason, it may be necessary to simultaneously
2024 consider maps that illustrate the vulnerability metric and the variation in raw data values present
2025 within each spatial unit.
2026 b. Extent of spatial units (HUC Levels)
2027 Aggregation of individual local measurements into a single metric frequently involves the
2028 extrapolation of information. Extrapolation may be appropriate in areas where sampling density
2029 is large enough to accurately describe the conditions, and that the extent of the local
2030 measurements coincides with the extent of the larger areal unit used to aggregate data. However,
2031 extrapolation may also result in the masking of low data density in cases where the extent of the
2032 aggregate unit is significantly different from the extent of the underlying data. The producers of
2033 maps must be sensitive to the limits of aggregation (and subsequent extrapolation) when
2034 choosing a spatial framework to represent a data source comprised of local measurements.
2035
2036 For example purposes, we rely here on 4-digit HUCs to illustrate patterns of vulnerability - we
2037 apply it consistently to compare across indicators. For some indicators, however, aggregation of
2038 data into this framework may mask low data density. Figure 7 illustrates this issue using 3
2039 different scales of HUC units and the same underlying data set. The visual contrast between the
Do Not Cite or Quote Page 73
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
2040 top and bottom maps demonstrates how low data density can be masked through aggregation into
2041 larger spatial units.
2042 All of the indicators we selected for mapping were chosen based on their ability to provide
2043 information on the relative vulnerability of water quality and aquatic ecosystems. As
2044 environmental measurements, the data collected and used for each indicator has an inherent level
2045 of uncertainty and error associated with it. Selecting a particular unit for presenting information
2046 in a set of maps is useful for making comparisons across the set. However, the data collected for
2047 the indicators were not available at consistent scales across the set of indicators. The data for
2048 most of the indicators was thus altered to present it at a consistent scale. Although manipulating
2049 the data changes the accuracy of the information, the manipulations help make the information
2050 presented more useful. For the most part, data manipulation required either a scaling up or down
2051 of data or transformation of the data from different geographic boundaries.
2052
2053 Data needing to be scaled up included point data. In all cases, the sample data used to calculate
2054 metrics for these indicators is not distributed homogeneously. As a result, dissimilar amounts of
2055 data are available within the HUC-4 unit boundaries. In cases where there are few sample points
2056 within a HUC-4 boundary, individual sites have a greater influence on the metric value that is
2057 calculated.
2058
2059 Data presented at the state level needed to be scaled down or transformed to match the HUC-4
2060 geographic boundary. Transforming the data from a state-based representation to a HUC-4
2061 representation requires an assumption that the distribution of the indicator is uniform within each
2062 state. Although this assumption is unlikely to be accurate, it allows for area-weighted metrics to
2063 be calculated for HUC-4 units that intersect more than one state.
2064
2065 Coastal data presented a unique challenge in mapping. As a watershed geographic unit, HUC-4
2066 has limited or no coverage for coastal and nearshore area data. This makes aggregation for the
2067 purposes of reporting at the HUC-4 scale problematic. To address this issue, we developed a
2068 special reporting unit for one indicator, the Coastal Vulnerability Index (#51).
2069
2070 Although necessary for creating useful and comparable maps, data manipulations change the
2071 quality of the data presented through assumptions about coverage and the representativeness of
2072 the data to nearby geographic areas. In most cases, data manipulations are likely to yield greater
2073 error and uncertainly than the original data. However, problems associate with data manipulation
2074 are likely to be more important for some indicators than others. For example, an indicator based
2075 on fine-scale data within a HUC-4 boundary will likely present a more accurate picture of
2076 relative regional vulnerability than an indicator based on transformed state-level data.
Do Not Cite or Quote Page 74
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
2077 Figure 7. Aggregation, Precision, Coverage, and Data Density
2078 The following maps display the Stream Habitat Quality (#284) indicator at various scales of
2079 HUC units, illustrating how low data density can be masked through aggregation into larger
2080 spatial units.
Average Rapid Assessment Score
40 - 109
^] 110-125
^B 126~ 136
^B 137~ 14T
^H 148 - 190
4-digit HUC units
Bioassessmeut Sample Location
8-digit HUC units
12-digit HUC units
2081
0 200 400 600 800 1,000 Miles
I I I I I I
Do Not Cite or Quote
Page 75
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
2082 c. Alternate Spatial Frameworks
2083 The selection of the spatial framework used to evaluate geographically-based data can have a
2084 significant influence on the graphical display of spatial information and for the assessment and
2085 management of resources (Omernik and Griffith, 1991). In some cases, different units of analysis
2086 can result in maps that provide difference perceptions using the same set of underlying data. Two
2087 spatial frameworks, watersheds and ecoregions, are often associated with ecosystem
2088 management. Each of these frameworks has advantages, and the tradeoffs between the two
2089 systems reinforce the concept that there is no single best spatial framework for displaying
2090 indicators of water quality and aquatic ecosystem condition or vulnerability.
2091 1. Watersheds (and hydrologic units)
2092 Watersheds are often advocated as the appropriate unit for ecosystem management because they
2093 encompass the area of land that influences a connected system of water bodies (Montgomery et
2094 al. 1995, U.S. EPA 1995). To address the practical need for a system of management units that
2095 serve as a standardized base for inventorying hydrologic data, the US Geological Survey
2096 delineated hydrologic units. These units are commonly identified by their hydrologic unit codes
2097 (HUCs) (Seaber et al., 1987). The term "HUC" is often used to describe the hydrologic unit, not
2098 just the unit code). HUCs are assigned at several hierarchical spatial scales. The HUC-4 units (n
2099 = 204) used in this study have a mean area of 38,542 km2.
2100
2101 It is noteworthy that many HUCs are true watersheds, while others are combinations of multiple
2102 smaller watersheds or segments of a larger watershed. HUCs provide non-overlapping,
2103 continuous coverage of a given area, and are typically used in place of true watersheds for
2104 mapping environmental data.
2105 2. Ecoregions
2106 Ecoregions are alternative spatial units, introduced by Omernik (1987), that are specifically
2107 designed to be internally homogeneous with regard to factors that affect water quality, such as
2108 vegetation, soils, land forms, and land use. Similar to HUCs, ecoregions are designated at several
2109 hierarchical spatial scales. The size of individual ecoregions varies more than individual HUCs.
2110 For example, the 87 ecoregions at the Level 3 scale range in size from 649 to 357,000 sq. km.
2111
2112 The shortcoming of ecoregions is that they rarely encompass a single hydrologically connected
2113 area, making it difficult to identify the location(s) where cumulative stresses will be felt.
2114
2115 Figure 8 illustrates differences resulting from the use of different spatial frameworks. Although
2116 the national spatial patterns are similar, there are local differences that may influence
2117 vulnerability interpretations. Specifically, differences between the maps are most evident in the
2118 western United States - particularly within the Rocky Mountains - and in northern Wisconsin.
2119 These differences are reasonable, given the basis for delineating individual areas within each of
2120 these frameworks. HUCs, which are based loosely on watershed boundaries, tend to integrate a
2121 wider range of physical/topographical characteristics than ecoregions. These local physical
2122 characteristics may have a significant influence on the ratio of snow to total precipitation at any
2123 one point, resulting in a wide range of values within a HUC. Ecoregions, on the other hand, are
2124 specifically intended to describe regions with physical/topographical similarities. Thus, one
Do Not Cite or Quote Page 76
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
2125 would expect that ecoregions would contain less within-unit variation for Indicator #218. Maps
2126 of the 25 mappable indicators by ecoregion are presented in Appendix H. Appendix I contains
2127 detailed descriptions of each of these maps. From a visual comparison of these maps with the
2128 HUC maps presented in Appendix F, it is evident that the choice of similarly sized spatial units
2129 (i.e., HUC4 vs. Ecoregion Level 3) has little effect on our results at the national scale.
2130 3. Coastal Areas
2131 Coastal areas are worthy of focus in national scale vulnerability assessments because they are of
2132 great national importance and pose unique challenges. Coastal areas may be more prone to the
2133 effects of climate change, but the limited geographic extent of coastal areas necessitates the use
2134 of a different analysis framework. For example, the indicator Coastal Vulnerability Index (#51)
2135 uses data available from a USGS database. The data are limited to only coastal and nearshore
2136 areas. Although this indicator provides complete coverage of coastal areas, aggregation into
2137 HUC-4 units or ecoregions would not provide meaningful results. To address this issue, a set of
2138 special reporting units for coastal areas was developed for this indicator. Each unit extends
2139 approximately 20 miles inland and includes approximately 150 miles of coastline (Figure 9).
2140
2141 A/,
2142 It is common to symbolize numerical data using chloropleth maps, which use a range of colors
2143 that correspond to the underlying data values. Determining how each color is assigned to the
2144 range of data values is classic cartographic challenge that applies to most any mapping project,
2145 this study included. For numerical data, the methods used to delineate breaks between data
2146 classes can affect the spatial patterns conveyed in a map, and the subsequent interpretation of
2147 those data. Thus, care must be taken in the development of maps based on numerical data,
2148 especially if the resulting spatial patterns may be used to develop policy.
2149
2150 Figure 10 illustrates how a single set of data can be used to create alternate maps simply by
2151 altering the number of data classes and the breaks used to distinguish between individual data
2152 classes.
Do Not Cite or Quote Page 77
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
2153 Figure 8. Data Represented by Different Spatial Frameworks
2154 The following maps display the Ratio of Snow to Precipitation (#218) indicator using 4-digit
2155 HUC units and Omernik's (1987) ecoregions, illustrating how the same underlying data appear
2156 different when using different spatial frameworks.
Ratio of Total Suo«fall to Total Precipitation
| |0%- 5%
| | 5.01%-10%
^| 10.01%-15%
^H 15.01%-20%
^H 20.01%-100%
4-cligit HUC unil
Ecoregions
2157
Do Not Cite or Quote
Page 78
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
2158
2159
2160
Figure 9. Spatial Framework for Coastal Zone Indicators
The following map displays the Coastal Vulnerability Index (#51), a coastal indicator, for which a set of special reporting units for
coastal areas was developed. Each coastal unit extends 20 miles inland and includes approximately 150 miles of coastline.
Coastal Vulnerability Index
Low Vulnerability
Medium Vulnerability
High Vulnerability
0 100 200 300' 400 500 Miles
I I I I I I
2161
Do Not Cite or Quote
Page 79
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
2162
2163
2164
2165
Figure 10. Different Breaks to Distinguish Data Classes
The following map displays the Stream Habitat Quality (#284) indicator, illustrating how the
same underlying data appear different when displayed using three different data breaks
(quantiles, equal intervals^ and natural breaks or jenks).
3 Quantiles
Average Rapid Assessment Score
^H 40-119
^H 120 - 138
139 - 190
2166
Equal Interval
Average Rapid Assessment Score
^| 40 - 50
^| 51 - 75
•• 76 - 100
^M 101 - 125
126- 150
151- 175
176 - 200
Natural Breaks (Jenks)
5 Classes
Average Rapid Assessment Score
! I 40 - 97
98 - 120
^H 121 - 136
^| 137- 153
^H 154- 190
Do Not Cite or Quote
Page 80
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
216? VII. Challenges Part IV: Combining Indicators
2168 A.
2169 Exposure to future stresses associated with external stressors such as climate and land-use
2170 change is likely to vary spatially. Scenarios derived from climate models can be used to map
2171 changes in exposure across the plausible range of future changes. A more comprehensive
2172 evaluation of future stresses could directly incorporate such scenarios in a vulnerability
2173 indicator-based assessment. Figure 11 displays an approach for combining indicators identified
2174 in this report with other variables. This approach allows the identification of locations that are
2175 both vulnerable to stress and are likely to experience additional stress in the future. Four
2176 indicators that are related to potential water shortages are presented in the context of simulated
2177 changes in temperature and precipitation derived from the IPCC 4th Assessment Report (TPCC,
2178 2007b) and population derived from EPA's Integrated Climate and Land Use Scenarios (ICLUS)
2179 project (USEPA, 2009d). Increasing temperature and population and decreasing precipitation all
2180 tend to increase the likelihood of water shortages. These plots are examples meant to illustrate
2181 how one might go about highlighting regions where we might see a convergence between an
2182 already stressed water supply system, a warmer, drier climate, and significant population growth.
2183
2184 While all of the indicators in Figure 11 relate to water supply, they deal with different aspects of
2185 vulnerability. For example, Precipitation Elasticity of Streamflow (#437) is based only on natural
2186 variation in water availability, whereas Groundwater Reliance (#125), Ratio of Withdrawals to
2187 Streamflow (#219), and Water Availability: Net Streamflow per Capita (#623) either directly
2188 incorporate current rates of water use or infer it through population. These plots illustrate how
2189 high water withdrawals in some regions may be unsustainable under the chosen temperature and
2190 precipitation scenario, or how locations that have low water availability per capita might also be
2191 places where we expect to see the greatest population increases in the future. In general, under
2192 the scenarios used here, current sensitivity and future exposure tend to co-vary, and thus the
2193 places that are vulnerable now are likely to become more vulnerable in the future.
2194
Do Not Cite or Quote Page 81
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
2195 Figure 11. Current and Future Vulnerability to Water Shortages
2196 The following plots displays values of some example indicators with a sample scenario of
2197 temperature and precipitation (based on the Bl greenhouse gas storyline) drawn from the IPCC
2198 Summary for Policymakers (IPCC, 2007b) and a population scenario from the Integrated
2199 Climate and Land Use Scenarios (ICL US) project. All variables are scaled as changes over a
2200 100 year period from 2000 to 2100. Each point represents a single HUC-4 and is shaded
2201 according to values of the indicator.
2202
2203 A. Groundwater Reliance (#125) (white, 0-10%; grey, 11-60%; black, 61-100%).
20
2204
300
Do Not Cite or Quote
Page 82
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
2205 B. Ratio of Withdrawals to Stream/low (#219) (white, 0-0.11; grey, 0.12-0.75; red, 0.75-59).
2206
2207
2208
20
300
C. Precipitation Elasticity of Stream/low (#437) (white, 0.43-1.59; grey, 1.60-2.06; black, 2.07-
2.96).
20
-200
2209
?te'
-100
**"*
300
Do Afo/ Cite or Quote
Page 83
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
2210 D. Net Stream/low per Capita (#623) (white, 8,493-1,779,536; grey, 888-8,493; black, 0-877).
20
300
2211
2212 B. Composites of Vulnerability Indicators
2213 Because individual indicators provide information on limited dimensions of aquatic ecosystem
2214 and water quality vulnerability, effective management planning would likely require that these
2215 dimensions be integrated into a more holistic perspective on vulnerability. Assuming issues
2216 specific to individual indicators can be resolved, there are several possible quantitative methods
2217 for integrating multiple indicators.
2218 a. Creating a Composite Map
2219 Mapped indicators could, potentially, be overlayed into a composite map, such that the averages
2220 of all indicator values for each of the HUC units are represented on a single map. This is
2221 challenging, however, for a number of reasons. One major reason is that the distinction between
2222 relative and real (i.e., functionally significant) differences in vulnerability, while not necessarily
2223 as critical for interpretation of individual indicator maps, is extremely important for the
2224 construction of a composite vulnerability map. For example, if the range of values for an
2225 indicator only reflect one category of vulnerability (e.g., very high vulnerability), differences in
2226 relative vulnerability may be functionally insignificant. If this type of indicator is given equal
2227 importance in a composite score to one whose values span a functionally significant range, the
2228 composite score will be inaccurate. As a consequence, the vulnerability of individual locations
2229 may be under- or over-estimated, depending on the relative frequency of high vulnerability
2230 values from these two classes.
2231
Do Not Cite or Quote
Page 84
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
2232 Another way to aggregate indicators could be by identifying geographic units where further
2233 stresses (including climate change) will cause the most harm across all system dimensions (e.g.,
2234 see Lin and Morefield, 2010). This can be done as follows:
2235 o Assign numeric scores to the vulnerability categories (e.g., 3 for highest, 2 for medium,
2236 and 1 for lowest). Sum the scores across all indicators.
2237 o For each geographic unit, calculate the percentage of indicators that are in the highest
2238 vulnerability category.
2239
2240 Once any technical deficiencies and data gaps have been addressed through data collection
2241 efforts, construction of a composite vulnerability map should consider the following:
2242
2243 • The relative importance of system dimensions. The importance of individual indicators is
2244 dependent on management objectives and the degree to which indicators are redundant
2245 with one another.
2246 • Range of indicator values. Only indicators whose values span functionally significant
2247 ranges should be used for a composite vulnerability map. This will lead to a more
2248 accurate representation of relative vulnerability.
2249 • How an integrated vulnerability rating will translate into management or adaptation
2250 efforts- Locations with high integrated vulnerability may either be moderately vulnerable
2251 for most attributes, or highly vulnerable for a few attributes. While both of these
2252 scenarios point to the need for planning, the specific suite of relevant strategies would
2253 differ. Thus, the production of multiple visualization tools may often be a helpful
2254 exercise.
2255 b. Characterizing vulnerability profiles
2256 The aim of this type of integrative procedure is to identify commonalities in the types of
2257 vulnerabilities among regions. A vulnerability profile for a given location can be defined as the
2258 set of values for all the vulnerability indicators. The proposed analysis allows watersheds with
2259 similar vulnerability profiles to be identified, and might be useful in the transfer of successful
2260 management or adaptation strategies from one location to another. Specifically, if a selected
2261 watershed is vulnerable in certain ways and in need of an adaptation strategy, other locations
2262 with similar vulnerability profiles could be identified. Successful adaptation strategies in those
2263 other locations could then be assessed for their applicability at the selected watershed.
2264
2265 Similarities in vulnerability profiles among locations can be summarized numerically through
2266 multivariate statistical analyses, such as Principal Components Analysis (PCA), which is a useful
2267 method for finding patterns in data. PCA is used to consolidate the information in a large number
2268 of variables into a smaller number of artificial variables (called principal components) that will
2269 account for most of the variability in the original variables. The first component extracted in a
2270 PCA accounts for the greatest amount of total variance in the original variables, and the second
2271 and subsequent components account for progressively less variance.
2272
2273 The principal components (PCs) are described in terms of loadings of the original variables. A
2274 PC may be heavily loaded on at least one variable, and usually on more than one. A high loading
2275 indicates that the PC is strongly related to that variable (either negatively or positively depending
2276 upon the sign of the loading). Variables for which a PC is heavily loaded are correlated with each
Do Not Cite or Quote Page 85
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
other, creating clusters of related variables that should be interpretable from a conceptual
standpoint. The PCs themselves, however, are uncorrelated with one another. One benefit of
conducting a PCA for this study is that reducing the full set of indicators to its principal
components helps to avoid overemphasis on system properties that are represented by multiple
similar indicators.
As an example, we conducted a PCA on 24 of the 25 mapped indicators (we excluded the
Coastal Vulnerability Index (#51) because of its unique spatial units). We normalized indicators
with non-normal frequency distributions with log or square root transformations. We inverted the
scales of some indicators so that high vulnerability was always represented by high values of the
indicator. We used the correlation matrix of these standardized variables for the PCA. When no
data were available for an indicator, the HUC was assigned the median value for that indicator.
We rotated the PCA (Varimax) and specified a maximum of six principal components - these six
cumulatively account for about 57% of the total variance, with 35 % coming from the first three.
Table 11 shows the six PCs generated in the PCA analysis. These PCs help demonstrate which
types of processes or environmental factors are driving a large part of the variability in the data.
PCI is heavily loaded on indicators related to at-risk species, which are negatively correlated
with the ratio of snow to total precipitation. PC2 is correlated with variables indicative of
streamflow availability and usage. PC3 represents pesticides in surface water. PC4 is loaded on
indicators related to macroinvertebrates and stream habitat quality. For PCS, the most heavily
loaded indicator is meteorological drought indices, which is moderately correlated with at-risk
freshwater plant communities. Finally, PC6 is loaded on herbicides in groundwater, but not
pesticides in groundwater.
Table 11. Principal Components Loadings for the
PCA Analysis
Twenty Four Indicators Included in the
Indicator
Acid neutralizing capacity (#1)
At-risk freshwater plant communities (#22)
At-risk native freshwater species (#24)
Groundwater reliance (#125)
Meteorological drought indices (#165)
Ratio of snow to total precipitation (#218)
Ratio water withdrawal to annual streamflow
(#219)
Stream habitat quality (#284)
Wetland species at risk (#326)
Erosion rate (#348)
Instream use/total streamflow (#351)
Total use/total streamflow (#352)
Pesticide toxicity index (#364)
Herbicide concentrations in streams (#367)
Insecticide concentrations in streams (#369)
PCI
0.166
0.401
0.863
0.087
0.006
-0.774
-0.071
0.092
0.789
0.387
0.132
0.017
0.082
0.078
0.070
PC2
-0.367
0.220
0.167
0.196
0.182
0.033
0.873
-0.018
-0.102
-0.056
0.262
0.753
0.009
-0.112
0.025
PC3
-0.231
-0.007
0.068
0.242
-0.138
-0.167
-0.089
0.170
0.017
-0.058
0.144
0.048
0.889
0.769
0.870
PC4
-0.071
0.153
0.051
0.291
-0.038
-0.193
0.036
0.687
0.026
-0.076
-0.104
0.126
-0.027
0.111
-0.033
PCS
-0.233
0.604
0.149
-0.033
0.771
0.120
0.035
0.196
-0.204
0.131
0.005
-0.052
-0.003
-0.028
-0.020
PC6
-0.265
0.090
0.117
-0.313
0.019
0.300
0.056
0.056
0.200
0.504
-0.456
-0.211
-0.041
-0.112
0.033
Do Not Cite or Quote
Page 86
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
Organochlorines in bed sediment (#371)
Herbicides in groundwater (#373)
Insecticides in groundwater (#374)
Precipitation elasticity of streamflow (#437)
Ratio of reservoir storage to mean annual runoff
(#449)
Runoff (variability) (#453)
Macroinvertebrate index of biotic condition
(#460)
Macroinvertebrate observed/expected (#461)
Water availability: streamflow per capita (#623)
Proportion of variability explained
PCI
0.092
0.018
0.191
0.628
-0.117
0.160
0.051
-0.156
-0.150
0.120
PC2
0.089
0.212
0.080
-0.073
-0.250
0.504
0.074
0.030
0.839
0.117
PCS
0.515
0.160
0.078
0.156
-0.090
0.036
-0.043
0.080
-0.127
0.113
PC4
0.016
-0.009
-0.139
0.207
0.074
-0.056
0.845
-0.754
0.002
0.085
PCS
-0.358
-0.239
-0.537
0.153
-0.151
0.256
0.007
0.066
-0.009
0.073
PC6
0.109
0.721
0.355
-0.107
0.110
0.137
-0.112
-0.055
-0.005
0.065
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
The map in Figure 12 is another way of using and displaying the results of the PCA. This map
shows the similarity of an example focal watershed (shown in blue) to watersheds across the
U.S. We defined the similarity of two watersheds as the weighted Euclidean distance (Dw)
among the values of the first six principal components:
=1
where x; and y; are the values of component i for the two watersheds, and w; is the weight for
component i, which is defined as the proportion of the total variance in the entire dataset
explained by that component. This approach is similar to the methods used by Tran et al. (2006).
As discussed above, because this kind of analysis and map allows watersheds with similar
vulnerability profiles to be identified, it might be useful in the transfer of successful adaptation
strategies from one location to another. Specifically, the map could help to identify locations
with the most similar multi-dimensional vulnerability profiles to that of a selected focal
watershed in need of adaptation strategies. Successful adaptation strategies in those other
locations could then be assessed for their applicability at the focal watershed.
While relative similarity could identify the closest matches to the focal watershed, its mean
absolute similarity to all other locations would be a measure of its uniqueness. The similarity of
all pairwise combinations of watersheds could be cataloged in a vulnerability similarity matrix to
expand the applicability of this approach. Such a matrix would include every watershed on the
Do Not Cite or Quote
Page 87
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
2326 Figure 12. Vulnerability Profile Similarity
2327 The following map displays the results of the PCA conducted on 24 of the 25 mapped indicators. It shows the similarity of the focal
2328 HUC watershed (blue) to the remaining 203 watersheds.
\ | States
^B Fccjl HUC
Vulnerability Profile Similarity
• Similar
2329
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
Page 88
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
2330 horizontal axis, and these same watersheds on the vertical axis. Each central cell of the matrix
2331 would contain a value that documents (according to the formula above) the similarity of the two
2332 watersheds defined by that cell. In addition, the vulnerability profile approach could be further
2333 refined by applying weights to indicators to account for differences in accuracy or relevance to
2334 climate change or other stressors of interest.
2335 VIII. Summary and Recommendations
2336 This report investigates the issues, challenges, and lessons associated with identifying,
2337 calculating, and mapping indicators of the relative vulnerability of watersheds across the United
2338 States to the potential adverse impacts of external stresses such as long-term climate and land-
2339 use change. It is our hope that this report will be a useful building block for future work on
2340 multi-stressor global change vulnerability assessments.
2341
2342 It is important to clarify here that this report does not attempt any kind of direct evaluation of the
2343 potential impacts of climate change or other global change stressors on ecosystems and
2344 watersheds. Instead, it deals only with the question of how to estimate the impacts of current
2345 stressors. We argue that a systematic evaluation of the impacts of existing stressors is a key input
2346 to any comprehensive climate change vulnerability assessment, as the impacts of climate change
2347 will be expressed via often complex interaction with such stressors - i.e., through their potential
2348 to reduce overall resilience, or increase overall sensitivity, to climate change. This argument is
2349 not new, and in fact it has been a staple of writing on climate change impacts, vulnerability, and
2350 adaptation, particularly of large assessments like those of the IPCC and U.S. Global Change
2351 Research Program. However, to date there has been relatively little exploration of the practical
2352 challenges associated with comprehensively assessing how the resilience of ecosystems and
2353 human systems in the face of global change may vary as a function of existing stresses and
2354 maladaptations.
2355
2356 A.
2357 Our approach in this report has two basic elements. First, we have collected, evaluated the
2358 quality of, processed, and aggregated a large quantity of data on water quality and aquatic
2359 ecosystem indicators across the nation that have been reported on in the ecological, hydrological,
2360 and management literature. Second, we have used this set of indicators as a testbed for
2361 identifying best practices, challenges, and gaps in ideas, methods, data, and tools for calculating
2362 and mapping vulnerability nationally.
2363
2364 Specifically, we conducted a literature search and compiled a comprehensive list of 623
2365 indicators of water quality or aquatic ecosystems, including those relating to ambient surface and
2366 groundwater quality, drinking water quality, ecosystem structure and function, individual
2367 species, and the provision of ecosystem services. This then formed the set of indicators for
2368 exploring a number of subsequent challenges. These challenges fall into four broad categories:
2369
2370 1. Challenges associated with identifying those indicators that speak specifically to
2371 vulnerability, as opposed to those reflecting simply a state or condition. In this context, we
2372 define vulnerability as adverse impacts accrued over time and associated with external
2373 stresses from, for example, climate or land-use change;
Do Not Cite or Quote Page 89
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
2374 2. Challenges associated with determining relative vulnerability using indicators, including
2375 interpreting gradients of indicator values, and, when possible, establishing important
2376 indicator thresholds that reflect abrupt or large changes in the vulnerability of water quality
2377 or aquatic ecosystems;
2378 3. Challenges associated with mapping these vulnerability indicators nationally, including data
2379 availability and spatial aggregation of the data;
2380 4. Challenges associated with combining and compositing indicators and developing multi-
2381 indicator indices of vulnerability.
2382
2383 For this work, we relied on published research and on studies by EPA, other federal agencies, the
2384 Heinz Center, the Pew Center, and a number of other sources, both for indicator definitions and
2385 for the data to support the mapping of indicators. Our intent was to examine what could be
2386 accomplished with existing indicators and datasets, and for the most part we did not attempt at
2387 this point to conceive of new indicators or collect new data. As part of this work, we developed a
2388 number of example maps, and we use some of these maps in this report for illustrative purposes.
2389 We hope that the lessons we learned while developing strategies for compiling and mapping
2390 national-level indicator datasets under this project would likely be useful for indicator-based
2391 vulnerability assessments in general. Here we summarize the main findings of the report,
2392 organized according to the four challenges listed above.
2393 a. Challenges Part I: Indicator Classification
2394 There is on ongoing debate in the literature on the meaning of vulnerability and the elements of
2395 which it is composed, particularly in the context of climate change. For the purposes of this
2396 report, we generally took as our starting point the IPCC definition, i.e., "The degree to which a
2397 system is susceptible to, or unable to cope with, adverse effects of climate change, including
2398 climate variability and extremes. Vulnerability is a function of the character, magnitude, and rate
2399 of climate variation to which a system is exposed, its sensitivity, and its adaptive capacity."
2400 (IPCC, 2007a.) Most of what we define as "vulnerability indicators" in this report primarily
2401 encompass sensitivity and exposure to environmental stresses, and we do not focus on adaptive
2402 capacity. The indicators we discuss relate generally to the vulnerability of aquatic ecosystems,
2403 ecosystem services, and drinking water supplies.
2404
2405 Our first challenge was to identify guidelines for classifying the comprehensive suite of 623
2406 indicators. The goal was to divide them into vulnerability indicators versus those indicators that
2407 merely measure the current state of a resource. The vulnerability indicators, at least in principle,
2408 could measure the degree to which the resource being considered (e.g., watershed, ecosystem,
2409 human population) is susceptible to, and unable to cope with, adverse effects of externally forced
2410 change. Such change potentially includes climate or any other global change stressor.
2411
2412 We determined that, in practical terms, the essence of a vulnerability indicator is that it should
2413 inherently include some kind of relative or value judgment, e.g., comparing one watershed to
2414 another, comparing it to some objectively defined threshold or possible state, or reporting on its
2415 change over time, as opposed to measuring water quality or ecological condition at a point in
2416 time without reference to anything else. Applying these criteria, we winnowed the original list of
2417 623 indicators down to 53, and in the report we discuss the degree to which indicators from this
Do Not Cite or Quote Page 90
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
2418 reduced set might reflect vulnerability of water quality and aquatic ecosystems to challenges
2419 from long-term global change stresses.
2420
2421 b. Challenges Part II: Determining Relative Vulnerability
2422 Determination of the relative vulnerability of a particular location using a given vulnerability
2423 indicator (or an index, if multiple indicators have been combined), can be accomplished by
2424 comparing the value of the indicator to a gradient of values measured at different locations.
2425 Alternatively, one can capitalize on objective vulnerability thresholds for some indicators. Such
2426 thresholds reflect abrupt or large changes in the vulnerability of water quality or aquatic
2427 ecosystems in response to a small change in a stressor. Such thresholds are most useful when
2428 they distinguish between acceptable and unacceptable conditions.
2429
2430 We searched for thresholds for our 53 vulnerability indicators from three different categories:
2431 human health-based thresholds, ecological thresholds, and sustainability thresholds. In the
2432 literature, we most often encountered the use of arbitrary cutoffs to separate relative vulnerability
2433 categories (e.g., high, medium, and low). We were only able to map objective thresholds for a
2434 small subset of the indicators, though in some cases we suggested modification of an indicator
2435 definition to facilitate the identification of thresholds. The lack of available functional break
2436 points for most indicators is to be expected. Many indicators respond to stress linearly or along a
2437 gradual gradient. For others, objective break points may be characterized through additional
2438 research, either through meta-analysis of previous research efforts or through new data collection
2439 and analysis. Future research may also yield additional insights into how break points for some
2440 indicators vary spatially (Link, 2005).
2441
2442 c. Challenges Part III: Mapping Vulnerability
2443 The effort to produce indicator maps for this report faced a number of classic cartographic
2444 challenges. Most of these challenges fell into the following two major categories: data
2445 availability and mappability, and spatial aggregation.
2446 1. Data and mappability
2447 Data availability and suitability were the most serious limitations in evaluating whether or not we
2448 could produce maps for the 53 vulnerability indicators. Issues we encountered included the
2449 following:
2450
2451 • Lack of national coverage
2452 • Varying scales of the data
2453 • Varying duration of the data records
2454 • Multiple datasets needed to be combined
2455 • A model needed to be run to generate the data for the indicator
2456 • The indicator was conceptual only, with no underlying dataset
2457 • Data collection was in progress
2458 • Data was too out of date
2459
Do Not Cite or Quote Page 91
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
2460 These data availability and suitability issues were sometimes readily apparent, but sometimes
2461 they emerged only after beginning the process of attempting to create maps. A major lesson we
2462 learned from this project was that it may often be impossible to establish mappability without
2463 beginning the process of manipulating and mapping the various datasets involved.
2464
2465 Overall, these data and mappability issues reduced the starting set of 53 vulnerability indicators
2466 to a set of 25 vulnerability indicators for which we were able to create example maps.
2467 2. Spatial Aggregation
2468 To create a national map for a given indicator of vulnerability, one must aggregate data collected
2469 at discrete locations and calculate summary statistics that describe conditions across a larger
2470 area, such as the mean value of an indicator or the percentage of sites that exceed a threshold
2471 value. As noted above, a major research gap is the lack of objective, functional thresholds
2472 between "vulnerable" and "not vulnerable" for most of the indicators we investigated. A
2473 complementary challenge is that, even if such functional breakpoints can be found, it may be
2474 difficult to aggregate in such a way that these breakpoints remain meaningful.
2475
2476 The major issues we encountered were the following:
2477 • Local variation and spatial heterogeneity in data collection sites;
2478 • The choice of spatial frameworks (e.g., watersheds, ecoregions, coasts);
2479 • The extent (resolution) of the spatial unit chosen.
2480
2481 As illustrated with a variety of example maps, these methodological choices can lead to very
2482 different results, and hence different conclusions about relative vulnerability in one location
2483 compared to another.
2484
2485 A systematic process for refining or re-defining indicators of vulnerability to account for the
2486 challenges summarized above is likely to be valuable. Such a process is presented in Figure 12.
2487 For example, the Acid Neutralizing Capacity (#1) indicator is defined as the ability of a stream to
2488 buffer acidic inputs from acid rain or acid mine drainage. This indicator can be refined to
2489 measure the percent of sites that with ANC less than 100 millequivalents/L to account for the
2490 aggregation challenge. In addition, indicators can be refined to more explicitly incorporate the
2491 exposure component of vulnerability. If elements of environmental change, such as temperature
2492 or precipitation, can be explicitly incorporated into the indicator, then future changes in this
2493 indicator can be modeled using predicted changes in the values of these elements. This
2494 strengthens the ties between the indicator and changes that may occur in the future, and
2495 facilitates the generation of more useful forecasts for decision-makers.
2496
2497 d. Challenges Part IV: Combining Indicators
2498 Ultimately, the value for global change assessments of a database of indicators, and their maps,
2499 rests in how they can be examined holistically. Such indicators and their maps can also be
2500 examined in combination with scenarios of changes in critical external stressors, such as climate
2501 and land use. We showed some simple examples of how one might use such scenario data to
2502 highlight locations around the country where, for example, we might see a convergence between
2503 an already stressed water supply system, a warmer, drier climate, and significant population
Do Not Cite or Quote Page 92
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
2504 growth. One of several more sophisticated approaches involves designing indicators that
2505 explicitly include a functional dependence on a stressor that is expected to change over time,
2506 such as temperature, precipitation, or population.
2507
2508 We also considered the challenges associated with compositing multiple indicators in some way
2509 and mapping the result. This brings up issues of determining the functional equivalency of the
2510 different levels of relative vulnerability measured by the very different indicators, with no
2511 absolute standard as an anchor point for weighting their contributions. Creation of a uniform
2512 scoring system (e.g., 1, for lowest, and 5 for highest, vulnerability) resolves the practical
2513 difficulties of mapping but not the conceptual ones of establishing the relative contribution of
2514 each indicator to overall vulnerability. Appendix K includes an evaluation of the effects of
2515 aggregation on the validity of theoretical breakpoints for each of the mapped indicators based on
2516 the process outlined in Figure 13.
2517
2518 A possible way forward is in the development of what we refer to as "vulnerability profiles,"
2519 based on multivariate statistical analyses such as Principal Components Analysis (PCA). As a
2520 simple example, we conducted a PCA on the mapped indicators. The six principal components
2521 we extracted tended to be associated with different potential dimensions of vulnerability: i.e.,
2522 PCI with at-risk species; PC2 with streamflow availability and usage; PC3 with pesticides in
2523 surface water; PC4 with macroinvertebrates and stream habitat quality; PCS with meteorological
2524 drought indices; and PC6 with herbicides in groundwater. This kind of analysis allows the
2525 identification of watersheds or other geographic units with similar vulnerability profiles. This has
2526 the potential to be useful in the transfer of successful management or adaptation strategies from
2527 one location to another.
2528
Do Not Cite or Quote Page 93
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
Figure 13. Indicator Evaluation Process
This process can be used to evaluate and guide the modification of potential indicators. The
questions are oriented around the definition of vulnerability and the suitability of the indicator
for mapping.
Indicatorselection
Can the indicate:
be modified to
describe
vulnerability?
Does the
indicator describe
vulnerability?
Not appropriate for
vulnerability assessments.
Yes
dicatordisplay
\
YesT
\
i
Are objective
breakpoints in
the range of
vulnerability
documented?
Can objective
breakpoints be
identified?
Are the
breakpoints st
valid when the
data are
aggregated?
Mappablewith objective
breakpoints
Mappablewith arbitrary
breakpoints
B. Recommendations for Future Research
As a result of exploring the challenges and issues described above, we have identified a number
of areas where additional research is likely to contribute significantly to our ability to carry out
indicator-based vulnerability assessments - both in the specific context of the indicators
discussed in this report, and more generally.
a. Assessment of non-mappable indicators
Some indicators were designated as non-mappable due to the need for additional processing of
available data, statistical analyses, evaluation of modeled data, or other tasks that were beyond
the scope of this study. Additional effort to address these needs may yield highly useful maps of
these indicators.
Do Not Cite or Quote
Page 94
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
2545 Examples of the data evaluation needs include:
2546
2547 • Acquiring and assembling national-scale wetland data: Wetlands may be significantly
2548 affected by climate and land-use change. Unfortunately, one important indicator for
2549 wetlands, Wetland Loss (#325), was designated as non-mappable, due to the effort required
2550 to download and process the data from the National Wetlands Inventory (NWI). The online
2551 ordering system requires users to download individual datasets at the 7.5 minute (1:24K) or
2552 15 minute (1:100K) scales. In the lower 48 states, the USGS has designated approximately
2553 56,500 l:24K-scale quadrangles. It may be possible to acquire national wetlands coverage
2554 from the U. S. Fish & Wildlife Service, and conduct subsequent analyses that would result in
2555 a national wetlands indicator.
2556
2557 • Assessment of the National Inventory of Dams database: In-stream connectivity (#620) is an
2558 important measure that can be used to make inferences about drinking water availability (e.g.
2559 large reservoirs) and aquatic ecosystem functions (e.g. migration of species). To produce an
2560 accurate assessment of connectivity, it is important to have a comprehensive source of dam
2561 locations and diversions in the United States. The National Inventory of Dams, managed by
2562 the U.S. Army Corps of Engineers, is an attempt at such a data set, but some data (especially
2563 data pertaining to small dams) is absent from the database, available digital maps of the
2564 stream network are of varying quality and detail across the country, and the available data for
2565 dams are frequently inaccurate. An assessment of this database is needed and, if possible,
2566 additional dam data should be obtained to produce a map for this indicator. Work by the
2567 USGS on the National Hydrography Dataset and the NHD-plus is currently underway and
2568 should provide useful data in the coming years. A challenge to reporting this indicator will be
2569 evaluating what percentage of dams is omitted because they are too small to be registered in
2570 the national database on dams.
2571
2572 • Digitization and analysis of national flood plain data: The Population Susceptible to Flood
2573 Risk (#209) indicator evaluates the human population currently residing within a 500-year
2574 flood plain. A map for this indicator could be obtained by overlaying estimates of the 500-
2575 year flood plain from the Federal Emergency Management Agency (FEMA) with population
2576 data from the U.S. Census Bureau. However, according to FEMA's Map Service Center,
2577 GIS-compatible digital flood plain data were not available at the time of this study for several
2578 areas within the U.S. FEMA is currently working on a multi-year project to update and
2579 digitize national flood plain data. In the absence of a national flood plain data set, it would be
2580 useful to utilize existing digital flood plain data for urbanized areas to evaluate the
2581 percentage of metropolitan populations that may be prone to flooding.
2582 b. Identifying opportunities to enhance source data
2583 The indicators evaluated during this study were associated with data sets with varying degrees of
2584 completeness, ranging from large national assessment efforts, to indicators with no clear data
2585 source. Additional research is needed to identify opportunities to enhance the utility of national
2586 data sets and fill significant data gaps.
2587
2588 Examples of large national data sets that were used for this study include the EPA Wadeable
2589 Streams Assessment or the USGS National Water Quality Assessment (NAWQA) Program.
Do Not Cite or Quote Page 95
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
2590 These are unique data sets that yield high-quality data, but even these excellent data collection
2591 efforts fall short of providing the data density required to produce robust analyses of
2592 vulnerability over large scales, e.g., at the scale of a 4-digit HUC unit, as calculated values may
2593 be highly sensitive to a few or even a single measurement taken at a discrete location within the
2594 spatial aggregation unit. Additional research is needed to evaluate data collection effort required
2595 to enhance the statistical power of these key datasets.
2596
2597 In addition, some example maps produced for this study could be improved by addressing
2598 significant gaps in the source data. For example, the data set used to produce In-stream Use /
2599 Total Streamflow (#351) did not include estimates of groundwater recharge, one of the input
2600 variables for this indicator, for some regions. For these regions, we assumed recharge was equal
2601 to withdrawals. The accuracy of this indicator in these areas would be improved by acquiring
2602 better estimates for the missing variable.
2603
2604 Furthermore, some data sets that are regularly updated through ongoing data collection activities
2605 may have quality problems. For example, the Centers for Disease Control and Prevention's
2606 (CDC) Waterborne Disease and Outbreak Surveillance System (WBDOSS), a potential data set
2607 for the Waterborne Human Disease Outbreaks (#322) indicator, relies on voluntary reporting of
2608 water-related disease outbreaks by public health departments of U.S. states, territories, and local
2609 governments. The data are inconsistent and of variable quality. Ideally, data would be reported
2610 regularly for all parts of the country and consistently documented by a single responsible entity.
2611 Alternatively, if voluntary data collection by multiple entities continues, stringent guidelines
2612 might be set forth to ensure the quality of the data in this database.
2613
2614 Finally, some of the indicators that we deemed to be non-mappable because we could not
2615 identify any existing data source have the potential to be highly useful measures. Additional
2616 research to identify the data needed to calculate appropriate vulnerability metrics, collect new
2617 data, or transform existing data to calculate and map these indicators would be valuable.
2618 c. Development of new indicators from available data sets
2619 A direct follow-up effort to the methodology employed for this study would be a review of
2620 existing national-scale environmental data sets to determine which might lend themselves to the
2621 development of new, useful indicators. This would allow for more opportunities to create
2622 indicators that are specifically tailored to the needs of local planners and decision-makers. For
2623 example, a new indicator, Water Demand, defined as the total water withdrawals in millions of
2624 gallons per day, can be created based on data available from the USGS' National Water-Use
2625 Data set. A map of this indicator is shown in Figure 5. Assessment of vulnerability using this
2626 indicator, perhaps in combination with indicators of water availability such as Groundwater
2627 Depletion (#121) and Net Streamflow per Capita (#623), may be useful at a variety of scales,
2628 from national to local, for understanding the water budgets of communities. This would facilitate
2629 responses with, for example, improved conservation policies in areas subject to severe water
2630 shortages.
2631
2632 Using available data as a starting point would also enhance our ability to work with indicators
2633 with objective thresholds that distinguish between acceptable and degraded condition. For
2634 example, in the present study a set of five pesticide indicators [#367, #369, #371, #373, and
Do Not Cite or Quote Page 96
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
2635 #374] were mapped using USGS' NAWQA data set. These indicators were designed by USGS
2636 to provide a cumulative assessment of multiple pesticides present in ambient water by
2637 calculating an average concentration. It is difficult to determine thresholds for these indicators
2638 given the diversity of pesticides and the varying levels of risks they pose. Instead, the
2639 development of new indicators for individual pesticides, using the same data set, would allow us
2640 to map the data using established thresholds, such as MCLs, to categorize vulnerability.
2641 Individual pesticide indicators may present regional patterns and identify regional water quality
2642 concerns, whereas the combined indicators developed by USGS and used in this study may mask
2643 local and regional vulnerability.
2644 d. Use of indicators for future studies
2645 The focus of the present study was to identify indicators of water quality and aquatic ecosystem
2646 condition that represented vulnerability and could be mapped at the national scale. 598 indicators
2647 were eliminated from the original comprehensive list of indicators for various reasons that made
2648 them unsuitable for a national-scale vulnerability assessment. However, many of these indicators
2649 may be valuable for other studies or purposes.
2650
2651 Many indicators were eliminated because their associated data sets did not have comprehensive
2652 national coverage or may only be relevant in some areas. Although these indicators had limited
2653 utility for the present study, they are likely to be valuable for conducting vulnerability
2654 assessments at regional or local scales. For example, EPA National Coastal Assessment data for
2655 the Water Clarity Index [#318] and Water Quality Index [#319] indicators are only available for
2656 the Gulf coast region. Similarly, Snowpack Depth [#440] is only measured in regions where
2657 rivers and other surface water sources are primarily fed by snowmelt, such as in the Colorado
2658 River basin. Mangrove Cover [#63] is only relevant where these trees grow - a small portion of
2659 the Gulf Coast. Each of these indicators may be highly useful for monitoring changes over time
2660 in local systems and for guiding local decisions in response to observed or expected changes. A
2661 useful follow-up effort to this study would be the development of an indicator compendium that
2662 would describe the geographic extent and available data sources for indicators that are relevant at
2663 local and regional scales. Local decision makers could use this resource in conjunction with the
2664 national-scale indicators presented in this study to guide local planning efforts.
2665
2666 Indicators whose data were based on future projections were also eliminated because the present
2667 study only examined current vulnerability. For example, data for Heat-Related Illnesses
2668 Incidence [#392] are available as estimates of mortality from the National Center for Health
2669 Statistics (NCHS) based on three climate change scenarios for the years 2020 and 2050. Data for
2670 land cover or land use indicators, such as Coastal Wetlands (acreage) (#52) and Urban and
2671 Suburban Areas (acreage) (#308), Population susceptible to flood risk (#209), and other
2672 population-related indicators, may be projected into the future using output data from General
2673 Circulation Models, earth system models, and regional climate models. These data, while not
2674 useful for the present study, are useful in understanding future vulnerability, particularly when
2675 taking into account the effects of climate change on human and natural environments.
2676 Understanding future vulnerability is a crucial component of many ongoing and planned research
2677 studies aimed at strategic planning for adaptation to the effects of global climate change.
Do Not Cite or Quote Page 97
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
2678 e. Establishment of stress-response curves, vulnerability thresholds, and baseline
2679 conditions
2680 In this report we focused on the development of methods to assess relative vulnerability.
2681 Additional research to evaluate how individual indicators respond to stress (e.g. sensitivity,
2682 threshold response, resistance, etc.) will facilitate assessments of absolute vulnerability linked to
2683 system function. There is a large body of basic ecological and sociological research that will
2684 need to be created before this issue can be comprehensively addressed. The issue of thresholds,
2685 much discussed above, is of course intimately related.
2686
2687 Furthermore, observationally establishing baseline conditions, and implementing more routine
2688 monitoring for locally relevant indicators, would enable water resource managers to identify
2689 significant water quality and ecological changes over time, which would allow the development
2690 of additional indicators, or more accurate calculation of existing indicators, for assessment.
2691 f. Drawing on other established approaches for combining indicators
2692 In particular, a comparison of the traditional multivariate approaches for combining indicators to
2693 the approaches used by EPA's ReVA program, such as the generalized weighted distance
2694 method, may be fruitful. Future research efforts could apply the ReVA aggregation methods to
2695 the indicators in this report, which are topically and spatially broader. Such aggregation would
2696 also allow relationships between components of vulnerability for the indicators specified in this
2697 study to be addressed. Future work could include the design of new, robust indicators using
2698 existing data sources.
2699 g. Incorporating landscape metrics
2700 Landscape metrics, such as percent natural cover, roads crossing streams, and agriculture on
2701 slopes, can provide additional context for the indicators presented in the report. Metrics such as
2702 these may assist with the interpretation of sensitivity. Measurements of human impact may
2703 explain an indicator's vulnerability score or may suggest an alternative interpretation. In
2704 addition, some metrics, such as population growth rate, can be used to assess future exposure to
2705 stress (see, for example, Figure 11).
2706 h. Incorporating metrics of adaptive capacity
2707 Vulnerability to future changes depends in part on choices made by society today and into the
2708 future. In the context of climate change in particular, adaptive capacity is the ability of an
2709 ecosystem or society to continue to perform its range of functions despite changes in factors that
2710 affect those functions. A system has inherent adaptive capacity when its natural attributes make
2711 it resilient to stress, whereas institutional adaptive capacity includes policies, practices, and
2712 infrastructure that create options for meeting human and ecosystem needs in the face of an
2713 uncertain future. The specific attributes or actions that create adaptive capacity are largely
2714 different for aquatic life and human uses of water, although there is some overlap among these
2715 categories.
2716
2717 Differentiating inherent and institutional adaptive capacity is useful because it points to two
2718 different management approaches. Systems with inherent adaptive capacity are less vulnerable,
2719 even when they are sensitive and exposed to stress. Thus, many advocate directing planning and
Do Not Cite or Quote Page :
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
2720 management efforts toward systems lacking this capacity. Institutional adaptive capacity can be
2721 built in many ways (for examples, see IPCC, 2007a). Many of these strategies require a
2722 significant shift from short to long term planning, which is typically resisted by institutional and
2723 infrastructural inertia. Many specific practices involve diversification and the creation of
2724 redundancy, which can be hard to justify in the context of current conditions. Some also require
2725 acknowledgement of fundamental uncertainty about the future.
2726
2727 Community-based analyses have shown that the conditions that interact to shape exposures,
2728 sensitivities, adaptive capacities, and hence create needs and opportunities for adaptation, are
2729 community-specific (Smit and Wandel, 2006). This finding suggests that any attempt to transfer
2730 adaptive strategies among regions must look for commonalities both in the magnitude of
2731 vulnerability and in its qualitative, multi-dimensional profile. As described above, some of the
2732 techniques described in this report (e.g., the development of vulnerability profiles and similarity
2733 maps) could, in principle, be used to identify such commonalities among regions, which, in
2734 combination with case studies of successful adaptation, would provide guidance for potential
2735 policy transfer, or serve as a screening tool for the feasibility of adaptive strategy transfer.
2736
2737 As we said above, we hope that this report will be a useful building block for future work on
2738 multi-stressor global change vulnerability assessments. Ultimately, we believe the work
2739 described here is a preliminary contribution toward bridging disconnects between the decision
2740 support needs of the water quality and aquatic ecosystem management communities and the
2741 priorities and capabilities of the global change science data and modeling communities; to the
2742 synthesis of insights across more detailed, place-based, system-based, or issue-based case studies
2743 (e.g., in individual watersheds, wetlands, urban ecosystems) to obtain national-scale insights
2744 about impacts and adaptation; and to prioritization of future work in developing adaptation
2745 strategies for global change impacts.
Do Not Cite or Quote Page 99
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
2746 IX. References
2747 Adger, W.N. 1999. Social vulnerability to climate change and extremes in coastal Vietnam.
2748 World Development. 27: 249-269.
2749 Adger, W. N. 2006. Vulnerability. Global Environmental Change. 16 (3): 268-281.
2750 Adger, W. N., Brooks, N., Bentham, G., Agnew, M., and S. Eriksen. 2004. New Indicators of
2751 Vulnerability and Adaptive Capacity. Tyndall Centre for Climate Change Research.
2752 Technical Report 7. January 2004. Project III. 11.
2753 Barsugli, J., C. Anderson, J.B. Smith, and J.M. Vogel. 2009. Options for Improving Climate
2754 Modeling to Assist Water Utility Planning for Climate Change. Water Utility Climate
2755 Alliance, http://www.wucaonline.org/assets/pdf/actions whitepaper 120909.pdf
2756 Accessed April 27, 2010.
2757 Bates, B. C., Z. W. Kundzewicz, S. Wu, and J. P. Palutikof, Eds. 2008. Climate Change and
2758 Water. Technical Paper of the Intergovernmental Panel on Climate Change (IPCC)
2759 Secretariat. Geneva, 210 pp.
2760 Beisner, B.E., D.T. Haydon, and K. Cuddington. 2003. Alternative stable states in ecology.
2761 Frontiers in Ecology and the Environment. 1 (7): 3 76-3 82.
2762 Bloomfield, J.P., R.J. Williams, D.C. Gooddy, J.N. Cape, and P. Guha. 2006. Impacts of climate
2763 change on the fate and behaviour of pesticides in surface and groundwater—a UK
2764 perspective. Science of the Total Environment. 369 (1-3): 163-177.
2765 Boxall, A.B.A., A. Hardy, S. Beulke, T. Boucard, L. Burgin, P. D. Falloon, P.M., Haygarth, T.
2766 Hutchinson, R.S. Kovats, G. Leonard!, L.S. Levy, G. Nichols, S.A. Parsons, L. Potts, D.
2767 Stone, E. Topp, D.B. Turley, K. Walsh, E.M.H., Wellington, and RJ. Williams. 2009.
2768 Impacts of climate change on indirect human exposure to pathogens and chemicals from
2769 agriculture. Environmental Health Perspectives. 117(4).
2770 Breiman, L., Friedman, J., Stone, C.J., and R.A. Olshen. 1984. Classification and Regression
2771 Trees. Chapman and Hall/CRC. 368 pp.
2772 Brooks, N. 2003. Vulnerability, risk and adaptation: a conceptual framework. Working Paper 38,
2773 Tyndall Centre for Climate Change Research, Norwich, UK.
2774 Brown, C., W. Werick, W. Leger, and D. Fay. 2010. A decision analytic approach to managing
2775 climate risks - Application to the Upper Great Lakes. Journal of the American Water
2776 Resources Association, in review.
2777 Burlando, P., and R. Rosso. 2002. Effects of transient climate change on basin hydrology. 2.
2778 Impacts on runoff variability in the Arno River, central Italy. Hydrological Processes. 16:
2779 1177-1199.
2780 Chen, C-C, and B.A. McCarl. 2001. An investigation into the relationship between pesticide
2781 usage and climate change. Climatic Change. 50 (4): 475-487.
2782 Dai, A., K. Trenberth, and T. Qian. 2004. Global Dataset of Palmer Drought Severity Index for
2783 1870-2002: Relationship with Soil Moisture and Effects of Surface Warming. Journal of
2784 Hydrometerology. 5: 1117-1130.
2785 Dai, A., K.E. Trenberth, and T.R. Karl. 1998. Global Variations in Droughts and Wet Spells:
2786 1900-1995. Geophysical Research Letters. 25: 3367-3370.
2787 Day, J. W. Jr., Barras, J., Clairain, Johnston, J., Justic, D., Kemp, G. P., Ko, J-Y., Lane, R.,
2788 Mitsch, W. J., Steyer, G., Templet, M., and A. Yanez-Arancibia. 2005. Implications of
Do Not Cite or Quote Page 100
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
2789 Global Climatic Change and Energy Cost and Availability for the Restoration of the
2790 Mississippi Delta. Ecological Engineering. 24 (4): 253-265.
2791 Dessai, S., M. Hulme, R. Lempert, and R. Pielke Jr., 2009. Climate prediction: a limit to
2792 adaptation? Adapting to Climate Change: Thresholds, Values, Governance [W.N. Adger,
2793 I. Lorenzoni, and K.L. O'Brien, Eds.], Cambridge University Press.
2794 DeWalle, D. R., B. R. Swistock, T. E. Johnson, and K. J. McGuire. 2000. Potential Effects of
2795 Climate Change and Urbanization on Mean Annual Streamflow in the United States,
2796 Water Resources Research. 36 (9): 2655-2664.
2797 Feng, S., and Q. Hu. 2007. Changes in winter snowfall/precipitation ratio in the contiguous
2798 United States. Journal of Geophysical Research. 112: 12.
2799 Fischhoff, B. 1994. What forecasts (seem to) mean. InternationalJournal of Forecasting. 10:
2800 387-403.
2801 Fiske, A., C.A. Myrick, and L.J. Hansen. 2005. Potential Impacts of Global Climate Change on
2802 Freshwater Fisheries. WWF'- World Wide Fundfor Natura, Gland, Switzerland.
2803 http://assets.panda.org/downloads/fwfishreport902nov05.pdf. Accessed July 16, 2010.
2804 Fraser, E., W. Mabee, and O. Slaymaker. 2003. Mutual vulnerability, mutual dependence: the
2805 reflective notion between human society and the environment. Global Environmental
2806 Change 13:137-144.
2807 Fiissel, H.-M. 2007. Vulnerability: A Generally Applicable Conceptual Framework for Climate
2808 Change Research. Global Environmental Change. 17(2): 155-167.
2809 Garvey, J.E., E.A. Marschall, and R.A. Wright. 1998. From Star Charts to Stoneflies: Detecting
2810 Relationships in Continuous Bivariate Data. Ecology. 79: 442 447.
2811 Gedney, N., Cox, P.M., Betts, R.A., Boucher, O., Huntingford, C., and P. A. Stott. 2006.
2812 Detection of a Direct Carbon Dioxide Effect in Continental River Runoff Records.
2813 Nature. 439: 835-838.
2814 Gilliom, R.J., J.E. Barbash , C.G. Crawford, P.A. Hamilton, J.D. Martin, N. Nakagaki, L.H.
2815 Nowell, J.C. Scott, P.E. Stackelberg, G.P. Thelin, and D.M Wolock. 2006. The Quality of
2816 Our Nation's Waters - Pesticides in the Nation's Streams and Ground Water, 1992-2001.
2817 U.S. Geological Survey Circular 1291.
2818 Graf, W. L. 1999. Dam Nation: A Geographic Census of American Dams and Their Large-Scale
2819 Hydrologic Impacts. Water Resources Research. 35 (4): 1305-1311.
2820 Groffman, P. M., Baron, J. S., Blett, T., Gold, A. J., Goodman, I, Gunderson, L. H., Levinson,
2821 B. M., Palmer, M. A., Paerl, H. W., Peterson, G D., Poff, N. L., Rejeski, D. W.,
2822 Reynolds, J. F., Turner, M. G., Weathers, K. C., and J. Wiens. 2003. Ecological
2823 Thresholds: The Key to Successful Environmental Management or an Important Concept
2824 with No Practical Application?. Ecosystems. 9 (1): 1-13.
2825 Grossman, D. H., D. Faber-Langendoen, A. S. Weakley, M. Anderson, P. Bourgeron, R.
2826 Crawford, K. Goodin, S. Landaal, K. Metzler, K. D. Patterson, M. Pyne, M. Reid, and L.
2827 Sneddon. 1998. International classification of ecological communities: terrestrial
2828 vegetation of the United States. Volume I. The National Vegetation Classification
2829 System: development, status, and applications. The Nature Conservancy, Arlington,
2830 Virginia, USA. http://www.natureserve.org/library/vol 1 .pdf. Accessed July 13, 2010.
2831 Heim, R.R. 2002. A Review of Twentieth-Century Drought Indices Used in the United States.
2832 Bulletin of the American Meteorological Society. 83: 1149-1165.
Do Not Cite or Quote Page 101
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
2833 H. John Heinz III Center for Science, Economics, and the Environment (Heinz Center). 2002.
2834 The State of the Nation's Ecosystems 2002: Measuring the Lands, Waters, and Living
2835 Resources of the United States. New York, NY: Cambridge University Press.
2836 Heinz Center. 2008. The State of the Nation's Ecosystems 2008: Measuring the Land, Waters,
2837 and Living Resources of the United States. Washington, DC: Island Press.
2838 Holling, C.S. 1973. Resilience and stability of ecological systems. Annual Review of Ecological
2839 Systems. 4:1-24.
2840 Holling, C.S. 2001. Understanding the complexity of economic, ecological, and social systems.
2841 Ecosystems. 4: 390-405.
2842 Huntington, T.G., G.A. Hodgkins, B.D. Keim, and R.W. Dudley. 2004. Changes in the
2843 Proportion of Precipitation Occurring as Snow in New England (1949-2000). Journal of
2844 Climate 17: 2626-2636.
2845 Kurd, B., N. Leary, R. Jones, and J. Smith. 1999. Relative Regional Vulnerability of Water
2846 Resources to Climate Change. Journal of the American Water Resources Association. 35
2847 (6): 1399-1409.
2848 Kurd, B., N. Leary, R. Jones, and K. Spiecker. 1998. Water Climate Change: A National
2849 Assessment of Regional Vulnerability. Report prepared for the United States
2850 Environmental Protection Agency. September 1998.
2851 Intergovernmental Panel on Climate Change (IPCC). 2007a. Summary for Policymakers.
2852 Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working
2853 Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate
2854 Change. Eds. M. L. Parry, O. F. Canziani, J. P. Palutikof, P. J. van der Linden, and C. E.
2855 Hanson. Cambridge, UK: Cambridge University Press.
2856 IPCC. 2007b. Summary for Policymakers. Climate Change 2007: The Physical Science Basis.
2857 Contribution of Working Group I to the Fourth Assessment Report of the
2858 Intergovernmental Panel on Climate Change Eds. S. Solomon, D. Qin, M. Manning, Z.
2859 Chen, M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller. Cambridge University Press,
2860 Cambridge, United Kingdom and New York, NY, USA.
2861 Johnson, L. K. Hayhoe, G. Kling, J. Magnuson, and B. Shuter. 2007. Confronting Climate
2862 Change in the Great Lakes Region - Technical Appendix: Wetland Ecosystems. Union of
2863 Concerned Scientists. http://www.ucsusa.org/greatlakes/pdf/wetlands.pdf Accessed July
2864 16,2010.
2865 Johnson, T.E., and C.P. Weaver. 2009. A Framework for Assessing Climate Change Impacts on
2866 Water and Watershed Systems. Environmental Management, 43 (1): 118-134.
2867 Jones, R., 2001. An environmental risk assessment/management framework for climate change
2868 impact assessments. Natural Hazards. 23: 197-230.
2869 Karl, T.R., and W.E. Riebsame. 1989. The impact of decadal fluctuations in mean precipitation
2870 and temperature on runoff: a sensitivity study over the United States. Climatic Change
2871 15: 423-447.
2872 Karl, T.R., P.Y. Groisman, R.W. Knight, and R.R. Heim, Jr. 1993. Recent Variations of Snow
2873 Cover and Snowfall in North America and Their Relation to Precipitation and
2874 Temperature Variations. Journal of Climate. 6: 1327-1344.
2875 Kelly, P.M., Adger, W.N., 2000. Theory and practice in assessing vulnerability to climate change
2876 and facilitating adaptation. Climatic Change. 47: 325-352.
Do Not Cite or Quote Page 102
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
2877 Kirwan, M. L., and G. R. Guntenspergen. 2010. Influence of Tidal Range on the Stability of
2878 Coastal Marshland. Journal of Geophysical Research. 115 (F02009).
2879 doi:10.1029/2009JF001400.
2880 Knowles, N., M.D. Dettinger, and D.R. Cayan. 2006. Trends in Snowfall versus Rainfall in the
2881 Western United States. Journal of Climate. 19: 4545-4559.
2882 Kundzewicz, Z. W., L. J. Mata, N. W. Arnell, P. Doll, P. Kabat, B. Jimenez, K. A. Miller, T.
2883 Oki, Z. Sen, and I. A. Shiklomanov. 2007. Chapter 3: Freshwater Resources and their
2884 Management. Climate Change 2007: Impacts, Adaptation and Vulnerability.
2885 Contribution of Working Group II to the Fourth Assessment Report of the
2886 Intergovernmental Panel on Climate Change, M. L. Parry, O. F. Canziani, J. P. Palutikof,
2887 P. J. van der Linden and C. E. Hanson, Eds., Cambridge University Press, Cambridge,
2888 UK: 173-210.
2889 Kurtz, J. C., Jackson, L. E., and W. S. Fisher. 2001. Strategies for Evaluating Indicators Based
2890 on Guidelines from the Environmental Protection Agency's Office of Research and
2891 Development. Ecological Indicators. 1 (1): 49-60.
2892 Larson, S. J., C. G. Crawford, and R.J. Gilliom. 2004. Development and Application of
2893 Watershed Regressions for Pesticides (WARP) for Estimating Atrazine Concentration
2894 Distributions in Streams. U.S. Geological Survey Water-Resources Investigations Report
2895 03-4047, 68 p.
2896 Lempert, R., N. Nakicenovic, D. Sarewitz, and M. Schlesinger. 2004. Characterizing climate-
2897 change uncertainties for decision-makers. Climatic Change. 65: 1-9.
2898 Lettenmaier, D., D. Major, L. Poff, and S. Running. 2008. Chapter 4: Water Resources. The
2899 Effects of Climate Change on Agriculture, Land Resources, Water Resources, and
2900 Biodiversity. A Report by the U.S. Climate Change Science Program and the
2901 Subcommittee on Global Change Research. Washington, DC: U.S. Climate Change
2902 Science Program.
2903 Lewontin, R.C. 1969. The meaning of stability. Brookhaven Symposia in Biology. 22:13-23.
2904 Lin, B.B., and P.E. Morefield. 2010. The vulnerability cube: A multi-dimensional framework for
2905 assessing relative vulnerability. Environmental Science and Policy, submitted.
2906 Lind, A.J. Amphibians and Reptiles and Climate Change. Nevada Research Center, Pacific
2907 Southwest Research Station, Davis, CA. Accessible at:
2908 http://www.fs.fed.us/ccrc/topics/amphibians-reptiles.shtml.
2909 Link, J. S.. 2005. Translating Ecosystem Indicators into Decision Criteria. ICES Journal of
2910 Marine Science. 62: 569-576.
2911 Lucier, A., M. Palmer, H. Mooney, K. Nadelhoffer, D. Ojima, and F. Chavez. 2006. Ecosystems
2912 and Climate Change: Research Priorities for the U.S. Climate Change Science Program.
2913 Recommendations from the Scientific Community. Report on an Ecosystems Workshop,
2914 prepared for the Ecosystems Interagency Working Group. Special Series No. SS-92-06,
2915 University of Maryland Center for Environmental Science, Chesapeake Biological
2916 Laboratory, Solomons, MD, USA. 50pp.
2917 http://www.usgcrp.gov/usgcrp/Library/ecosystems/eco-workshop-report-jun06.pdf.
2918 Accessed July 16, 2010.
2919 Luers, A. L. 2005. The Surface of Vulnerability: An Analytical Framework for Examining
2920 Environmental Change. Global Environmental Change Part A. 15 (3): 214-223.
2921 Maurer, E.P., D.P. Lettenmaier, and N.J. Mantua. 2004. Variability and potential sources of
2922 predictability of North American runoff. Water Resources Research 40, 13pp.
Do Not Cite or Quote Page 103
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
2923 May, R.M. 1977. Thresholds and breakpoints in ecosystems with a multiplicity of states. Nature.
2924 267:471-477.
2925 Mendelssohn, I. A., Morris, J.T., 2000. Eco-physiological Controls on the Productivity of
2926 Spartina alterniflora loisel. In: Weinstein, M.P., Kreeger, D.A. (Eds.), Concepts and
2927 Controversies in Tidal Marsh Ecology. Kluwer Academic Publishers, Boston, MA, pp.
2928 59-80.
2929 Meyer, J.L., Sale, M.J., Mulholland, P.J., and Poff, N.L., 1999. Impacts of climate change on
2930 aquatic ecosystem functioning and health. Journal of the American Water Resources
2931 Association. 35(6), 1393-1386.
2932 Miller, K.A., and D. Yates (NCAR). 2005. Climate Change and Water Resources: A Primer for
2933 Municipal Water Providers. AWWA Research Foundation, Denver, CO.
2934 Monmonier, Mark S. 1996 (2nd ed.). How to lie with maps. Chicago: University of Chicago
2935 Press. 207pp.
2936 Montgomery, D. R., G. E. Grant, and K. Sullivan, 1995. Watershed Analysis as a Framework for
2937 Implementing Ecosystem Management. Water Resources Bulletin. 31 (3): 369-386.
2938 Moreno, A., Becken, S. 2009. A climate change vulnerability assessment methodology for
2939 coastal tourism. Journal of Sustainable Tourism 17(4): 473-488.
2940 Moser, S.C., and A.L. Luers, 2008: Managing climate risks in California: The need to engage
2941 resource managers for successful adaptation to change. Climatic Change, 87, S309-S322.
2942 Moss, R.H., Brenkert, A.L., Malone, E.L., 2001. Vulnerability to climate change: a quantitative
2943 approach. Technical Report PNNL-SA-33642, Pacific Northwest National Laboratories,
2944 Richland, WA.
2945 Munn, M.D., and Gilliom, R.J., 2001, Pesticide toxicity index for freshwater aquatic organisms:
2946 U.S. Geological Survey Water-Resources Investigations Report 2001-4077, 55 p.
2947 Murdoch, P.S., Baron, J. S., and T. L. Miller. 2000. Potential Effects of Climate Change on
2948 Surface-Water Quality in North America. Journal of the American Water Resources
2949 Association. 36 (2): 347-366.
2950 National Research Council (NRC), 2009: Informing Decisions in a Changing Climate. Panel on
2951 Strategies and Methods for Climate-Related Decision Support, Committee on the Human
2952 Dimensions of Global Change. Division of Behavioral and Social Sciences and
2953 Education. The National Academies Press, Washington, DC, 188 pp.
2954 National Research Council. 1992. Restoration of Aquatic Ecosystems. National Academy Press,
2955 Washington, D.C. 552 pp.
2956 Nicholson, M. D., and S. Jennings. 2004. Testing Candidate Indicators to Support Ecosystem-
2957 based Management: The Power of Monitoring Surveys to Detect Temporal Trends in
2958 Fish Community Metrics. ICES Journal of Marine Science. 61 (1): 35-42.
2959 Norton, D.J., J.D. Wickham, T.G. Wade, K. Kunert, J.V. Thomas, and P. Zeph. 2009. A Method
2960 for Comparative Analysis of Recovery Potential in Impaired Waters Restoration.
2961 Environmental Management. 44: 356-368.
2962 Noyes, P.O., M.K. McElwee, H.D. Miller, B.W. Clark, L.A. Van Tiem, K.C. Walcott, K.N.
2963 Erwin, E.D. Levin, 2009. The Toxicology of Climate Change: Environmental
2964 Contaminants in a Warming World. Environment International. 35 (6): 971-986.
2965 O'Brien, K., Eriksen, S., Schjolen, A., Nygaard, L., 2004. What's in a word? Conflicting
2966 interpretations of vulnerability in climate change research. CICERO Working Paper
2967 2004:04, CICERO, Oslo University, Oslo, Norway.
Do Not Cite or Quote Page 104
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
2968 Olmos, S., 2001. Vulnerability and adaptation to climate change: concepts, issues, assessment
2969 methods. Climate Change Knowledge Network. Available at
2970 http ://www. cckn.net/pdf/va_foundation_fmal .pdf
2971 Omernik, J.M. 1987. Ecoregions of the conterminous United States. Map (scale 1:7,500,000).
2972 Annals of the Association of American Geographers. 77 (1): 118-125.
2973 Omernik, J. M., and G. E. Griffith. 1991. Ecological Regions versus Hydrologic Units -
2974 Frameworks for Managing Water Quality. Journal of Soil and Water Conservation. 46
2975 (5): 334-340.
2976 Palmer, M. A., Lettenmaier, D. P., Poff, L. N., Postel, S. L., Richter, B., and R. Warner. 2008.
2977 Climate Change and River Ecosystems: Protection and Adaptation Options.
2978 Environmental Management. 44 (6): 1053-1068.
2979 Poff, L.N., M.M. Brinson, and J.W.J. Day, 2002: Aquatic Ecosystems and Global Climate
2980 Change: Potential Impacts on Inland Freshwater and Coastal Wetland Ecosystems in the
2981 United States. Pew Center on Global Climate Change, Arlington, Virginia, 44 pp.
2982 Polsky, C., Neff, R., Yarnal, B. 2007. Building comparable global change vulnerability
2983 assessments: the vulnerability scoping diagram. Global Environmental Change 17: 472-
2984 485.
2985 Sankarasubramanian, A., and R. M. Vogel. 2001. Climate Elasticity of Streamflow in the United
2986 States. Water Resources Research. 37(6): 1771-1781.
2987 Sarewitz, D., R.A. Pielke, Jr., and R. Byerly, Jr. (Eds.), 2000: Prediction: Science, Decision
2988 Making, and the Future of Nature. Island Press, Washington, DC, 405 pp.
2989 Scheffer, M., S. Carpenter, J.A. Foley, C. Folke, and B. Walker. 2001. Catastrophic shifts in
2990 ecosystems. Nature. 413:591-596.
2991 Seaber, P. R., F. P. Kapinos, and G. L. Knapp. 1987. Hydrologic Unit Maps. Water-Supply
2992 Paper 2294. U.S. Geological Survey, Washington, D.C.
2993 Smit, B., and J. Wandel. 2006. Adaptation, Adaptive Capacity and Vulnerability. Global
2994 Environmental Change. 16 (3): 282-292.
2995 Sutherland, J.P. 1974. Multiple stable points in natural communities. The American Naturalist.
2996 108:859-873.
2997 Thieler, E. R., and E. S. Hammar-Klose. 2000. National Assessment of Coastal Vulnerability to
2998 Sea-Level Rise: Preliminary Results for the U.S. Gulf of Mexico Coast. United States
2999 Geological Survey Open-File Report 00-179. Accessed at:
3000 http://pubs.usgs.gov/of/2000/ofOO-179/index.html
3001 Tompkins, E.L., and W.N. Adger. 2004. Does adaptive management of natural resources
3002 enhance resilience to climate change? Ecology and Society 9, 10. Accesed at:
3003 http://www.ecologvandsocietv.org/vol9/iss2/artlO
3004 Iran, L.T., R.V. O'Neill, and E.R. Smith. 2006. A Generalized Distance Measure for Integrating
3005 Multiple Environmental Assessment Indicators. Landscape Ecology. 21 (4): 469-476.
3006 U. S. Climate Change Science Program (USCCSP). 2008. Preliminary Review of Adaptation
3007 Options for Climate-sensitive Ecosystems and Resources. U.S. Climate Change Science
3008 Program and the Subcommittee on Global Change Research. [Julius, S.H., J.M. West
3009 (eds.), J.S. Baron, L.A. Joyce, P. Kareiva, B.D. Keller, M.A. Palmer, C.H. Peterson, and
3010 J.M. Scott (Authors)]. U.S. Environmental Protection Agency, Washington, DC, USA,
3011 873pp.
Do Not Cite or Quote Page 105
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
3012 United States Environmental Protection Agency (USEPA). 1995. Watershed Protection: A
3013 Statewide Approach. EPA 841-R-95-004. Office of Water, U.S. Government Printing
3014 Office, Washington, D.C.
3015 USEPA. 2000a. Evaluation Guidelines for Ecological Indicators. EPA/620/R-99/005. May 2000.
3016 Eds. Jackson, L. E., Kurtz, J.C., and W. S. Fisher.
3017 USEPA. 2000b. Mid-Atlantic Stressor Profile Atlas. EPA/600/C-99/003. December 1999.
3018 Updated March 27, 2000.
3019 USEPA. 2004. Wadeable Streams Assessment Field Operations Manual: The States Assess the
3020 Nation's Streams. EPA 841-B-04-004. Final. July 2004.
3021 USEPA. 2006. Wadeable Streams Assessment: A Collaborative Survey of the Nation's Streams.
3022 EPA 841-B-06-002. Washington, DC: United States Environmental Protection Agency,
3023 Office of Water. December 2006.
3024 USEPA. 2008a. 2011 National Wetland Condition Assessment: Candidate Indicators of
3025 Ecological Condition. Unpublished Preliminary Rough Draft.
3026 USEPA. 2008b. Chapter 3: Water. U.S. EPA's 2008 Report on the Environment. Final Report.
3027 EPA/600/R-07/045F (NTIS PB2008-112484). Washington, DC: U.S. Environmental
3028 Protection Agency. May 2008.
3029 USEPA. 2008c. National Water Program Strategy Response to Climate Change. Report No.
3030 EPA800-R-08-001.
3031 USEPA. 2009a. Regional Vulnerability Assessment (ReVA) Program. Available on the Internet
3032 at: www.epa.gov/reva/. Accessed December 3, 2010. Last updated June 19, 2009.
3033 USEPA. 2009b. 2009 Edition of the Drinking Water Standards and Health Advisories. EPA 822-
3034 R-09-011. Update. October 2009.
3035 USEPA. 2009c. Final Contaminant Candidate List 3 Chemicals: Classification of the PCCL to
3036 CCL. EPA 815-R-09-008. August 2009.
3037 USEPA. 2009d. Land-Use Scenarios: National-Scale Housing-Density Scenarios Consistent
3038 with Climate Change Storylines. EPA/600/R-08/076F. Final Report.
3039 USEPA. 2010a. Environmental Monitoring & Assessment Program. Available on the Internet at:
3040 http://www.epa.gov/emap/. Accessed December 3, 2010. Last updated November 2,
3041 2010.
3042 USEPA. 201 Ob. National Aquatic Resource Surveys. Available on the Internet at:
3043 http://water.epa.gov/type/watersheds/monitoring/nationalsurveys.cfm. Accessed
3044 December 3, 2010. Last updated November 30, 2010.
3045 USEPA. 2010c. Impaired Waters and Total Maximum Daily Loads. Available on the Internet at:
3046 http://water.epa.gov/lawsregs/lawsguidance/cwa/tmdl/index.cfm. Accessed December 3,
3047 2010. Last updated November 16, 2010.
3048 USEPA. 201 Od. Watershed Assessment, Tracking & Environmental Results: Water Quality
3049 Assessment and Total Maximum Daily Loads Information (ATTAINS). Available on the
3050 Internet at: http://www.epa.gov/waters/ir/index.html. Accessed December 3, 2010. Last
3051 updated September 1, 2010.
3052 USEPA. 2010e. Climate Change Indicators in the United States. EPA 430-R-10-007, U.S.
3053 Environmental Protection Agency, Washington, DC, 74 pp.
3054 USEPA. 201 Of. Implications of climate change for bioassessment programs and approaches to
3055 account for effects (external review draft). Washington, D.C.: Office of Research and
3056 Development. Report no. EPA/600/R-10/xxx.
Do Not Cite or Quote Page 106
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
3057 USEPA. 2010g. Regulating Public Water Systems and Contaminants Under the Safe Drinking
3058 Water Act. Available online at: http://www.epa.gov/safewater/standard/setting.html.
3059 Accessed August 3, 2010.
3060 United States Geological Survey (USGS). 1999. The Quality of Our Nation's Waters-Nutrients
3061 and Pesticides. National Water-Quality Assessment Program. Circular 1225. Reston, VA:
3062 United States Geological Survey (USGS).
3063 Villa, F. and H. McLeod. 2002. Environmental Vulnerability Indicators for Environmental
3064 Planning and Decision-Making: Guidelines and Applications. Environmental
3065 Management. 29: 335 - 348.
3066 Vogel, R.M., I. Wilson, and C. Daly. 1999. Regional Regression Models of Annual Streamflow
3067 for the United States. Journal of Irrigation and Drainage Engineering. 125(3): 148-157.
3068 WRC. 1978. The Nation's Water Resources: 1975-2000 (Vol. 2). U.S. Government Printing
3069 Office, Washington D.C.
3070 Wolock, D.M., and GJ. McCabe. 1999. Explaining Spatial Variability in Mean Annual Runoff
3071 in the Conterminous United States. Climate Research. 11: 149-159.
3072 Yang, D., Kanae, S., Oki, T., Koike, T., and K. Musiake. 2003. Global Potential Soil Erosion
3073 with Reference to Land Use and Climate Changes. HydrologicalProcesses. 17: 2913-
3074 2928.
Do Not Cite or Quote Page 107
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
X. Appendices
Do Not Cite Or Quote
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
The following appendices accompany the report entitled Approach for Developing a National
Atlas of Vulnerabilities for U.S. Water Quality and Aquatic Ecosystems.
Appendix A. Bibliography
Appendix B. Comprehensive List of Indicators
Appendix C. Data Sources and Supporting Information
Appendix D. Technical Notes
Appendix E. Mapping Methodology
Appendix F. Example Maps for Indicators of Water Quality and Aquatic Ecosystem
Vulnerability by HUC-4 Watershed
Appendix G. Descriptions of Example Indicator Maps by HUC-4 Watershed
Appendix H. Example Maps for Indicators of Water Quality and Aquatic Ecosystem
Vulnerability by Ecoregion
Appendix I. Descriptions of Example Indicator Maps by Ecoregion
Appendix J. Vulnerability Category Matrix
Appendix K. Evaluation and Potential Modification of Vulnerability Indicators
Appendix L. Contact Information
Do Not Cite Or Quote
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Do Not Cite or Quote Page A-l
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
The literature sources in the following list were reviewed to identify indicators of vulnerability of
water quality and aquatic ecosystems. Core Literature studies (i.e., those identified by the Global
Change Research Program (GCRP) as starting points for building the comprehensive list of
indicators for this study) are identified by a * before the authors' names.
Adger, W. N., N. Brooks, M. Kelly, G. Bentham, and S. Eriksen. 2004. New Indicators of
Vulnerability and Adaptive Capacity. Technical Report 7. Norwich, UK: Tyndall Centre
for Climate Change Research. January 2004.
Anderson, J., K. Arblaster, J. Bartsley, T. Cooper, M. Kettunen, T. Kaphengst, A. Leipprand, C.
Laaser, K. Umpfenbach, E. Kuusisto, A. Lepisto, and M. Holmberg. 2007. Climate
Change-induced Water Stress and its Impact on Natural and Managed Ecosystems.
IP/A/CLIM/ST/2007-06. Brussels, Belgium: European Parliament.
Arnell, N. 1998. Climate Change and Water Resources in Britain. Climatic Change. 39 (1): 83-
110.
Arnell, N. 1999. Climate Change and Global Water Resources. Global Environmental Change. 9
(Supplement 1): S31-S49.
Barnett, T. P., J. C. Adam, and D. P. Lettenmaier. 2005. Potential Impacts of a Warming Climate
on Water Availability in Snow-Dominated Regions. Nature. 438 (7066): 303-309.
Bergstrom, S., B. Carlsson, M. Gardelin, G. Lindstrom, A. Pettersson, andM. Rummukainen.
2001. Climate Change Impacts on Runoff in Sweden - Assessments by Global Climate
Models, Dynamical Downscaling and Hydrological Modeling. Climate Research. 16:
101-112.
Bexfield, L. M. 2008. Decadal-Scale Changes of Pesticides in Ground Water of the United
States, 1993-2003. Journal of'Environmental Quality. 37: S226-S239.
Bradbury, J. A., S. L. Dingman, and B. D. Keim. 2002. New England Drought and Relations
with Large Scale Atmospheric Circulation Patterns. Journal of the American Water
Resources Association (JAWRA). 38 (5): 1287-1299.
Brezonik, P. L., L. G. Olmanson, M. E. Bauer, and S. M. Kloiber. 2007. Measuring Water
Clarity and Quality in Minnesota Lakes and Rivers: A Census-Based Approach Using
Remote-Sensing Techniques. Center for Urban and Regional Affairs (CURA) Reporter.
1>1 (3): 3-13. Minneapolis, MN: University of Minnesota.
Brooks, N., W. N. Adger, and P. M. Kelly. 2005. The Determinants of Vulnerability and
Adaptive Capacity at the National Level and the Implications for Adaptation. Global
Environmental Change. 15 (2): 151-163.
Bunn, S. E., and A. H. Arthington. 2002. Basic Principles and Ecological Consequences of
Altered Flow Regimes for Aquatic Biodiversity. Environmental Management. 30 (4):
492-507.
Do Not Cite or Quote Page A-2
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Burke, E., S. J. Brown, andN. Christidis. 2006. Modeling the Recent Evolution of Global
Drought and Projections for the Twenty-First Century with the Hadley Centre Climate
Model. Journal ofHydrometeorology. 7 (5): 1113-1125.
Burkett, V., and J. Kusler. 2000. Climate Change: Potential Impacts and Interactions in Wetlands
in the United States. Journal of the American Water Resources Association (JAWRA). 36
(2): 313-320.
Chesapeake Bay Program. 2008. Chesapeake Bay: Health & Restoration Assessment 2007. EPA-
903-R-08-002 (CBP/TRS-291-08). A Report to the Citizens of the Bay Region.
Annapolis, MD: Chesapeake Bay Program. March 2008.
*Coastal States Organization (CSO). 2007. The Role of Coastal Zone Management Programs in
Adaptation to Climate Change: Synthesis Report from the CSO Climate Change Work
Group.
Conway, D., and M. Hulme. 1996. The Impacts of Climate Variability and Future Climate
Change in the Nile Basin on Water Resources in Egypt. International Journal of Water
Resources Development. 12 (3): 277-296.
Crabbe, P. and M. Robin. 2006. Institutional Adaptation of Water Resource Infrastructures to
Climate Change in Eastern Ontario. Climatic Change. 78 (1): 103-133.
Dai, A., K. E. Trenberth, and T. R. Karl. 1999. Effects of Clouds, Soil Moisture, Precipitation,
and Water Vapor on Diurnal Temperature Range. Journal of Climate. 12 (8): 2451-2473.
Day, J. W., J. Barras, E. Clairain, J. Johnston, D. Justic, G. P. Kemp, J. Y. Ko, R. Lane, W. J.
Mitsch, G. Steyer, P. Templet, and A. Yanez-Arancibia. 2005. Implications of Global
Climatic Change and Energy Cost and Availability for the Restoration of the Mississippi
Delta. Ecological Engineering. 24 (4): 253-265.
Day, J. W., D. F. Boesch, E. J. Clairain, G. P. Kemp, S. B. Laska, W. J. Mitsch, K. Orth, H.
Mashriqui, D. J. Reed, L. Shabman, C. A. Simenstad, B. J. Streever, R. R. Twilley, C. C.
Watson, J. T. Wells, and D. F. Whigham. 2007. Restoration of the Mississippi Delta:
Lessons from Hurricanes Katrina and Rita. Science. 315 (5819): 1679-1684. March 23,
2007.
Day, J. W., R. R. Christian, D. M Boesch, A. Yanez-Arancibia, J. Morris, R. R. Twilley, L.
Naylor, L. Schaffner, and C. Stevenson. 2008. Consequences of Climate Change on the
Ecogeomorphology of Coastal Wetlands. Estuaries and Coasts. 31 (3): 477-491.
de Loe, R., R. Kreutzwiser, and L. Moraru. 2001. Adaptation Options for the Near Term:
Climate Change and the Canadian Water Sector. Global Environmental Change. 11 (3):
231-245.
de Wit, M., and J. Stankiewicz. 2006. Changes in Surface Water Supply Across Africa with
Predicted Climate Change. Science. 311 (5769): 1917-1921. 31 March 2006.
Do Not Cite or Quote Page A-3
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
*Ebi, K. L., G. A. Meehl, D. Bachelet, J. M. Lenihan, R. P. Neilson, R. R. Twilley, D. F. Boesch,
V. J. Coles, D. G. Kimmel, and W. D. Miller. 2007. Regional Impacts of Climate
Change: Four Case Studies in the United States. Prepared for the Pew Center on Global
Climate Change. Arlington, VA: Pew Center on Global Climate Change. December
2007.
Frumhoff, P. C., J. C. McCarthy, J. M. Melillo, S. C. Moser, and D. J. Wuebbles. 2006. Climate
Change in the U.S. Northeast. A Report of the Northeast Climate Impacts Assessment.
October 2006. Cambridge, MA: Union of Concerned Scientists (UCS).
*Frumhoff, P. C., J. C. McCarthy, J. M. Melillo, S. C. Moser, and D. J. Wuebbles. 2007.
Confronting Climate Change in the U.S. Northeast: Science, Impacts, and Solutions.
Synthesis Report of the Northeast Climate Impacts Assessment (NECIA). Cambridge,
MA: Union of Concerned Scientists (UCS).
Gibson, J. J., T. D. Prowse, and D. L. Peters. 2006. Partitioning Impacts of Climate and
Regulation on Water Level Variability in Great Slave Lake. Journal of Hydrology. 329
(1-2): 196-206.
*Gilliom, R. J., J. E. Barbash, C. G. Crawford, P. A. Hamilton, J. D. Martin, N. Nakagaki, L. H.
Nowell, J. C. Scott, P. E. Stackelberg, G. P. Thelin, and D. M. Wolock. 2006. The
Quality of Our Nation's Waters — Pesticides in the Nation's Streams and Ground Water,
1992-2001. National Water-Quality Assessment Program. Circular 1291: 172 pp. Reston,
VA: United States Geological Survey (USGS).
*Gleick, P. H., and D. B. Adams. 2000. Water: The Potential Consequences of Climate
Variability and Change for the Water Resources of the United States. The Report of the
Water Sector Assessment Team of the National Assessment of the Potential
Consequences of Climate Variability and Change for the U.S. Global Change Research
Program. Oakland, CA: Pacific Institute for Studies in Development, Environment, and
Security. September 2000.
Grimm, N. B., A. Chacon, C. N. Dahmn, S. W. Hosteller, O. T. Lind, P. L. Starkweather, and W.
W. Wurtsbaugh. 1997. Sensitivity of Aquatic Ecosystems to Climatic and Anthropogenic
Changes: The Basin and Range, American Southwest and Mexico. Hydrological
Processes. 11 (8): 1023-1041.
*H. John Heinz III Center for Science, Economics, and the Environment, The (Heinz Center,
The). 2002. The State of the Nation's Ecosystems 2002: Measuring the Lands, Waters,
and Living Resources of the United States. New York, NY: Cambridge University Press.
*Heinz Center, The. 2008. The State of the Nation's Ecosystems 2008: Measuring the Land,
Waters, and Living Resources of The United States. Washington, DC: Island Press.
Do Not Cite or Quote Page A-4
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
*Hamilton, P. A., T. L. Miller, and D. N. Myers. 2004. Water Quality in the Nation's Streams
and Aquifers — Overview of Selected Findings, 1991-2001. National Water-Quality
Assessment Program. Circular 1265. Reston, VA: United States Geological Survey
(USGS).
Hayashi, M., and D. O. Rosenberry. 2002. Effects of Ground Water Exchange on the Hydrology
and Ecology of Surface Water. Ground Water. 40 (3): 309-316.
Hayslip, G., L. Edmond, V. Partridge, W. Nelson, H. Lee, F. Cole, J. Lamberson, and L. Caton.
2006. Ecological Condition of the Estuaries of Oregon and Washington. EPA 910-R-06-
001. Seattle, WA: United States Environmental Protection Agency, Office of
Environmental Assessment, Region 10. March 2006.
Hodgkins, G. A., R. W. Dudley, and T. G. Huntington. 2003. Changes in the Timing of High
River Flows in New England over the 20th Century. Journal of Hydrology. 278 (1-4):
244-252.
Hughes, T. P., A. H. Baird, D. R. Bellwood, M. Card, S. R. Connolly, C. Folke, R. Grosberg, H.
Hoegh-Guldberg, J. B. C. Jackson, J. Kleypas, J. M. Lough, P. Marshall, M. Nystrom, S.
R. Palambi, J. M. Pandolfi, B. Rosen, and J. Roughgarden. 2003. Climate Change,
Human Impacts, and the Resilience of Coral Reefs. Science. 301 (5635): 929-933. 15
August 2003.
Huntington, T. G., G. A. Hodgkins, B. D. Keim, and R. W. Dudley. 2004. Changes in the
Proportion of Precipitation Occurring as Snow in New England (1940-2000). Journal of
Climate. 17 (13): 2626-2636.
*Hurd, B., N. Leary, R. Jones, and J. Smith. 1999. Relative Regional Vulnerability of Water
Resources to Climate Change. Journal of the American Water Resources Association. 35
(6): 1399-1409.
*Hurd, B., N. Leary, R. Jones, and K. Spiecker. 1998. Water Climate Change: A National
Assessment of Regional Vulnerability. Report prepared for the United States
Environmental Protection Agency. September 1998.
Intergovernmental Panel on Climate Change (IPCC). 2007. Summary for Policymakers. In:
Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working
Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate
Change. Eds. M. L. Parry, O. F. Canziani, J. P. Palutikof, P. J. van der Linden, and C. E.
Hanson. Cambridge, UK: Cambridge University Press.
Kling, G. W., K. Hayhoe, L. B. Johnson, J. J. Magnuson, S. Polasky, S. K. Robinson, B. J.
Shuter, M. M. Wander, D. J. Wuebbles, D. R. Zak, R. L. Lindroth, S. M. Moser, and M.
L. Wilson. 2003. Confronting Climate Change in the Great Lakes Region: Impacts on
Our Communities and Ecosystems. Cambridge, MA: Union of Concerned Scientists, and
Washington, D.C.: Ecological Society of America. April 2003.
Do Not Cite or Quote Page A-5
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Kundzewicz, Z. W., L. J. Mata, N. W. Arnell, P. Doll, B. Jimenez, K. Miller, T. Oki, Z. Sen, and
I. Shiklomanov. 2008. The Implications of Projected Climate Change for Freshwater
Resources and their Management. Hydrological Sciences - Journal des Sciences
Hydrologiques. 53 (1): 3-10.
Lehner, B. 2006. Estimating the Impact of Global Change on Flood and Drought Risks in Europe
- A Continental, Integrated Analysis. Climatic Change. 75 (3): 273-299.
*Lettenmaier, D., D. Major, L. Poff, and S. Running. 2008. Chapter 4: Water Resources. In: The
Effects of Climate Change on Agriculture, Land Resources, Water Resources, and
Biodiversity. A Report by the U.S. Climate Change Science Program and the
Subcommittee on Global Change Research. Washington, DC: U.S. Climate Change
Science Program.
Lins, H. F., and J. R. Slack. 2005. Seasonal and Regional Characteristics of U.S. Streamflow
Trends in the U.S. from 1940 to 1999. Physical Geography. 26 (6): 489-501.
Long Island Sound Study. 2008. Sound Health 2008: Status and Trends in the Health of Long
Island Sound. Stamford, CT: Long Island Sound Study.
Luers, A. L., D. R. Cayan, G. Franco, M. Hanemann, and B. Croes. 2006. Our Changing
Climate: Assessing the Risks to California. A Summary Report from the California
Climate Change Center. CEC-500-2006-077. Sacramento, CA: California Energy
Commission and California Department of Environmental Protection. July 2006.
McCabe, G. J. and D. M. Wolock. 2002. A Step Increase in Streamflow in the Conterminous
United States. Geophysical Research Letters. 29 (24): 2185-2188.
Meyer, J. L., M. J. Sale, P. J. Mulholland, and N. L. Poff. 1999. Impacts of Climate Change on
Aquatic Ecosystem Functioning and Health. Journal of the American Water Resources
Association (JAWRA). 35 (6): 1373-1386.
Middelkoop, H., K. Daamen, D. Gellens, W. Grabs, J. C. J. Kwadijk, H. Lang, B. W. A. H.
Parmet, B. Schadler, J. Schulla, and K. Wilke. 2001. Impact of Climate Change on
Hydrological Regimes and Water Resources Management in the Rhine Basin. Climatic
Change. 49 (1-2): 105-128.
*Millennium Ecosystem Assessment (MEA). 2005a. Ecosystems and Human Wellbeing:
Wetlands and Water - Synthesis. Washington, DC: World Resources Institute.
*MEA [Agardy, T., J. Alder; Lead Authors: P. Dayton, S. Curran, A. Kitchingman, M. Wilson,
A. Catenazzi, J. Restrepo, C. Birkeland, C. Blaber, S. Saifullah, G. Branch, D. Boersma,
S. Nixon, P. Dugan, N. Davidson, and C. Vorosmarty]. 2005b. Chapter 19: Coastal
Systems. In: Ecosystems and Human Well-being: Current State and Trends, Volume 1,
eds. R. Hassan, R. Scholes, andN. Ash. Washington, DC: Island Press.
Do Not Cite or Quote Page A-6
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
*MEA [Finlayson, C. M., R. D'Cruz; Lead Authors: N. Aladin, D. R. Barker, G. Beltram, J.
Brouwer, N. Davidson, L. Duker, W. Junk, M. D. Kaplowitz, H. Ketelaars, E. Kreuzberg-
Mukhina, G. de la Lanza Espino, C. Leveque, A. Lopez, R. G. Milton, P. Mirabzadeh, D.
Pritchard, C. Revenga, M. Rivera, A. S. Hussainy, M. Silvius, M. Steinkamp]. 2005c.
Chapter 20: Inland Water Systems. In: Ecosystems and Human Well-being: Current State
and Trends, Volume 1, eds. R. Hassan, R. Scholes, andN. Ash. Washington, DC: Island
Press.
Murdoch, P. S., J. S. Baron, and T. L. Miller. 2000. Potential Effects of Climate Change on
Surface-Water Quality in North America. Journal of the American Water Resources
Association (JAWRA). 36 (2): 347-366.
*National Assessment Synthesis Team. 2000a. Chapter 14: Potential Consequences of Climate
Variability and Change for the Water Resources of the United States. In: Climate Change
Impacts on the United States: The Potential Consequences of Climate Variability and
Change — Foundation Report. Washington, DC: US Global Change Research Program.
*National Assessment Synthesis Team. 2000b. Water Sector. In: Climate Change Impacts on the
United States: The Potential Consequences of Climate Variability and Change —
Overview Report. Washington, DC: US Global Change Research Program.
Nicholls, R. J., and F. M. J. Hoozemans. 1996. The Mediterranean: Vulnerability to Coastal
Implications of Climate Change. Ocean and Coastal Management. 31 (2-3): 105-132.
Ojima, D., L. Garcia, E. Elgaali, K. Miller, T. G. F. Kittel, and J. Lackett. 1999. Potential
Climate Change Impacts on Water Resources in the Great Plains. Journal of the
American Water Resources Association (JAWRA). 35 (6): 1443-1454.
Paerl, H. W., J. Dyble, P. H. Moisander, R. T. Noble, M. F. Piehler, J. L. Pinckney, T. F. Steppe,
L. Twomey, and L. M. Valdes. 2003. Microbial Indicators of Aquatic Ecosystem Change:
Current Applications to Eutrophication Studies. FEMSMicrobiology Ecology. 46 (3):
233-246.
Palmer, M. A., C. A. R. Liermann, C. Nilsson, M. Florke, J. Alcamo, P. S. Lake, and N. Bond.
2008. Climate Change and the World's River Basins: Anticipating Management Options.
Frontiers in Ecology and Environment. 6(2): 81-89.
*Poff, N. L., M. M. Brinson, and J. W. Day. 2002. Aquatic Ecosystems and Global Climate
Change: Potential Impacts on Inland Freshwater and Coastal Wetland Ecosystems in the
United States. Prepared for the Pew Center on Global Climate Change. Arlington, VA:
Pew Center on Global Climate Change. January 2002.
Roderick, M. L., and G. D. Farquhar. 2002. The Cause of Decreased Pan Evaporation over the
Past 50 Years. Science. 298 (5597): 1410-1411. 15 November 2002.
Sankarasubramanian, A., R. M. Vogel, and J. F. Limbrunner. 2001. Climate Elasticity of
Streamflow in the United States. Water Resources Research. 37 (6): 1771-1781.
Do Not Cite or Quote Page A-7
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Schmitt, C. V., K. E. Webster, J. M. Peckenham, A. L. Tolman, and J. L. McNelly. 2008.
Vulnerability of Surface Water Supplies in Maine to the 2001 Drought. Journal of the
New England Water Works Association. 122(2): 104-116.
Shen, Y., T. Oki, N. Utsumi, S. Kanae, andN. Hanasaki. 2008. Projection of Future World
Water Resources Under SRES Scenarios: Water Withdrawal. Hydrological Sciences-
Journal des Sciences Hydrologiques. 53 (1): 11-33.
Twilley, R. R., E. J. Barren, H. J. Gholz, M. A. Harwell, R. L. Miller, D. J. Reed, J. B. Rose, E.
H. Siemann, R. G. Wetzel, and R. J. Zimmerman. 2001. Confronting Climate Change in
the Gulf Coast Region: Prospects for Sustaining Our Ecological Heritage. Cambridge,
MA: Union of Concerned Scientists, and Washington, D.C.: Ecological Society of
America. October 2001.
U.S. Climate Change Science Program, The (CCSP) [Titus, J. G.; K. E. Anderson, D. R. Cahoon,
D. B. Gesch, S. K. Gill, B. T. Gutierrez, E. R. Thieler, and S. J. Williams]. 2008. Coastal
Elevations and Sensitivity to Sea Level Rise: A Focus on the Mid-Atlantic Region. A
report by the U.S. Climate Change Science Program and the Subcommittee on Global
Change Research. Public Review Draft for Synthesis and Assessment Product 4.1.
Washington D.C.: U.S. Environmental Protection Agency. February 2008.
United States Environmental Protection Agency (USEPA). 1995. The Great Lakes: An
Environmental Atlas and Resource Book. Toronto, Ontario, Canada: Government of
Canada, and Chicago, IL: United States Environmental Protection Agency, Great Lakes
National Program Office. Available online at:
http://www.epa.gov/glnpo/atlas/index.html. Accessed July 6, 2009.
USEPA. 2002. State of the Waters 2002 Region 5. EPA 905-R-02-007. Chicago, IL: United
States Environmental Protection Agency, Region 5. September 2002.
USEPA. 2006a. National Estuary Program: Coastal Condition Report. EPA 842-B-06-001.
Washington, DC: U.S. Environmental Protection Agency. June 2007.
*USEPA. 2006b. Wadeable Streams Assessment: A Collaborative Survey of the Nation's
Streams. EPA 841-B-06-002. Washington, DC: United States Environmental Protection
Agency, Office of Water. December 2006.
*USEPA. 2008a. 2011 National Wetland Condition Assessment: Candidate Indicators of
Ecological Condition. Unpublished Preliminary Rough Draft.
*USEPA. 2008b. Chapter 3: Water. In: U.S. EPA's 2008 Report on the Environment. Final
Report. EPA 600-R-07-045F (NTIS PB2008-112484). Washington, DC: U.S.
Environmental Protection Agency. May 2008.
USEPA. 2008c. National Water Program Strategy: Response to Climate Change. EPA 800-R-08-
001. Washington, DC: United States Environmental Protection Agency, Office of Water.
September 2008.
Do Not Cite or Quote Page A-8
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
USEPA. 2008d. The Analysis of Regulated Contaminant Occurrence Data from Public Water
Systems in Support of the Second Six-Year Review of National Primary Drinking Water
Regulations. Unpublished Draft.
*United States Geological Survey (USGS). 1999. The Quality of Our Nation's Waters-Nutrients
and Pesticides. National Water-Quality Assessment Program. Circular 1225. Reston, VA:
United States Geological Survey (USGS).
United States Government Accountability Office (USGAO). 2000. Water Quality: Key EPA and
State Decisions Limited by Inconsistent and Incomplete Data. Report to the Chairman,
Subcommittee on Water Resources and Environment, Committee on Transportation and
Infrastructure, House of Representatives. Washington, D.C.: United States Government
Accountability Office. GAO/RCED-00-54. March 2000.
USGAO. 2002. Water Quality: Inconsistent State Approaches Complicate Nation's Efforts to
Identify Its Most Polluted Waters. Report to Congressional Requesters. GAO-02-186.
Washington, D.C.: United States Government Accountability Office. January 2002.
USGAO. 2004. Watershed Management: Better Coordination of Data Collection Efforts Needed
to Support Key Decisions. Report to the Chairman, Subcommittee on Water Resources
and Environment, Committee on Transportation and Infrastructure, House of
Representatives. GAO-04-382. Washington, D.C.: United States Government
Accountability Office. June 2004.
USGAO. 2005. Status of Federal Data Programs That Support Ecological Indicators. GAO-05-
376. Washington, D.C.: United States Government Accountability Office. September
2005.
Vincent, W. F., and R. Pienitz. 2006. Vulnerability of Northern Lake Ecosystems to Climate
Change. In: Workshop: Vulnerability of Cryospheric and Socio-Economic Systems (Feb.
26-28, 2006), Session 6: Vulnerability of Northern Social Systems. Peter Wall Institute
for Advanced Studies. February 2006.
Vorosmarty, C. J., P. Green, J. Salisbury, andR. B. Lammers. 2000. Global Water Resources:
Vulnerability from Climate Change and Population Growth. Science. 289 (5477): 284-
288.14 July 2000.
Wuebbles, D. J., and K. Hayhoe. 2002. Climate Change: A Real Issue with Real Concerns for
the Midwest. Proceedings of the International Conference on Climate Change and
Environmental Policy, November 11-12, 2002. Urbana, 111.
Yamin, F., A. Rahman, and S. Huq. 2005. Vulnerability, Adaptation and Climate Disasters: A
Conceptual Overview. IDS Bulletin. 36 (4): 1-14.
Do Not Cite or Quote Page A-9
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
*Zogorski, J. S., J. M. Carter, T. Ivahnenko, W. W. Lapham, M. J. Moran, B. L. Rowe, P. J.
Squillace, and P. L. Toccalino. 2006. The Quality of Our Nation's Waters - Volatile
Organic Compounds in the Nation's Ground Water and Drinking-Water Supply Wells.
National Water-Quality Assessment Program. Circular 1292. Reston, VA: United States
Geological Survey (USGS).
Do Not Cite or Quote Page A-10
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Do Not Cite or Quote Page B-l
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
The following table lists the 623 indicators gathered from a review of the 86 documents cited in Appendix A. Each indicator was
randomly assigned an Indicator ID#. This unique identifier serves as an easy way to identify and refer to the indicator throughout the
report. Indicator definitions included in this table were obtained, when possible, from the literature source that identified the indicator.
(Note: Some text may be verbatim from the source.). Definitions for some indicators were revised when the data used were different
(e.g., more recent) than those cited by the literature. These revised definitions are marked with * in the Indicator Definition column.
The references cited in the Literature Source column refer to the literature source from which the indicator was obtained. Full citations
for these references can be found in Appendix A. Some indicators from this list were determined to be duplicates; for these, the ID# of
the duplicate indicator is listed in the Duplicate Indicator column. After selection of one indicator for further assessment from each
duplicate group, the remaining duplicates were eliminated. Selection of indicators from groups of duplicates was based on
completeness of the indicator definition, national (as opposed to regional or local) focus of the literature source, etc. Eliminated
duplicate indicators are marked with an x next to the indicator ID#.
The 53 indicators identified as vulnerability indicators in the report appear in boldface font in this table. (The remaining 559 indicators
were considered to be state variables). Of these, the 25 example indicators that were mapped have a ** next to the indicator ID#.
Indicator
ID#
1**
2
Indicator
Acid Neutralizing Capacity
(ANC)
Agricultural Inputs -
Durable Goods (Units of
durable goods per unit of
output)
Definition
Percent of streams with low acid-neutralizing capacity (ANC) i.e., below
100 milliequivalents/liter. ANC is a measure of the water's ability to
buffer additional acid deposition or drainage from acid mines."1
Tractors are an example of durable goods. This indicator reports the
amount of inputs used to produce one unit of output, with 1975 as the
base year. For example, all fertilizers used on U.S. farms were divided by all
agricultural outputs — even if different amounts of fertilizer were used to
produce each commodity. So, for any input, the index value for a given year
describes whether more or less of that input was used to produce a unit of
output in that year than in 1975.
Literature Source
[See Appendix A for full
citation]
USEPA, 2006b.
Heinz Center, 2002; Heinz
Center, 2008
Duplicate
Do Not Cite or Quote
Page B-2
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
Indicator
Definition
Literature Source
[See Appendix A for full
citation]
Duplicate
Agricultural Inputs -
Energy (Units of energy
per unit of output)
This indicator reports the amount of energy inputs used to produce one
unit of output, with 1975 as the base year. For example, all fertilizers used
on U.S. farms were divided by all agricultural outputs — even if different
amounts of fertilizer were used to produce each commodity. So, for any
input, the index value for a given year describes whether more or less of
that input was used to produce a unit of output in that year than in 1975.
Heinz Center, 2002; Heinz
Center, 2008
Agricultural Inputs -
Fertilizers (Units of
fertilizers per unit of
output)
This indicator reports the amount of fertilizer inputs used to produce one
unit of output, with 1975 as the base year. For example, all fertilizers used
on U.S. farms were divided by all agricultural outputs — even if different
amounts of fertilizer were used to produce each commodity. So, for any
input, the index value for a given year describes whether more or less of
that input was used to produce a unit of output in that year than in 1975.
Heinz Center, 2002; Heinz
Center, 2008
Agricultural Inputs- Labor
(Units of labor per unit of
output)
This indicator reports the amount of labor inputs used to produce one unit
of output, with 1975 as the base year. For example, all fertilizers used on
U.S. farms were divided by all agricultural outputs — even if different
amounts of fertilizer were used to produce each commodity. So, for any
input, the index value for a given year describes whether more or less of
that input was used to produce a unit of output in that year than in 1975.
Heinz Center, 2002; Heinz
Center, 2008
Agricultural Inputs - Land
(Units of land per unit of
output)
This indicator reports the amount of land inputs used to produce one unit
of output, with 1975 as the base year. For example, all fertilizers used on
U.S. farms were divided by all agricultural outputs — even if different
amounts of fertilizer were used to produce each commodity. So, for any
input, the index value for a given year describes whether more or less of
that input was used to produce a unit of output in that year than in 1975.
Heinz Center, 2002; Heinz
Center, 2008
Do Not Cite or Quote
Page B-3
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
Indicator
Definition
Literature Source
[See Appendix A for full
citation]
Duplicate
Agricultural Inputs -
Pesticides (Units of
pesticides per unit of
output)
This indicator reports the amount of pesticide inputs used to produce one
unit of output, with 1975 as the base year. For example, all fertilizers used
on U.S. farms were divided by all agricultural outputs — even if different
amounts of fertilizer were used to produce each commodity. So, for any
input, the index value for a given year describes whether more or less of
that input was used to produce a unit of output in that year than in 1975.
Heinz Center, 2002; Heinz
Center, 2008
Agricultural Outputs -
Crops (Units of output per
year)
The indicator reports agricultural outputs over time, with 1975 as the base
year.
Heinz Center, 2002; Heinz
Center, 2008
Agricultural Outputs -
Meat, Dairy, Eggs, and
Other Products (Units of
output per year)
The indicator reports U.S. agricultural outputs over time, with 1975 as the
base year.
Heinz Center, 2002; Heinz
Center, 2008
10
Agricultural Outputs-
Total (Units of output per
year)
The indicator reports U.S. agricultural outputs over time, with 1975 as the
base year.
Heinz Center, 2002; Heinz
Center, 2008
11
Agricultural products
(economic production)
This indicator reports the production of food and fiber and the withdrawals
of water (agricultural products), using an index with 1980 as the base year.
Heinz Center, 2002; Heinz
Center, 2008
12
Agricultural water use
share
Agricultural sector withdrawals (QWag) as a share of total average annual
withdrawals. Method of calculation: QWag/QW
Hurdetal., 1998.
13
Air Quality - High Ozone
Levels: At least 1 day per
year (Percent of
urban/suburban air
monitoring stations with 1
day exceedance)
This indicator reports the percentage of air pollution monitoring stations in
urban and suburban areas with "high" ozone concentrations at least 1 day
a year. Ground-level ozone is considered high when the 8-hour average
concentration exceeds 0.08 parts per million (ppm).
Heinz Center, 2002; Heinz
Center, 2008
Do Not Cite or Quote
Page B-4
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
Indicator
Definition
Literature Source
[See Appendix A for full
citation]
Duplicate
14
Air Quality - High Ozone
Levels: At least 2 days per
year (Percent of
urban/suburban air
monitoring stations with 2
day exceedance)
This indicator reports the percentage of air pollution monitoring stations in
urban and suburban areas with "high" ozone concentrations at least 2 days
a year. Ground-level ozone is considered high when the 8-hour average
concentration exceeds 0.08 parts per million (ppm).
Heinz Center, 2002; Heinz
Center, 2008
15
Air Quality - High Ozone
Levels: At least 3 days per
year (Percent of
urban/suburban air
monitoring stations with 3
day exceedance)
This indicator reports the percentage of air pollution monitoring stations in
urban and suburban areas with "high" ozone concentrations at least 3 days
a year. Ground-level ozone is considered high when the 8-hour average
concentration exceeds 0.08 parts per million (ppm).
Heinz Center, 2002; Heinz
Center, 2008
16
Air Quality - High Ozone
Levels: At least 4 days per
year (Percent of
urban/suburban air
monitoring stations with 4
day exceedance)
This indicator reports the percentage of air pollution monitoring stations in
urban and suburban areas with "high" ozone concentrations at least 4 days
a year. Ground-level ozone is considered high when the 8-hour average
concentration exceeds 0.08 parts per million (ppm).
Heinz Center, 2002; Heinz
Center, 2008
17
Altered Freshwater
Ecosystems (percent
miles changed)
This indicator of alteration reports the percentage of:
Heinz Center, 2002; Heinz
Center, 2008
18
Ambient toxicity (chemical
concentration)
Metals, pesticides, PCBs, and organic contaminants.
USEPA, 2006a.
19
Animal Deaths and
Deformities (events)
This indicator reports on unusual mortality events for waterfowl, fish,
amphibians, and mammals, and on deformity events for amphibians. Only
data on waterfowl mortality can be reported at this time.
Heinz Center, 2002; Heinz
Center, 2008
20
Aquatic life mobility
N/A
ME A, 2005c.
Do Not Cite or Quote
Page B-5
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
Indicator
Definition
Literature Source
[See Appendix A for full
citation]
Duplicate
21
Areas with depleted
oxygen (percent monthly
exposure)
The percentage of brackish water exposed to a range of oxygen
concentrations for at least 1 month will be reported as anoxic (no oxygen),
hypoxic (>0 and <2 parts per million [ppm]), low (2-4 ppm), or sufficient
(>4ppm).
Heinz Center, 2002; Heinz
Center, 2008
22*
At-Risk Freshwater Plant
Communities (% area at
risk)
This indicator reports on the percentage of wetland and riparian plant
communities that are at risk of extinction. These status ranks are based
on such factors as the remaining number and condition of occurrences of
the community, the remaining acreage, and the severity of threats to the
community type.*
Heinz Center, 2002; Heinz
Center, 2008
467
23
At-Risk Native Forest
Species (Percent of all
forest species that are at
risk)
This indicator reports on the relative risk of extinction of native forest
species. The risk categories are based on such factors as the number and
condition of individuals and populations, the area occupied by the species,
population trends, and known threats. Degrees of risk reported here range
from very high ("critically imperiled" species are often found in five or
fewer places or have experienced very steep declines) to moderate
("vulnerable" species are often found in fewer than 80 places or have
recently experienced widespread declines). In all cases, a wide variety of
factors contribute to the overall ratings. "Forest species" live in forests
during at least part of their life and depend on forest habitats for survival.
Heinz Center, 2002; Heinz
Center, 2008
24*
At-Risk Native Freshwater
Species (relative rank)
This indicator reports on percentage of native freshwater species that are
at risk of extinction. The risk categories are based on such factors as the
number and condition of individuals and populations, the area occupied
by the species, population trends, and known threats.*
Heinz Center, 2002; Heinz
Center, 2008
Do Not Cite or Quote
Page B-6
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
Indicator
Definition
Literature Source
[See Appendix A for full
citation]
Duplicate
25
At-Risk Native Grassland
and Shrubland Species -
By Category (Percent of all
at-risk species in a certain
category)
Categories include: Extinct, Critically Imperiled, Imperiled, Vulnerable, and
All At-Risk. This indicator reports on the status of native grassland and
shrubland species with respect to their relative risk of extinction. These
status ranks are based on multiple factors: the number and condition of
individuals and populations, the area occupied by the species, population
trends, and known threats. Degrees of risk reported here range from very
high ("critically imperiled" species often are found in five or fewer places or
have experienced very steep declines) to moderate ("vulnerable" species
often are found in fewer than 80 places or have recently experienced
widespread declines). In all cases, a wide variety of factors contribute to
overall ratings. "Grassland and shrubland species" live in these habitats
during at least part of their life cycle and depend on them for survival.
Heinz Center, 2002; Heinz
Center, 2008
26
At-Risk Native Grassland
and Shrubland Species -
By Region (Percent of all
at-risk species in a certain
region)
Regions include: Northeast/Mid-Atlantic, Southeast, Midwest, Southwest,
Rocky Mountain, Pacific Coast, and Hawaii. This indicator reports on the
status of native grassland and shrubland species with respect to their
relative risk of extinction. These status ranks are based on multiple factors:
the number and condition of individuals and populations, the area
occupied by the species, population trends, and known threats. Degrees of
risk reported here range from very high ("critically imperiled" species often
are found in five or fewer places or have experienced very steep declines)
to moderate ("vulnerable" species often are found in fewer than 80 places
or have recently experienced widespread declines). In all cases, a wide
variety of factors contribute to overall ratings. "Grassland and shrubland
species" live in these habitats during at least part of their life cycle and
depend on them for survival.
Heinz Center, 2002; Heinz
Center, 2008
27
At-Risk native marine
species (relative risk)
Relative risk of extinction of native marine species, both plants and
animals. The risk categories are based on such factors as the number and
condition of individuals and populations, the area occupied by the
species, population trends, and known threats.
Heinz Center, 2002; Heinz
Center, 2008
28
At-Risk native species
(relative rank)
This indicator reports on the relative risk of extinction of native plant and
animal species.
Heinz Center, 2002; Heinz
Center, 2008
Do Not Cite or Quote
Page B- 7
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
29
30
31
32X
33X
34
35
36X
37
Indicator
Bay grasses
Beach closings (driven by
bacterial contamination)
Benthic Index (several)
Benthic
Macroinvertebrates
Benthic organisms
(abundance, diversity)
Bottom habitat (diversity,
abundance, biomass)
Carbon Storage - Forests
(Weight of carbon stored
overtime)
Carbon Storage -
Grasslands/Shrublands
(Weight of carbon stored
overtime)
Chemical contaminants
(exceedence of regulatory
value)
Definition
N/A
Measure of bacterial contamination.
Shannon-Weiner Diversity Index, and other Indices.
Benthic communities are largely composed of macroinvertebrates, such as
annelids, mollusks, and crustaceans. These organisms inhabit the bottom
substrates of estuaries and play a vital role in maintaining sediment and
water quality. They also are an important food source for bottom-feeding
fish, invertebrates, and birds.
Benthic abundance, species richness/diversity.
Attainment of the benthic restoration goal was determined by examining:
benthic biodiversity measures, measures of assemblage abundance and
biomass, life history strategy measures, activity beneath the sediment
surface, and feeding guild measures.
This indicator reports how much carbon— an essential component of all
organisms— is stored in forests, including trees, soil, and plant litter on the
forest floor, and in wood products.
This indicator will report the total amount of carbon stored in soil and
plants in grasslands and shrublands.
Metals, PCBs, tributyltin, and priority organics found exceeding total
maximum daily loads (TMDLs).
Literature Source
[See Appendix A for full
citation]
Chesapeake Bay Program,
2008.
USEPA, 1995.
USEPA, 2006a.
USEPA, 2008b.
Hayslip etal., 2006.
Chesapeake Bay Program,
2008.
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Chesapeake Bay Program,
2008.
Duplicate
461, 33
32,461
36,617
35,617
Do Not Cite or Quote
Page B-&
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
Indicator
Definition
Literature Source
[See Appendix A for full
citation]
Duplicate
38
Chemical Contamination
in Urban Streams -
Contaminant Occurrence
(Number of contaminants
detected)
This indicator reports on contaminants found in urban and suburban
streams. Compounds reported here include many pesticides, select
pesticide breakdown products, ammonia, and nitrate (because nitrate and
ammonia occur naturally, they are not included in the graphs showing
contaminant occurrence).
Heinz Center, 2002; Heinz
Center, 2008
39
Chemical Contamination
in Urban Streams-
Contaminant
Concentrations above
Standards or Guidelines
(Percent sites with
exceedances)
This indicator reports on contaminants found in urban and suburban
streams. Compounds reported here include many pesticides, select
pesticide breakdown products, ammonia, and nitrate (because nitrate and
ammonia occur naturally, they are not included in the graphs showing
contaminant occurrence).
Heinz Center, 2002; Heinz
Center, 2008
40
Chemical Contamination
in Urban/Suburban Soils
This indicator reports on contaminants found in urban and suburban soils.
Compounds reported here include many pesticides, selected pesticide
breakdown products, ammonia, and nitrate (because nitrate and ammonia
occur naturally, they are not included in the graphs showing contaminant
occurrence).
Heinz Center, 2002; Heinz
Center, 2008
41*
Chlorophyll a
N/A
Chesapeake Bay Program,
2008.
42
42
Chlorophyll a (surface
concentration)
Good: Surface concentrations are less than 5 ng/L (less than 0.5 ng/L for
tropical ecosystems). Fair: Surface concentrations are between 5 ng/L and
20 ng/L (between 0.5 ng/L and 1 ng/L for tropical ecosystems). Poor:
Surface concentrations are greater than 20 |jg/L (greater than 1 ng/L for
tropical ecosystems).
USEPA, 2006a.
41
43
Chlorophyll
concentrations (within 25
miles of shore)
Chlorophyll concentration in estuaries and ocean waters within 25 miles of
shore. For ocean waters, the indicator reports the average value for the
season with the highest concentration, for each region. For estuaries, the
indicator reports the percentage of area in three ranges: below 5 parts per
billion (ppb), between 5 and 20 ppb, and above 20 ppb, using data for the
season with the highest average concentration.
Heinz Center, 2002; Heinz
Center, 2008
Do Not Cite or Quote
Page B-9
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
Indicator
Definition
Literature Source
[See Appendix A for full
citation]
Duplicate
44
Climate, storm, and
extreme event variability
Climate fluctuations, mostly related to flood and drought events.
Gleick and Adams, 2000.
45
Coastal Bird Populations
(2 indicator species)
Indicators consist of two birds (piping plover and least tern) that inhabit
Long Island Sound beaches, plus wading birds that forage in tidal marshes.
Long Island Sound Study,
2008.
46
46A
Coastal birds
N/A
MEA, 2005b.
45
47
Coastal erosion (managed
vs. unmanaged area)
How much of the U.S. coast is managed in an attempt to control erosion
and how much remains in a "natural" state, with no erosion control. For
unmanaged areas, the indicator reports what fraction is eroding, accreting
(gaining land area), or stable.
Heinz Center, 2002; Heinz
Center, 2008
48A
Coastal Fish Tissue
Contaminants
N/A
USEPA, 2008b.
58, 99,
579
49
Coastal land loss
N/A
Twilley et al., 2001.
50A
Coastal Sediment Quality
N/A
USEPA, 2008b.
250
51*
Coastal Vulnerability
Index (to sea level rise)
Index of coastal ecosystem vulnerability to sea level rise, he index allows
the six physical variables to be related in a quantifiable manner that
expresses the relative vulnerability of the coast to physical changes due
to sea-level rise. This method yields numerical data that cannot be
equated directly with particular physical effects. It does, however,
highlight those regions where the various effects of sea-level rise might
be the greatest. The six variables are: a = Geomorphology; b = Coastal
Slope (%); c = Relative sea-level change (mm/year); d = Shoreline
erosion/accretion (m/year); e = Mean tide average (m); e = Mean wave
height (m). Once each section of coastline is assigned a risk value based
on each specific data variable, the coastal vulnerability index is calculated
as the square root of the geometric mean, or the square root of the
product of the ranked variables divided by the total number of variables
as: CVI = [(a*b*c*d*e*f*)/6)]Al/2. *
Day eta I.,2005.
Do Not Cite or Quote
Page B-10
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
Indicator
Definition
Literature Source
[See Appendix A for full
citation]
Duplicate
52
Coastal wetlands
(extent/acreage)
Acreage of coastal habitats whose defining feature is that they are
composed of living organisms (such as seagrasses, mangrove forests, and
coastal wetlands) or are built by them (such as coral reefs or shellfish
beds).
Heinz Center, 2002; Heinz
Center, 2008
53
53A
Coasts and Oceans
(extent/acreage)
This indicator presents the area coastal land as a percentage of the total
U.S. land area, for the most recent 50-year period and compared to
presettlement estimates.
Heinz Center, 2002; Heinz
Center, 2008
52
54
Commercial fish and
shellfish landings (weight)
Reports the weight offish, shellfish, and other products taken from U.S.
waters. Landings, plus certain aquaculture harvests, are shown for five
regions that cover all waters out to the 200-mile territorial limit.
Heinz Center, 2002; Heinz
Center, 2008
55
Commercially important
fish stocks (size)
Tracks the percentage of commercially important fish species, or "stocks,"
that are increasing or decreasing in size. Only stocks whose population
increased or decreased by at least 25% are reported. Trends are based on
the estimated weight, or "biomass," of the entire stock.
Heinz Center, 2002; Heinz
Center, 2008
56
Condition of bottom-
dwelling animals (percent
area inhabited)
Describes the condition of worms, clams, snails, and shrimplike animals in
bottom sediments ("benthic communities") by reporting the percentage of
area in which these communities are in "undegraded," "moderate," and
"degraded" condition. The index reflects changes in benthic community
diversity and the abundance of pollution-tolerant and pollution-sensitive
species. A low benthic index rating indicates that the benthic communities
are less diverse than expected, are populated by more than expected
pollution-tolerant species, and contain fewer than expected pollution-
sensitive species. The data in this report reflect an assessment of benthic
communities as "good" (high index score), "fair" (moderate index score), or
"poor" (low index score).
Heinz Center, 2002; Heinz
Center, 2008
57
Constructed Materials-
30% or Greater Area
Covered by Constructed
Materials (area)
This indicator would report on the percentage of land area covered by 30%
or more constructed materials.
Heinz Center, 2002; Heinz
Center, 2008
Do Not Cite or Quote
Page B-11
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
Indicator
Definition
Literature Source
[See Appendix A for full
citation]
Duplicate
58
Contaminants in fish and
shellfish (chemical
concentration)
Measures the concentration of PCBs, mercury, and DDT in the edible tissue
of seafood from U.S. coastal waters.
Heinz Center, 2002; Heinz
Center, 2008
48, 99,
579
59
Contamination in bottom
sediments (concentration)
Information on the concentration, in coastal bottom sediments, of four
major classes of contaminants that can harm fish and other aquatic
organisms and can adversely affect human health if ingested while
consuming fish or shellfish.
Heinz Center, 2002; Heinz
Center, 2008
60
Cropland (total area of
land used for crops)
This indicator reports the amount of land used for crops, including pasture
and hay. Acreage that is enrolled in long-term set-aside programs, such as
the Conservation Reserve Program (CRP) is not considered to be part of this
indicator.
Heinz Center, 2002; Heinz
Center, 2008
61
Croplands
(extent/acreage)
This indicator presents the area of croplands as a percentage of the total
U.S. land area, for the most recent 50-year period and compared to
presettlement estimates.
Heinz Center, 2002; Heinz
Center, 2008
62
Cropped Land (area,
ecosystem condition)
This indicator would report on the percentage of land area that is cropped
land (not including interspersed natural areas).
Heinz Center, 2002; Heinz
Center, 2008
63
Delta accretion rate
The rate of sediment accretion in a river delta.
Day et al., 2008.
64
Depth to Shallow
Groundwater (Percent of
shallow groundwater with
a certain depth)
This indicator will describe the depth to shallow groundwater in grassland
and shrubland areas. Specifically, it will report the percentage of grassland
and shrubland areas where the depth to groundwater falls within several
ranges (less than 5 feet, 5 to 10 feet, 10 to 20 feet and more than 20 feet).
Heinz Center, 2002; Heinz
Center, 2008
65
Disruptive Species - By
Region (Number of
disruptive species by
region)
This indicator would report the number and type of "disruptive" species
found in metropolitan areas. Disruptive species are those that have
negative effects on natural areas and native species or cause damage to
people and property. This indicator would report the number of disruptive
native and non-native plant and animal species on a regional basis, for the
most current year.
Heinz Center, 2002; Heinz
Center, 2008
Do Not Cite or Quote
Page B-12
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
66
67X
68
69
70
71X
72x
73X
Indicator
Disruptive Species in
Metropolitan Areas
(Number of disruptive
species over time)
Dissolved inorganic
nitrogen
Dissolved Inorganic
Nitrogen (DIN) (surface
concentration)
Dissolved Inorganic
Phosphorus (DIP) (surface
concentration)
Dissolved Organic Carbon
Dissolved oxygen
Dissolved oxygen
Dissolved oxygen
Definition
This indicator would report the number and type of "disruptive" species
found in metropolitan areas. Disruptive species are those that have
negative effects on natural areas and native species or cause damage to
people and property. Specifically, the indicator will report the number of
larger metropolitan areas with 5 or fewer, from 6 to 10, from 11 to 20, and
more than 20 disruptive plant and animal species.
N/A
Good: Surface concentrations are less than 0.1 mg/L (NE, SE, Gulf), 0.5
mg/L (West), or 0.05 mg/L (tropical). Fair: Surface concentrations are 0.1-
0.5 mg/L (NE, SE, Gulf), 0.5-1.0 mg/L (West), or 0.05-0.1 mg/L (tropical).
Poor: Surface concentrations are greater than 0.5 mg/L (NE, SE, Gulf), 1.0
mg/L (West), or 0.1 mg/L (tropical).
Good: Surface concentrations are less than 0.01 mg/L (NE, SE, Gulf), 0.01
mg/L (West), or 0.005 mg/L (tropical). Fair: Surface concentrations are
0.01-0.05 mg/L (NE, SE, Gulf), 0.01-0.1 mg/L (West), or 0.005-0.01 mg/L
(tropical). Poor: Surface concentrations are greater than 0.05 mg/L (NE, SE,
Gulf), 0.1 mg/L (West), or 0.01 mg/L (tropical).
N/A
Good: Concentrations are greater than 5 mg/L. Fair: Concentrations are
between 2 mg/L and 5 mg/L. Poor: Concentrations are less than 2 mg/L.
N/A
N/A
Literature Source
[See Appendix A for full
citation]
Heinz Center, 2002; Heinz
Center, 2008
Hayslipetal., 2006.
USEPA, 2006a.
USEPA, 2006a.
Lettenmaier et al., 2008.
USEPA, 2006a.
Chesapeake Bay Program,
2008.
Hayslip etal., 2006.
Duplicate
68
67
72, 73,
74, 75,
337, 131
71, 73,
74, 75,
337, 131
71, 72,
74, 75,
337, 131
Do Not Cite or Quote
Page B-13
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
74
75*
76X
77
78
79
80
81
82
83
Indicator
Dissolved Oxygen
Dissolved oxygen
concentration
Dissolved solids (total)
Drained or Impounded
Wetlands (area,
ecosystem condition)
Drought events (severity)
Dryness Ratio (ratio of
precipitation/evapotransp
iration)
Duration of Dry Periods in
Grassland/Shrubland
Streams and Rivers
(Percent of streams with
substantially
shorter/longer dry-
periods overtime)
Ecosystem extent
(classification of area)
Ecosystem heat sensitivity
Ecosystem ice cover
sensitivity
Definition
Percent of observations of ambient concentrations less than 5 mg/L
N/A
N/A
This indicator would reports the percentage of land area drained or
impounded wetlands (areas that remain wetlands but have been highly
altered).
N/A
Share of total average annual precipitation (P) that is lost through
evapotranspiration (ET), where ET is defined as P-QS.
Duration of dry-period compared to a 50-year average). The indicator
tracks the frequency and duration of zero-flow conditions for streams and
rivers in grassland/shrubland regions. It reports the percentage of streams
and rivers that have at least one no-flow day per year, and the percentage
where the duration of zero-flow periods for a given period is substantially
longer or shorter than the long-term (50-year) average.
Type of ecosystem.
The average annual number of days with maximum temperatures
exceeding 90°F (32°C). *
The average annual number of days with average temperatures below 32°F
(0°C). *
Literature Source
[See Appendix A for full
citation]
Hurdetal., 1999.
USEPA, 1995.
USEPA, 2006b.
Heinz Center, 2002; Heinz
Center, 2008
Lettenmaier et al., 2008.
Hurdetal., 1999.
Heinz Center, 2002; Heinz
Center, 2008
U.S. Climate Change
Science Program, 2008.
Hurdetal., 1998.
Hurdetal., 1998.
Duplicate
71, 72,
73, 75,
337, 131
71, 72,
73, 74,
337, 131
345
Do Not Cite or Quote
Page B-14
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
84
85
86
87X
88
89
90
Indicator
Electrical Conductivity
Estuarine Waters
Contamination (chemical
occurrence)
Evaporation and
Transpiration
Evaporation and
Transpiration
Expenditure on Dredging
Activities in Waterways
(economy)
Extreme temperatures
(number)
Extreme/heavy rainfall
events (number)
Definition
N/A
This indicator reports on contaminants found in estuarine waters.
Contaminants reported here include many pesticides, selected degradation
products, polychlorinated biphenyls (PCBs), polyaromatic hydrocarbons
(PAHs), volatile organic compounds, other industrial contaminants, trace
elements, nitrate, and ammonium. (Because nitrate, ammonium, and trace
elements such as cadmium and chromium occur naturally, they are not
included in the contaminant occurrence graphs).
N/A
N/A
Average annual expenditures on dredging activities in navigable
waterways.
Number of Threshold Exceedances per Year —Thresholds: Daily Maximum
Temperature of 97°F/36°C.
Number of 24-hour and 7-day intense rainfall events.
Literature Source
[See Appendix A for full
citation]
USEPA, 2006b.
Heinz Center, 2002; Heinz
Center, 2008
Lettenmaier et al., 2008.
Gleick and Adams, 2000.
Hurdetal., 1999.
Klingetal., 2003 using
data from Wuebbles and
Hayhoe, 2003.
Klingetal., 2003 using
data from Wuebbles and
Hayhoe, 2003.
Duplicate
603
87
86
Do Not Cite or Quote
Page B-15
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
Indicator
Definition
Literature Source
[See Appendix A for full
citation]
Duplicate
91
Farmland Landscape
(Cropland as a percentage
of total farmland)
This indicator reports the percentage of the farmland landscape that is
actively used for crop production, pasture, or haylands. The "farmland
landscape" includes croplands and the forests or woodlots, wetlands,
grasslands and shrublands, and the like that surround or are intermingled
with them. This indicator describes the degree to which croplands
dominate the landscape, or, conversely, the degree to which these other
lands are intermingled. This indicator also describes the composition of the
noncropland portion of the farmland landscape by reporting the
percentage of these lands that are forests, grasslands and shrublands,
wetlands, developed areas, and other lands and waters. The noncropland
elements of the farmland landscape (other than developed) provide
wildlife habitat, serve as streamside buffers and windbreaks, and lend a
distinctive visual character to the landscape. (Pasture and haylands are
intermediate in character between "natural" grasslands and cultivated
croplands; for this indicator, they are counted as croplands.)
Heinz Center, 2002; Heinz
Center, 2008
92
Finfish (abundance,
biomass)
Indicators include abundance of popular sport fish (such as striped bass
and bluefish), fish biomass, and abundance of fish that spawn in rivers in
the Sound's watershed.
Long Island Sound Study,
2008.
93
Fire Frequency (Percent of
forest land burned over
time)
This indicator describes the frequency with which forests are burned by
wildfire. It would report the fraction of forest lands that experience wildfire
much more or less frequently, moderately more or less frequently, or with
about the same frequency as in presettlement times. Thus, a forest that,
historically, burned every 50 years on average will be considered
moderately altered if it burns every 100 years, and significantly altered if it
burns only every 150 years, and about the same if it burns once every 50
years.
Heinz Center, 2002; Heinz
Center, 2008
Do Not Cite or Quote
Page B-16
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
94
95
96
97
98
ggX
Indicator
Fire Frequency (Percent of
grasslands/shrublands by
area that are burned over
time)
Fish and Bottom-Dwelling
Animals (comparison to
baseline)
Fish and shellfish
populations (5 species)
Fish range distribution
(species)
Fish Tissue Contaminants
Index (exceedence of
regulatory value)
Fish-tissue contaminants
(chemical levels)
Definition
This indicator will describe how often grassland and shrublands are burned
by wildfire. Specifically, it will report the fraction of grassland and
shrubland areas that burn much more or less often, moderately more or
less often, or about as often as before European settlement. So, for
example, an area that historically burned every 5 years on average might
be considered moderately altered if it now burns every 10 years and
significantly altered if it now burns only every 25 years. An area that
historically burned every 80 years might be considered moderately altered
if it now burns every 40 years and significantly altered if it now burns every
20 years. (Presettlement conditions are used here as a reference against
which to compare current conditions, not as an implied management goal).
This indicator reports on "biological integrity" — the degree to which the
suite of fish and bottom-dwelling animals in a lake or stream resembles
what one might find in a relatively undisturbed lake or stream in the
same region. Tests assess the number of different species, the number
and condition of individuals, and food chain interactions for fish and
bottom-dwelling (or benthic) animals, which include insects, worms,
mollusks, and crustaceans.
Blue Crab, Oyster, Striped Bass, Shad, and Juvenile Menhaden population
characteristics.
Northern/southern limit.
Good: Concentrations of all chemical contaminants fall below the range of
the EPA Advisory Guidance. Fair: Concentration of at least one chemical
contaminant falls within the range of the EPA Advisory Guidance. Poor:
Concentrations of at least one chemical contaminant exceeds the
maximum value in the range of the EPA Advisory Guidance.
Inorganic arsenic, cadmium, lead, mercury, selenium, silver, zinc, DDT.
Literature Source
[See Appendix A for full
citation]
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Chesapeake Bay Program,
2008.
Kling et al., 2003.
US EPA, 2006a.
Hayslipetal., 2006.
Duplicate
58, 48,
579
Do Not Cite or Quote
Page B-l 7
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
100
101
102
103
104
105
Indicator
Flood events (frequency)
Forest Age (Percent of
forest lands by age group)
Forest Area and
Ownership (Area of forest
lands)
Forest Community Types
with Significantly Reduced
Area (Area occupied by
community type)
Forest Disturbance: Fire,
Insects, Disease (Area
affected by disturbance)
Forest Management
Categories (Change in
percent of forested area
with time)
Definition
N/A
This indicator reports the percentage of forest lands with stands in several
age classes.
This indicator reports how much forest land there is in the United States
and who owns it.
This indicator would report whether those forest community types that
cover significantly fewer acres than they did in presettlement times are
increasing or decreasing in area, and by how much. It would also report the
total area occupied by these much-reduced forest community types— those
that have been reduced by 70% or more in area.
This indicator reports the acreage of forest affected each year by several
important types of disturbance: forest fires, insects, and diseases of trees.
This indicator reports the percentage of forest area in several different
management categories. These range from "reserved lands" (forests in
national parks, wilderness areas, and other similar areas) to forests under
intensive management involving replanting after harvest. Other forest
lands in intermediate categories are subject to a wide variety of both
management practices and restrictions on use.
Literature Source
[See Appendix A for full
citation]
Lettenmaier et al., 2008.
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Duplicate
Do Not Cite or Quote
Page B-18
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
106
107
108
109
110
111
Indicator
Forest Pattern and
Fragmentation (Percent of
a tree's surroundings that
are forested)
Forest products
(economic production)
Forest Types (Area
covered by a certain
forest type)
Forests (extent/acreage)
Forests with Nursery
Stock (area, ecosystem
condition)
Fragmentation and
Landscape Pattern
Definition
This indicator describes a tree's forest neighborhood according to the
degree of forest cover within various distances. Thus, the "immediate
neighborhood" of a particular tree is everything within about 250 feet in all
directions. This "immediate neighborhood" is "mostly forest" if the land is
at least 90% forested. A tree's "local neighborhood" extends about 1/4
mile in all directions, and its "larger neighborhood" extends about 2 1/2
miles. This analysis relies upon computer analyses of satellite data on
millions of individual forest points. While these points (called "pixels") are
not individual trees— they are squares about 100 feet on a side— they serve
much the same purpose.
This indicator reports the production of food and fiber and the withdrawals
of water (forest products), using an index with 1980 as the base year.
This indicator reports the acreage of a variety of forest "cover types." Cover
types describe the dominant species of trees found in the forests (e.g.,
oak-hickory forests are dominated by oaks and hickories, but include other
kinds of trees as well).
This indicator presents the area of forests as a percentage of the total U.S.
land area, for the most recent 50-year period and compared to
presettlement estimates.
This indicator reports the percentage of land area that is comprised of
forests planted with nursery stock.
N/A
Literature Source
[See Appendix A for full
citation]
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Duplicate
Do Not Cite or Quote
Page B-19
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
112
113
114
115
116
117
118
Indicator
Fragmentation of
Farmland Landscapes by
Development (Ratio of
cropland: developed land)
Fresh Water Resources
Contamination (chemical
occurrence)
Fresh Waters
(extent/acreage)
Freshwater input to
coastal ecosystems
Freshwater Rivers and
Streams with Low Index
of Biological Integrity
(ecosystem condition)
Glaciers
Grassland and Shrubland
Areas (extent/acreage)
Definition
This indicator would report the degree to which suburban development
and other built-up areas break up (fragment) the farmland landscape
(defined as croplands plus intermingled "natural" areas such as forests,
wetlands, and grasslands and shrublands). Areas with a mosaic of cropland
and intermingled natural areas— but little or no development— would be
rated as "low" on the "fragmentation index" used for this indicator, while
those in which small patches of cropland are mixed into a backdrop of
suburban development would be rated as "high."
This indicator reports on contaminants found in freshwater resources.
Contaminants reported here include many pesticides, selected degradation
products, polychlorinated biphenyls (PCBs), polyaromatic hydrocarbons
(PAHs), volatile organic compounds, other industrial contaminants, trace
elements, nitrate, and ammonium. (Because nitrate, ammonium, and trace
elements such as cadmium and chromium occur naturally, they are not
included in the contaminant occurrence graphs).
This indicator presents the area of fresh waters.
Rate of freshwater input into coastal ecosystems.
This indicator would report on the percentage of freshwater rivers and
streams with low IBI (Index of Biological Integrity, a species-based
measure of disturbance).
N/A
This indicator presents the area of grasslands and shrublands as a
percentage of the total U.S. land area, for the most recent 50-year period
and compared to presettlement estimates.
Literature Source
[See Appendix A for full
citation]
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Day et al., 2008.
Heinz Center, 2002; Heinz
Center, 2008
Gleick and Adams, 2000.
Heinz Center, 2002; Heinz
Center, 2008
Duplicate
Do Not Cite or Quote
Page B-20
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
119
120
121
122
123
124X
125**
126
Indicator
Grassland Patches
(Percent of grassland
patches that cover a
certain area)
Grasslands and
Shrublands (Area covered
by grasslands/shrublands)
Groundwater Depletion -
Ratio of
Withdrawals/Baseflow
Groundwater Levels - Area
of Aquifer subject to
Change in Groundwater
Levels
Groundwater recharge
Groundwater recharge
Groundwater reliance
Growing season length
Definition
This indicator will describe the fraction of grassland area and shrubland
area that is in patches of different sizes. The total area occupied by patches
of a certain size will be reported as a percentage of the total area of either
grasslands or shrublands.
This indicator reports the acreage of U.S. grasslands and shrublands
(although data are not available for Hawaii).
Ratio of average groundwater withdrawals (QGW) in 1990 to annual
average baseflow (QBase), reflecting the extent that groundwater use
rates may be exceeding recharge.
This indicator would report the percentage of the area of the nation's
major regional aquifers in which water levels are increasing, decreasing, or
stable. The indicator would report what fraction of the aquifer area
declined, increased, or remained stable in comparison to a previous period,
and it would be reported every 5 years.
Rate of groundwater recharge (mm/year) based on WaterGAP Global
Hydrology Model and mapped on a 0.5 degree grid. It accounts for spatial
variation in precipitation, infiltration capacity, hydrogeology, topography,
and permafrost. *
N/A
Share of total annual withdrawals derived from groundwater in 1995.
Method of calculation: QGW/QW. *
Number of days between last spring frost and first autumn frost.
Literature Source
[See Appendix A for full
citation]
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Hurdetal., 1999.
Heinz Center, 2002; Heinz
Center, 2008
Lettenmaier et al., 2008.
Gleick and Adams, 2000.
Hurdetal., 1998.
Klingetal., 2003 using
data from Wuebbles and
Hayhoe, 2003.
Duplicate
124
123
Do Not Cite or Quote
Page B-21
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
127
128
129
130
13 lx
132
133
134X
135
136
Indicator
Harmful algal blooms
(occurrence)
Highest spring streamflow
date
Hydrologic isolation
Hydropower capacity
Hypoxia (dissolved oxygen
concentration)
Ice cover duration (freeze
and ice-out dates)
Ice cover on rivers
Ice -out date
Industrial water use
(average annual share
consumed)
Inland water extent
Definition
Defined as (1) an increase in the abundance of species that are known to
produce toxins harmful to marine animals or humans; (2) the occurrence
of lesions or mass mortalities of marine animals caused by HAB species;
and (3) the occurrence of human pathologies caused by HAB species. A
single event counts only once toward the relative intensity scale, even if it
produces multiple impacts (e.g., an increase in the abundance of a HAB
species that causes mass mortalities and an increased human health risk
will be counted as a single event).
Peak streamflow in early spring, indicative of maximum baseflow and/or
maximum snowmelt rate in a basin.
Indicates whether a system is a closed basin, or "isolated water" (i.e., a
single spring that flows for a short distance before re-infiltrating into
ground).
Distribution of regional hydroelectric power capacity, in megawatts (MW).
Levels of dissolved oxygen below 2 milligrams per liter (mg/L).
Number of days waterbodies are covered by ice, or are ice-free.
N/A
Date of spring ice-out on lakes in the Northeast.
Share of total industrial water use that is consumed (i.e., not returned to
the system). *
Total areal extent of inland waters (could be divided by area to calculate
density).
Literature Source
[See Appendix A for full
citation]
Heinz Center, 2002; Heinz
Center, 2008
Frumhoff etal., 2006.
MEA, 2005b.
Hurd et al., 1998.
USEPA, 2008b.
Kling etal., 2003.
Frumhoff etal., 2006.
Frumhoff etal., 2006.
Hurd etal., 1999.
MEA, 2005b.
Duplicate
71, 72,
73, 74,
75, 337
275, 423
Do Not Cite or Quote
Page B-22
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
137
138
139
140
141
142
143X
144
145
146
Indicator
Institutional Barriers to
Water Trading
In-stream fish habitat
Intactness of coastal
buffer
Intensively Grazed
Grassland (area,
ecosystem condition)
Invasive Plant Cover -
Grasslands and
Shrublands (Percent of
non-native plant cover)
Invasive plant coverage -
Forests (Percent area of
non-native cover)
Invasive species
Invasive species
Invasive species - Coasts
affected (area, ecosystem
condition)
Invasive species - Forests
Definition
Flexibility score (on a scale of integers from zero to five) is assigned to each
state based on the relative degree of barriers to water trading.
In-stream fish concealment features consisting of undercut banks,
boulders, large pieces of wood, brush, and cover from overhanging
vegetation within a stream and its banks.
The degree to which natural coastal landforms and vegetation are intact.
This indicator would report on the percentage of land area that is
intensively grazed grassland/shrubland.
This indicator will report the percentage of plant cover in grasslands and
shrublands that is made up of non-native species. The indicator will report
on both invasive non-native species (those that spread aggressively) and all
non-native species.
This indicator describes the degree to which non-native plants are found in
U.S. forests. It will report the percentage of the total area covered by
overstory (large trees that form the canopy) and understory (shrubs,
ground plants, and smaller trees) that is made up of non-native plants.
N/A
This indicator reports the percentage of watersheds with different numbers
of nonnative species with established breeding populations. "Non-native"
includes species not native to North America and those that are native to
this continent but are now found outside their historic range.
This indicator would report on the percentage of coastline length that is
heavily affected by invasive species.
This indicator would report on the percentage of land area that is forests
Literature Source
[See Appendix A for full
citation]
Hurdetal., 1999.
USEPA, 2006b.
Day et al., 2007.
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Twilley et al., 2001.
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Duplicate
463, 144,
618
618, 143,
463
Do Not Cite or Quote
Page B-23
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
147
148
149
150X
15 lx
Indicator
Affected (area, ecosystem
condition)
Invasive species -
Grasslands and
Shrublands Affected (area,
ecosystem condition)
Invasive species (range
expansion)
Invasive species in
estuaries (percent
influenced)
Irrigation withdrawal rate
Lake and Stream Acidity
Definition
heavily affected by invasive species.
This indicator would report on the percentage of land area that is
grasslands and shrublands heavily affected by invasives.
N/A
Percentage of major estuaries with high, medium, or low influence by
non-native species. Ratings of the degree of influence should incorporate
both the number of different species present and the degree to which
they occupy available habitat.
N/A
N/A
Literature Source
[See Appendix A for full
citation]
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
MEA, 2005b.
Heinz Center, 2002; Heinz
Center, 2008
MEA, 2005 b.
USEPA, 2008b.
Duplicate
444
604, 621
Do Not Cite or Quote
Page B-24
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
152
153X
154
155
156
157
158
159
160
161
162
163
Indicator
Lake Levels and
Conditions
Lake stratification
(thermal structure)
Land use
Lands and Waters with
Highly Altered Species Mix
(area, ecosystem
condition)
Latitude/altitude
Lined and Culverted
Streams (area)
Living Planet Index
Low flow sensitivity
(mean baseflow)
Low-flow events
M&l water use share
Major Crop Yields (Tons or
bushels per acre of land)
Mangrove cover
Definition
N/A
N/A
N/A
This indicator would report on the percentage of lands and waters with
highly altered species mix, such as would be characteristic of altered fire or
hydrologic regimes.
N/A
This indicator would report on the percentage of land area and stream
length that is lined and culverted streams.
N/A
Unregulated mean baseflow in cfslmiZ, i.e. the amount of streamflow
originating from groundwater outflow.
N/A
Municipal and industrial sector share of total average annual withdrawals.
This indicator reports the yields of corn, soybeans, wheat, hay, and cotton
(which account for 95% of crop production in the US), as an index with
1975 as the base year. Values above 1.0 indicate higher yields, typically
measured as tons or bushels per acre, than in 1975; values below 1.0
indicate lower yields than in 1975.
N/A
Literature Source
[See Appendix A for full
citation]
Gleick and Adams, 2000.
Klingetal., 2003.
U.S. Climate Change
Science Program, 2008.
Heinz Center, 2002; Heinz
Center, 2008
MEA, 2005b.
Heinz Center, 2002; Heinz
Center, 2008
MEA, 2005b.
Hurd et al., 1999.
Frumhoff etal., 2006.
Hurd etal., 1998.
Heinz Center, 2002; Heinz
Center, 2008
MEA, 2005b.
Duplicate
343
Do Not Cite or Quote
Page B-25
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
164
165**
166
167
168
169
170
171
172
173
Indicator
Marine fish landings
(economic production)
Meteorological drought
indices
Mid-channel clarity
Migratory bird use
Miles of Hardened
Coastline
Miles of Streams
Impounded to Lakes
(area, ecosystem
condition)
Modification of water
regimes
Monetary Value of
Agricultural Production -
Time-related (Billions of
1999 $ per year)
Monetary Value of
Agricultural Production -
Area-related (Thousands
of 1999 $ per square mile)
Native Vegetation in Areas
Dominated by Croplands
(Percentage of vegetation
native to a specific area)
Definition
Specifically refers to marine fish landings. This indicator reports the
production of food and fiber and the withdrawals of water, using an index
with 1980 as the base year.
Average Palmer Drought Severity Index value, 2003-2007. *
N/A
N/A
This indicator would report on the percentage of coastline length that is
hardened coastline.
This indicator would report on the percentage of stream length (at some
last baseline date) that has since been impounded into lakes.
A measure of the degree of anthropogenic influence, which may be
worsened by climate change.
This indicator reports the dollar value of the annual output of major crops
and livestock, in billions of 1999 dollars per year. The value is determined
by multiplying the amount of output by the prices received by farmers (in
1999 dollars).
This indicator reports the dollar value of the annual output of major crops
and livestock, in thousands of 1999 dollars per square mile. The value is
determined by multiplying the amount of output by the prices received by
farmers (in 1999 dollars).
This indicator would report, for areas where croplands account for a large
percentage of the land cover, how much of the remaining vegetation
(outside of croplands) is native to the area.
Literature Source
[See Appendix A for full
citation]
Heinz Center, 2002; Heinz
Center, 2008
National Assessment
Synthesis Team, 2000a.
Chesapeake Bay Program,
2008.
MEA, 2005 b.
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
MEA, 2005b.
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Duplicate
Do Not Cite or Quote
Page B-26
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
174
175
176X
177
178
179
Indicator
Natural coastal wetlands
Natural Ecosystem
Services
Natural Ecosystem
Services -
Urban/Suburban Lands
Nitrate and Pesticides in
Shallow Ground Water in
Agricultural Watersheds
Nitrate concentration -
major aquifers on
agricultural lands (Percent
of samples with drinking
water exceedances)
Nitrate concentration -
shallow groundwater
Definition
N/A
This indicator would report on the levels of key services provided by
"natural" ecosystems— forests, grasslands and shrublands, fresh waters,
and coasts and oceans. The goods, or products, these ecosystems
provide— such as fish, wood products, and food— can be counted, and a
monetary value often placed upon them.
Urban and suburban areas are defined by what people have built, but the
remaining "natural" components— trees, meadows, streams, wetlands, and
the like— provide valuable services to the residents of these developed
areas. Ecosystem services are the benefits, both tangible and intangible,
that these natural elements provide. For example, forested areas reduce
stormwater runoff, when compared to paved areas, and trees cool streets
and buildings, reducing energy consumption; trees also reduce urban noise
levels. Natural areas, including forests, grasslands and shrublands, beaches,
lakes, streams, and wetlands, also provide recreational opportunities,
increase property values and community amenities, and are aesthetically
pleasing.
N/A
Percentage of samples exceeding drinking-water standard for nitrate (10
milligrams per liter).
Median concentration of nitrate (in milligrams per liter).
Literature Source
[See Appendix A for full
citation]
MEA, 2005b.
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
USEPA, 2008b.
USGS, 1999.
USGS, 1999.
Duplicate
176
175
Do Not Cite or Quote
PageB-27
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
180
181
182
183
184
185
186X
187
188
Indicator
Nitrate in Farmland
Streams and Groundwater
Nitrate in Grassland and
Shrubland Groundwater
(Percent of groundwater
sites tested)
Nitrate in Streams
(Average nitrate
concentration)
Nitrate in Urban and
Suburban Streams
(Percent of streams with
certain level of detects)
Nitrogen (movement,
yield/load)
Nitrogen (total)
Nitrogen and Phosphorus
-large rivers
Nitrogen and Phosphorus
-streams in agricultural
watersheds
Nitrogen and Phosphorus
-wadeable streams
Definition
This indicator reports on the concentration of nitrate in representative
farmland streams and groundwater sites. Specifically, the indicator reports
the percentage of streams and groundwater wells with average nitrate
concentrations in one of four ranges, in areas that are primarily farmland.
This indicator reports on the concentration of nitrate in groundwater in
grassland and shrubland areas. Specifically, the indicator reports the
percentage of groundwater sites with average nitrate concentrations in
one of four ranges, in areas that are primarily grassland or shrubland.
This indicator reports on the concentration of nitrate in representative
streams in forested areas. Specifically, the indicator reports the percentage
of streams with average nitrate concentrations in one of four ranges, for
streams draining watersheds that are primarily forested.
This indicator reports the concentration of nitrate in streams in
representative urban areas. Specifically, the indicator reports the
percentage of streams with average nitrate concentrations in one of four
ranges, for streams draining watersheds that are primarily urban.
This indicator reports the yield of nitrogen from major watersheds: pounds
of nitrogen per square mile of watershed area that enters rivers and
streams through discharges, runoff, and other sources. It also reports the
load of nitrate, a common form of nitrogen, from major rivers: tons of
nitrate carried to the ocean each year by the four largest U.S. rivers.
This indicator reports the total nitrogen concentration in a system.
This indicator reports the nitrogen and phosphorus content of large rivers.
This indicator reports the nitrogen and phosphorus content of streams and
rivers in agricultural watersheds.
This indicator reports the nitrogen and phosphorus content of wadeable
streams.
Literature Source
[See Appendix A for full
citation]
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
USEPA, 2006b.
USEPA, 2008b.
USEPA, 2008b.
USEPA, 2008b.
Duplicate
459
Do Not Cite or Quote
Page B-28
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
189
190
191
192
193
194
195
196
Indicator
Nitrogen concentration -
streams (total)
Number of Dry Periods in
Grassland/Shrubland
Streams and Rivers
(Percent of streams with
dry periods over time)
Nutrient enrichment
(coastal wetlands)
Open Mines, Quarries,
and Gravel Pits (area)
Patches of Forest,
Grassland and Shrubland,
and Wetlands - By Region
(Percent area of natural
lands in urban/suburban
areas by region)
Patches of Forest,
Grassland and Shrubland,
and Wetlands - National
(Percent area of natural
lands in urban/suburban
areas)
Permafrost
Permafrost temperatures
Definition
Average annual concentration of total nitrogen (in milligrams per liter).
The indicator tracks the frequency and duration of zero-flow conditions
for streams and rivers in grassland/shrubland regions. It reports the
percentage of streams and rivers that have at least one no-flow day per
year, and the percentage where the duration of zero-flow periods for a
given period is substantially longer or shorter than the long-term average.
Nutrient enrichment of coastal wetland ecosystems.
This indicator would report on the percentage of land area that is open
mines, quarries, and gravel pits, measured from satellite.
This indicator reports how much of the "natural" area within urban and
suburban lands is in patches of varying size, from less than 10 acres to
greater than 10,000 acres. Natural areas include forests, grasslands and
shrublands (including most pasturelands — especially in the west), and
wetlands.
This indicator reports how much of the "natural" area within urban and
suburban lands is in patches of varying size, from less than 10 acres to
greater than 10,000 acres. Natural areas include forests, grasslands and
shrublands (including most pasturelands — especially in the west), and
wetlands.
N/A
N/A
Literature Source
[See Appendix A for full
citation]
USGS, 1999.
Heinz Center, 2002; Heinz
Center, 2008
Day et al., 2008.
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Gleick and Adams, 2000.
Lettenmaier et al., 2008.
Duplicate
Do Not Cite or Quote
Page B-29
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
197
198
199X
200
201
202
203
204
Indicator
Pesticide Exceedances in
Farmland Streams and
Groundwater (Percent of
waterbodies with
regulatory exceedances)
Pesticide Occurrence in
Farmland Streams and
Groundwater (Number of
pesticides detected in
waterbodies)
Pesticides in Streams in
Agricultural Watersheds
Phosphorus (total)
Phosphorus in Farmland
Streams
Phosphorus in Urban and
Suburban Streams
(Percent of streams with
certain level of detects)
Phosphorus -lakes,
reservoirs (concentration)
Phosphorus -large rivers
Definition
This indicator reports on pesticides and pesticide degradates found in
farmland streams and groundwater as the percentage of streams and
shallow groundwater wells with contaminant concentrations that exceeded
standards and guidelines (benchmarks) set for the protection of human
health or aquatic life. Data report currently used agricultural pesticides and
selected breakdown products of these pesticides, as well as selected
organochlorine insecticides that were widely used in the past but whose
use is no longer permitted in the United States.
This indicator reports on pesticides and pesticide degradates found in
farmland streams and groundwater as the average number of such
contaminants detected throughout the year in streams and shallow
groundwater wells. Data report currently used agricultural pesticides and
selected breakdown products of these pesticides, as well as selected
organochlorine insecticides that were widely used in the past but whose
use is no longer permitted in the United States.
N/A
Indicator reports total phosphorus concentration in water body
This indicator reports on the concentration of phosphorus in representative
farmland streams. Specifically, the indicator reports the percentage of
streams with average annual concentrations in one of four ranges, for
streams draining watersheds that are primarily farmland.
This indicator reports the concentration of phosphorus in representative
streams in urban areas. Specifically, the indicator reports the percentage of
streams with average annual concentrations in one of four ranges, for
streams draining watersheds that are primarily urban.
This indicator reports the average concentration of phosphorus in lakes and
rivers.
This indicator reports the average concentration of phosphorus in large
Literature Source
[See Appendix A for full
citation]
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
USEPA, 2008b.
USEPA, 2006b.
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Duplicate
199
198
Do Not Cite or Quote
Page B-30
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
205
206
207
208
209
210
211
212X
213X
Indicator
(concentration)
Phytoplankton
Plant Growth Index
Population
Population (human)
Population (human)
susceptible to flood risk
Population Trends in
Invasive and Non-invasive
Grassland/Shrubland Birds
(Percent change in
population size of species
overtime)
Potential for wetland
migration
Precipitation
Precipitation
Definition
rivers.
N/A
This indicator reports a plant growth index, based on satellite
measurements of the amount of solar energy absorbed by vegetation and
potentially used for photosynthesis. The index shows, for any given year,
whether plant growth in a region or for an ecosystem type was above or
below the 11-year average
N/A
This indicator reports the human population, using an index with 1980 as
the base year.
Population within the 500-year flood plain.
This indicator describes population trends for selected grassland/shrubland
bird species by comparing trends for selected "invasive" species with those
that are not invasive.
Area of tidal wetlands compared to area of land within one-half tide range
above spring high water. Ratio of tidal wetlands to dry land and all land.
N/A
N/A
Literature Source
[See Appendix A for full
citation]
Center, 2008
Chesapeake Bay Program,
2008.
Heinz Center, 2002; Heinz
Center, 2008
Twilley et al., 2001.
Heinz Center, 2002; Heinz
Center, 2008
Hurd et al., 1999.
Heinz Center, 2002; Heinz
Center, 2008
U.S. Climate Change
Science Program, 2008.
Twilley et al., 2001.
Lettenmaier et al., 2008.
Duplicate
208
207
360
213,214,
215
212,214,
215
Do Not Cite or Quote
Page B-31
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
214*
215
216
217
218**
219**
220
221
Indicator
Precipitation
Precipitation (average
daily)
Production of Cattle on
Grasslands and
Shrublands (Number of
cattle overtime)
Publicly Accessible Open
Space per Resident
(Percent of metropolitan
areas with a certain
amount of open space per
resident)
Ratio of Snow to
Precipitation (S/P)
Ratio of water
withdrawals to annual
stream flow
Recreation - Outdoor
(number of activities)
Recreation - Participation
in freshwater activities
(number of days)
Definition
N/A
N/A
This indicator reports the number of cattle grazing on grasslands and
shrublands (including pastures), rather than at feedlots, during July of each
year.
This indicator would report the amount of open space— land that is
dominated by "natural" surfaces, like grass or woods, along with lakes,
rivers, beaches, and wetlands— that is accessible to the general public in
large metropolitan areas. Specifically, the indicator would report the
percentage of metropolitan areas with different amounts of open space
per resident.
Average annual ratio of snowfall (in inches) to total precipitation (in
inches). *
Ratio of total annual surface and groundwater withdrawal in 1990 (QW)
to unregulated mean annual streamflow (Qs). Method of calculation:
QW/Qs
This indicator reports the number of times Americans over the age of 15
took part in a variety of outdoor recreational activities. (Each time
someone took part in an activity is counted: if the activity took place over
multiple days, each day counts as a separate event, and if a person took
part in several activities on a single day, each activity is counted as a
separate event.)
This indicator shows the number of days that people took part in a variety
of freshwater activities. A "recreation day" for this measure is any day
during which a person was engaged in the activity, whether for only a few
Literature Source
[See Appendix A for full
citation]
Gleick and Adams, 2000.
Klingetal., 2003 using
data from Wuebbles and
Hayhoe, 2003.
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Lettenmaier et al., 2008.
Hurdetal., 1999.
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Duplicate
212,213,
215
212,213,
214
259, 260
608
Do Not Cite or Quote
Page B-32
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
222
223
224
225
226
227
228X
Indicator
Recreation in Forests
(Number of days of
recreational activity per
year)
Recreation on Farmlands
(Number of days of
engagement in
recreational activities)
Recreation on Grasslands
and Shrublands (Number
of days of recreational
activity per year)
Recreational water quality
(beach-mile-days affected
by Enterococcus)
Red List Index
Relative Bed Stability
(RBS)
Relative sea level rise
(RSLR)
Definition
minutes or for many hours.
This indicator would report the number of days per year that people
engage in a variety of recreational activities in forests. Activities such as
walking, hiking and backpacking, fishing and hunting, wildlife viewing,
cross-country and downhill skiing, and snowmobiling would be included.
This indicator would report the number of days spent fishing, hunting,
viewing wildlife, or engaged in other recreational activities on farmland.
This indicator will report the number of days per year that people engage
in a variety of recreational activities on the nation's grasslands and
shrublands. Activities will include: hunting; off-road vehicle (ORV) driving,
motorsports, mountain biking, and snowmobiling; bird watching and
nature study; and hiking and camping.
This indicator will report the percentage of "beach-mile-days" affected by
various levels of Enterococcus, a bacterium that indicates contamination
with human or animal waste. A "beach-mile-day" is one mile of beach
affected for one day— 100 miles of beach affected for one day would count
the same as 1 mile affected for 100 days.
Index of threatened/endangered status for birds, based on a 1988 baseline.
Ratio that compares measures of particle size of observed sediments to the
size of sediments that each stream can move or scour during its flood stage
(based on measures of the size, slope, and other physical characteristics of
the stream channel).
Function of absolute sea level, changes in land level, and sediment delivery
Literature Source
[See Appendix A for full
citation]
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
MEA, 2005b.
USEPA, 2006b;USEPA,
2008b.
MEA, 2005 b.
Duplicate
229, 241,
405, 412
Do Not Cite or Quote
Page B-33
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
229
230
231
232
233
234
235X
236
237
238
239
240X
Indicator
Relative sea level rise
(RSLR)
Riparian Areas (extent,
acreage)
Riparian Condition
(Riparian Condition
Index)
River channel and
geomorphology
River flow and nitrogen
loads
Road surface (area)
Runoff
Runoff from precipitation
and snowmelt (mean
annual)
Runoff -large areas
Runoff patterns
Runoff -regions
Salinity
Definition
The net change in the relative elevations of the sea surface and coastal
lands
For streams and rivers, the indicator reports on the type of land cover on
their shorelines and adjacent areas ("riparian" areas): forest; grasslands,
shrublands, or wetlands; and urban/suburban or agricultural land.
This indicator will describe the condition of riparian (streamside) areas.
The condition of these areas will be rated using an index that combines
key factors such as water flows, streambed physical condition, riparian
vegetation composition and structure, and use by various species. Such a
measure should take into account multiple factors, including hydrology
(e.g., relationship to natural flow patterns), geomorphology (e.g., stream
sediment transport), and biology (e.g., canopy cover) to provide an
overall index of condition.
N/A
Indicator reports total nitrogen arriving in a water body (e.g., from the
Mississippi River system to the Gulf of Mexico).
This indicator would report on the percentage of land area that is road
surface (including unpaved roads).
N/A
Water-balance model (monthly precipitation and potential
evapotranspiration) and output from the two GCMs to estimate the effects
of climate change
N/A
N/A
N/A
N/A
Literature Source
[See Appendix A for full
citation]
Day et al., 2008.
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Gleick and Adams, 2000.
Chesapeake Bay Program,
2008.
Heinz Center, 2002; Heinz
Center, 2008
Twilley et al., 2001.
National Assessment
Synthesis Team, 2000a.
Gleick and Adams, 2000.
Lettenmaier et al., 2008.
Gleick and Adams, 2000.
USEPA, 2006b.
Duplicate
228, 241,
405, 412
236
235
603, 84
Do Not Cite or Quote
Page B-34
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
241*
242
243X
244
245
246
247
248X
249
Indicator
Sea level rise
Sea surface temperature
(difference from average)
Sea surface temperature
(trend)
Seagrass cover
Seals (number)
Sediment contaminants
(chemical levels)
Sediment Contamination
(exceedance of regulatory
value)
Sediment delivery
Sediment discharge (river
to coast)
Definition
N/A
Indicator describes whether sea surface temperature (SST) is above or
below average. Method used: (1) the seasonal average sea surface
temperature (SST) of near-shore water (shoreline out to 25 miles) was
calculated for the warmest season in each region (termed the "seasonal
mean maximum"), which typically occurred during summer or fall; (2) the
long-term mean (during the warmest seasons) for the period of
observation (1985-1998) was calculated; and (3) the long-term mean was
then subtracted from the seasonal mean maxima. Thus, values greater than
zero are positive "anomalies" (i.e., deviations from the long-term average),
and those less than zero are negative anomalies.
N/A
N/A
Number of seals observed in winter months at two monitoring locations.
Total organic carbon, metals, polynuclear aromatic hydrocarbons (PAH),
PCBs (Polychlorinated Biphenyls), pesticides.
Using ERM and ERL guidelines. Good: No ERM values are exceeded, and
fewer than five ERL values are exceeded. Fair: No ERM values are
exceeded, and five or more ERL values are exceeded. Poor: One or more
ERM values are exceeded.
N/A
Sediment load discharged from riverine systems into estuarine and coastal
systems
Literature Source
[See Appendix A for full
citation]
Gleick and Adams, 2000.
Heinz Center, 2002; Heinz
Center, 2008
Twilley et al., 2001.
MEA, 2005b.
Long Island Sound Study,
2008.
Hayslip et al., 2006.
USEPA, 2006a.
MEA, 2005 b.
Day et al., 2008.
Duplicate
228, 229,
405, 412
243
242
249
248
Do Not Cite or Quote
Page B-35
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
250
251
252
253
254
255
256
Indicator
Sediment Quality Index (3
components)
Sediment Total Organic
Carbon (TOC)
Sediment toxicity
Sediment Toxicity (species
test)
Shape of "Natural"
Patches (Ratio of
perimeter to area of
patch)
Shellfish (acreage,
harvest)
Shoreline types
(miles/category)
Definition
Based on three sediment quality component indicators: sediment toxicity,
sediment contaminants, and sediment TOC. Good: Less than 5% of the NEP
estuarine area is in poor condition, and more than 50% of the NEP
estuarine area is in good condition. Fair: 5% to 15% of the NEP estuarine
area is in poor condition, or more than 50% of the NEP estuarine area is in
combined poor and fair condition. Poor: More than 15% of the NEP
estuarine area is in poor condition.
Good: The TOC concentration is less than 2%. Fair: The TOC concentration
is between 2% and 5%. Poor: The TOC concentration is greater than 5%.
Acute sediment toxicity test.
Using a 10-day static toxicity test with the amphipod Ampelisca abdita.
Good: Mortality is less than or equal to 20%. Poor: Mortality is greater than
20%.
This indicator describes the shape of patches of "natural" lands in the
farmland landscape, by reporting on the percentage of patch area that is
found in "compact" patches (e.g., like a circle), "elongated" patches (e.g.,
like a long narrow rectangle), and an intermediate class of patch shape.
These classes are defined based on the ratio of the perimeter, or edge, of
each patch to its area; these perimeter-to-area ratios will be divided by
patch area for the sake of comparison. "Natural" areas include forest,
grasslands and shrublands, wetlands, and lands enrolled in the
Conservation Reserve Program (CRP).
Indicators include acreage of shellfish beds and harvests of important
commercial mollusks and crustaceans (oyster and lobster).
Reports the miles of coastline in several categories, including: beach; mud
or sand flats; steep sand, rock, or clay cliffs; wetlands; and coastline
"armored" with bulkhead or riprap. The coastline includes ocean-front
areas and the shoreline of estuaries and bays.
Literature Source
[See Appendix A for full
citation]
USEPA, 2006a.
USEPA, 2006a.
Hayslip et al., 2006.
USEPA, 2006a.
Heinz Center, 2002; Heinz
Center, 2008
Long Island Sound Study,
2008.
Heinz Center, 2002; Heinz
Center, 2008
Duplicate
50
Do Not Cite or Quote
Page B-36
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
257
258
259X
260X
261
262
263X
264
265
266X
267
Indicator
Shrubland Patches
(Percent of grassland
patches that cover a
certain area)
Silt-Clay content of
sediment
Snow influence
Snow Water Equivalent
(SWE)
Snowmelt event dates
Snowmelt runoff volume
Snowpack
Snowpack
Snowpack density
Snowpack depth
Soil Biological Condition
(Percentage of croplands
with variety of worms)
Definition
This indicator will describe the fraction of grassland area and shrubland
area that is in patches of different sizes. The total area occupied by patches
of a certain size will be reported as a percentage of the total area of either
grasslands or shrublands.
The proportion of fine grained materials (silt and clay) in the estuarine
sediments.
Share of average annual precipitation (P) that falls as snow (Ps) (i.e., snow
water equivalent).
N/A
N/A
N/A
N/A
N/A
N/A
N/A
This indicator reports the percentage of croplands in three different ranges
on the Nematode Maturity Index (NMI), an index that measures the types
of roundworms, or nematodes, in the soil. Calculation of the NMI is based
on the proportion of nematodes with different levels of tolerance for
disturbance. A map showing the percentage of cropland in each major
cropland region with low index values (indicating disturbed soils) would
accompany the nationwide map.
Literature Source
[See Appendix A for full
citation]
Heinz Center, 2002; Heinz
Center, 2008
Hayslip et al., 2006.
Hurdetal., 1998.
Lettenmaier et al., 2008.
Lettenmaier et al., 2008.
Lettenmaier et al., 2008.
Frumhoff et al., 2006;
Frumhoffetal., 2007.
Gleick and Adams, 2000.
Lettenmaier et al., 2008.
Lettenmaier et al., 2008.
Heinz Center, 2002; Heinz
Center, 2008
Duplicate
218, 260
259,218
264
263
438, 440
Do Not Cite or Quote
PageB-37
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
268
269*
270
271
272
273
274
275X
Indicator
Soil Erosion (Percentage
of U.S. farmlands prone to
erosion)
Soil moisture
Soil moisture
Soil Organic Matter
(Percentage of organic
matter by soil weight)
Soil Salinity (Percentage of
croplands with high
salinity)
Species Status (Percent of
metropolitan areas with
species at risk)
Spring high water
Spring ice-out dates
Definition
This indicator reports the percentage of U.S. farmlands according to their
potential for erosion by wind or water. These data are based on an index
that combines information on soil characteristics, topography, and
management activities such as tillage practices and whether crop residue is
left on the field or not. This indicator covers croplands (excluding pastures)
and Conservation Reserve Program (CRP) lands.
N/A
N/A
This indicator reports the amount of organic matter — partially decayed
plant and animal matter — in the top 4-6 inches of cropland soil. Soil
organic matter is usually measured as the percentage of organic matter (by
dry weight) in the top 4-6 inches of the soil, where human activities have
most influence on soil condition.
This indicator reports the percentage of cropland with different levels of
salt content, measured in decisiemens per meter (dS/m). A map showing
the percentage of land in each major cropland region with elevated salt
levels (i.e., over 4 dS/m), would accompany the nationwide map.
This indicator reports the degree to which "original" plants and animals are
either absent entirely or are at risk of being lost from metropolitan areas.
Original species are those that, before European settlement, inhabited the
lands now occupied by metropolitan areas. Specifically, the indicator will
report on the fraction of metropolitan areas where 25% or more, 50% or
more, and 75% or more of original species are at risk of being displaced or
are absent.
This indicator is the average high tide during a full or new moon. It
approximates the boundary between tidal wetlands and dry land.
This indicator reports the date at which lake ice cover ends.
Literature Source
[See Appendix A for full
citation]
Heinz Center, 2002; Heinz
Center, 2008
Twilley et al., 2001.
Gleick and Adams, 2000.
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
U.S. Climate Change
Science Program, 2008.
Lettenmaier et al., 2008.
Duplicate
270
269
134, 423
Do Not Cite or Quote
Page B-38
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
276
277
278
279
280X
281
282X
Indicator
Status of Animal
Communities in Urban
and Suburban Streams
(Percent of
urban/suburban sites
with undisturbed and
disturbed species)
Status of Animal Species
in Farmland Areas
Stream Bank Vegetation
Stream flow variability
(annual)
Stream flow variability
(daily and weekly)
Stream flow variability
(daily and weekly)
Stream flow variability
(daily)
Definition
This indicator reports on "biological integrity" in urban and suburban
streams. Biological integrity is a measure of the degree to which the suite
of fish and bottom-dwelling (or benthic) animals (including insects,
worms, mollusks, and crustaceans) resemble what one might find in a
relatively undisturbed stream in the same region. Tests assess the
number of different species, number and condition of individuals, and
food chain interactions. High scores indicate close resemblance to
"reference" or undisturbed conditions, and low scores indicate significant
deviation from them.
This indicator reports the status of wildlife in farmland areas.
This indicator describes the percentage of miles of stream (stream-miles) in
urban and suburban areas that are lined with trees, shrubs, and other
plants.
The coefficient of variation (CV) of unregulated streamflow is an indicator
of annual streamflow variability. It is computed as the ratio of the
standard deviation of unregulated annual streamflow (oQs) to the
unregulated mean annual streamflow (QS)'.
Trends in stream flow volumes based on daily flow data (same as
"Changing Stream Flows" from (2) Heinz Center, 2002) are indicators of
daily and weekly stream flow variability.
This indicator describes changes in the amount and timing of river and
stream flow by reporting the percentage of monitored streams or rivers
with major, moderate, and minimal changes in low flow, high flow, and the
timing of these two extreme events. The indicator also describes the nature
of major flow changes. Four subindicators were included in the analysis:
N/A
Literature Source
[See Appendix A for full
citation]
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Hurdetal., 1999.
USEPA, 2008b.
Heinz Center, 2002; Heinz
Center, 2008
Lettenmaier et al., 2008.
Duplicate
413, 415
281, 282
280, 282
280, 281
Do Not Cite or Quote
Page B-39
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
283*
284**
285
286
287
288
289
Indicator
Stream Habitat Quality -
Farmland Streams
(Presence of certain
attributes compared to
undisturbed streams)
Stream habitat quality
(comparison to baseline)
Streamflow (discharge)
Submerged Aquatic
Vegetation (SAV)
Suburban/Rural Land Use
Change
Suspended solids (total, 4
substances)
Temperature (average
annual)
Definition
This indicator describes the habitat quality of farmland streams by
comparing a number of key attributes to those of relatively undisturbed
streams in the same general area. The index would incorporate the
presence of riffles and pools, the size of streambed sediments and the
degree to which larger gravel and cobbles are buried in silt, the presence of
branches, tree trunks, and other large woody pieces, and the stability of
the bank.
This indicator is represented by the Rapid Bioassessment Protocol score,
an index that can be used to assess the condition of underwater and bank
habitats. The Rapid Bioassessment Protocol score is used to assess
habitat conditions based on field observations often variables: epifaunal
substrate/ available cover, embeddedness (for riffles) or pool substrate
characterization (for pools), velocity and depth regimes (for riffles) or
pool variability (for pools), sediment deposition, channel flow status,
channel alteration, frequency of riffles or bends (for riffles) or channel
sinuosity (for pools), bank stability (condition of banks), bank vegetative
protection, and riparian vegetated zone width.*
Discharge data from a national network of stream gages
Rooted aquatic plants that support the health of ecosystems by generating
food and habitat for waterfowl, fish, shellfish, and invertebrates.
This indicator describes the pattern and intensity, or density, of
development, both at the outer edge of suburban development around
cities, and in rural areas that, despite the lack of a large town center, are
growing rapidly toward suburban densities.
Suspended materials include soil particles (clay and silt), algae, plankton,
and other substances. Total suspended solids (TSS) refer to the matter that
is suspended in water.
Mean annual temperature.
Literature Source
[See Appendix A for full
citation]
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
National Assessment
Synthesis Team, 2000a.
USEPA, 2008b.
Heinz Center, 2002; Heinz
Center, 2008
Hayslip etal., 2006.
Gleick and Adams, 2000.
Duplicate
284
283
549
Do Not Cite or Quote
Page B-40
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
290
291
292
293
294
295
296
297
298
Indicator
Temperature (average
daily)
Temperature -streams
Thermal sensitivity
(changes in extremes)
Tidal wetlands area
Timber Growth and
Harvest- Private (Volume
of timber harvested vs.
grown overtime)
Timber Harvest - By
Region (Volume of timber
overtime)
Timber Harvest - By Use:
Fuelwood (Volume of
timber overtime)
Timber Harvest - By Use:
Logging Residues/Other
(Volume of timber over
time)
Timber Harvest - By Use:
Other Products (Volume
of timber overtime)
Definition
N/A
N/A
This indicator is based on the sensitivity to changes in extreme
temperatures. It combines the vulnerability of two sub-indicators: (1) heat
(the average annual number of days with maximum temperatures
exceeding 35 degrees C) and (2) cold (the average annual number of days
with average temperatures below O degrees C).
N/A
This indicator reports the annual amount of new wood grown and the
annual amount of wood harvested on public and private timberlands, by
region.
This indicator reports trends in timber harvest, by regions East and West,
and by primary product category (sawlogs, pulpwood, etc.).
This indicator reports trends in timber harvest, by region and by primary
product category (sawlogs, pulpwood, etc.).
This indicator reports trends in timber harvest, by region and by primary
product category (sawlogs, pulpwood, etc.).
This indicator reports trends in timber harvest, by region and by primary
product category (sawlogs, pulpwood, etc.).
Literature Source
[See Appendix A for full
citation]
Klingetal., 2003 using
data from Wuebbles and
Hayhoe, 2003.
Lettenmaier et al., 2008.
Hurdetal., 1999.
Chesapeake Bay Program,
2008.
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Duplicate
Do Not Cite or Quote
Page B-41
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
299
300
301
302
303
304
305
306
307
Indicator
Timber Harvest - By Use:
Pulpwood (Volume of
timber overtime)
Timber Harvest - By Use:
Sawlogs (Volume of
timber overtime)
Timber Harvest - By Use:
Veneer logs (Volume of
timber overtime)
Topography/Elevation
(LIDAR)
Total Impervious Area
(Percent of
urban/suburban areas
having a certain amount
of impervious area)
Trophic state
Trophic state (coastal
waters)
Tropical storm frequency
Unusual marine
mortalities
Definition
This indicator reports trends, by region, in the harvest of pulpwood.
This indicator reports trends, by region, in the harvest of sawlogs.
This indicator reports trends, by region, in the harvest of veneer logs.
This indicator includes: (1) Light Detection and Ranging (LIDAR) mapping,
and (2) shallow water-penetrating LIDAR.
This indicator classifies urban and suburban areas according to their
percentage of impervious surface (e.g., roads, parking lots, driveways,
sidewalks, rooftops, etc.). The indicator uses several thresholds: less than
10% impervious surface in the region, at least 10%, at least 20%, and at
least 30%.
This indicator integrates water clarity, phosphorus concentration, and
chlorophyll a.
This indicator refers to aspects of aquatic systems associated with the
growth of algae, decreasing water transparency, and lowering oxygen
levels in the lower water column that can harm fish and other aquatic life.
Frequency and intensity of tropical storms (current and predicted with
climate change)
Unusual mortality events (UME) are characterized by an abnormal number
of dead animals or by the appearance of dead animals in locations or at
times of the year that are not typical for that species.
Literature Source
[See Appendix A for full
citation]
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Heinz Center, 2002; Heinz
Center, 2008
Coastal States
Organization, 2007.
Heinz Center, 2002; Heinz
Center, 2008
USEPA, 1995.
USEPA, 2008b.
Day et al., 2007.
Heinz Center, 2002; Heinz
Center, 2008
Duplicate
Do Not Cite or Quote
Page B-42
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
Indicator
Definition
Literature Source
[See Appendix A for full
citation]
Duplicate
308
Urban and Suburban
Areas (extent/acreage)
This indicator presents the extent/acreage of urban and suburban areas as
a percentage of the total U.S. land area, for the most recent 50-year period
and compared to pre-settlement estimates. It also reports on a key
component of freshwater ecosystems (freshwater wetlands) and will report
on the area of brackish water, a key component of coastal and ocean
ecosystems when data become available.
Heinz Center, 2002; Heinz
Center, 2008
309
Urban and Suburban
Lands - By Region (Area
covered by
urban/suburban lands by
region)
This indicator reports the extent of urban and suburban lands, in acres.
Heinz Center, 2002; Heinz
Center, 2008
310
Urban and Suburban
Lands- By Region (Percent
area covered by
urban/suburban lands by
region)
This indicator reports the extent of urban and suburban lands as a
percentage of all land area in a region.
Heinz Center, 2002; Heinz
Center, 2008
311
Urban and Suburban
Lands - Composition of
Undeveloped Urban and
Suburban Lands (Percent
area of undeveloped lands
by region)
This indicator reports on the extent and composition of undeveloped lands,
such as wetlands, croplands, forest, or grassland and shrubland, contained
within urban and suburban areas.
Heinz Center, 2002; Heinz
Center, 2008
312
Urban Heat Island
(Percent of metropolitan
areas with having certain
differences in air
temperature between
rural and urban areas)
This indicator describes the difference between urban and rural air
temperatures for major U.S. metropolitan areas. Temperatures within
urban areas will be compared to those in less-developed surrounding
areas.
Heinz Center, 2002; Heinz
Center, 2008
313
Vegetative cover -riparian
(area of 3 classes)
This indicator reports the sum of the amount of woody cover provided by
three layers of riparian vegetation: the ground layer, woody shrubs, and
USEPA, 2006b.
Do Not Cite or Quote
Page B-43
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
314
315
316X
317X
318
319
320
321
Indicator
Water Availability:
Streamflow per Capita
(Vulnerability of Domestic
Water Uses)
Water Clarity
Water Clarity (seed,
remote sensing)
Water clarity (secci,
transmitivity)
Water Clarity Index (real
vs. reference)
Water Quality Index (5
components)
Water use to storage
Water withdrawals
Definition
canopy trees.
A measure estimating per capita water availability based on per capita
average annual streamflow (QS). Method of calculation: 1. QS/pop.
This indicator reports the percentage of lake and reservoir area with low-,
medium-, and high-clarity water. (Ponds are not included because of their
shallow depth.)
This indicator presents correlations between water clarity, as measured by
the Secchi disk, and light in the blue and red bands of the spectrum
reflected from lake water surfaces and measured as "brightness" by
satellite sensors.
Light transmitivity, secci depth.
Water clarity index (WCI) is calculated by dividing observed clarity at 1
meter by a regional reference clarity at 1 meter. This regional reference is
10% for most of the U.S., 5% for areas with naturally high turbidity, and
20% for areas with significant submersed aquatic vegetation (SAV) beds
or active SAV restoration programs. Good: WCI ratio is >2. Fair: WCI ratio
is between 1 and 2. Poor: WCI ratio < 1.
This indicator is based on 5 water quality component indicators (DIN, DIP,
chlorophyll a, water clarity, and dissolved oxygen).
This indicator is the ratio of total annual average surface and groundwater
withdrawals in 1990 (Qw) to total active basin storage (S). Method of
calculation: QW/S.
This indicator reports the total amount of surface water and groundwater
withdrawn for use in the municipal, rural, industrial, thermoelectric, and
Literature Source
[See Appendix A for full
citation]
Hurdetal., 1998.
Heinz Center, 2002; Heinz
Center, 2008
Brezonik et al., 2007.
Hayslipetal., 2006.
USEPA, 2006a.
USEPA, 2006a.
Hurd et al., 1998.
Heinz Center, 2002; Heinz
Center, 2008
Duplicate
316, 317
315, 317
315, 316
447
Do Not Cite or Quote
Page B-44
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
322
323
324
325
326**
327
328
329
Indicator
Waterborne human
disease outbreaks
(events)
Watershed forest cover
(area)
Wetland Extent, Change,
and Sources of Change
Wetland loss
Wetland species at risk
(number of species)
Wetlands, Lakes,
Reservoirs, and Ponds
(extent, acreage)
Ratio of water use to safe
yield
Percent urbanized
Definition
irrigation sectors.
This indicator reports the number of disease outbreaks (i.e., at least two
people getting sick) attributed to drinking water that is untreated or
where treatment has failed to remove disease-causing organisms, or to
swimming or other recreational contact at lakes, streams, and rivers.
N/A
N/A
Rate or total extent of wetland loss relative to original extent of wetland.
Number of wetland and freshwater species at risk, either threatened or
endangered. *
This indicator reports the area of wetlands and lakes, reservoirs, and ponds
and the length of small, medium, and large streams and rivers.
Safe yield provides an estimate of the maximum quantity of water that
can be withdrawn during an extended dry period without depleting the
source beyond its ability to be replenished in naturally 'wet years.' It is
measured as the water balance of inflows, usable storage, and
evapotranspiration determined between Jun-Oct in a median year, using
an 80% likelihood of water level recovery by the following spring. This
indicator reports the ratio of water use to safe yield.
N/A
Literature Source
[See Appendix A for full
citation]
Heinz Center, 2002; Heinz
Center, 2008
Chesapeake Bay Program,
2008.
USEPA, 2008b.
MEA, 2005b.
Hurd et al., 1999.
Heinz Center, 2002; Heinz
Center, 2008
Schmitt et al., 2008.
Schmitt et al., 2008.
Duplicate
Do Not Cite or Quote
Page B-45
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
330
331
332
333
334
335
336
337X
338
339
Indicator
Summer retail sales
Seasonal housing
Peak drinking water use
nitrate in drinking water
select pesticides in
drinking water
Total Trihalomethanes
(TTHMs) in drinking water
Cyanotoxins (the 3 on the
Contaminant Candidate
Lists orCCLS) in drinking
water
Dissolved oxygen
concentration
Lake stratification timing
Lake volume
Definition
N/A
N/A
Peak drinking water use at individual public water systems.
Percent of public water systems (PWSs) nationally with at least one
compliance monitoring sample detection of nitrate at a concentration
greater than one-half the nitrate drinking water standard.
Percent of PWSs nationally with at least one detection of alachlor, atrazine,
carbofuran, endothall, and/or simazine at a concentration greater than
one-half the respective drinking water standard.
Percent of PWSs nationally with at least one compliance monitoring sample
detection of Total Trihalomethanes (TTHMs, disinfection byproducts)
greater than one-half the TTHM drinking water standard. TTHMs include
chloroform, bromodichloromethane, dibromochloromethane, and
bromoform.
Percent of PWSs sites with at least one detection of one of the three
cyanotoxins.
N/A
The occurrence and timing of thermal stratification in lakes.
N/A
Literature Source
[See Appendix A for full
citation]
Schmittetal., 2008.
Schmittetal., 2008.
Schmittetal., 2008.
USEPA, 2008d.
USEPA, 2008d.
USEPA, 2008d.
USEPA, 2008d.
Murdoch etal., 2000.
Murdoch etal., 2000.
Murdoch etal., 2000.
Duplicate
71, 72,
73, 74,
75, 131
Do Not Cite or Quote
Page B-46
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
340
341
342
343
344
345
346
347
348**
349
350
351**
352**
353
Indicator
Ice cover duration
Metabolic rate
Epilimnion volume
Hypolimnion temperature
Permafrost and glacial
extent
Solute concentration
Surface water extent
Dominant flowpath
Erosion rate
Water clarity
Water residence time
Instream use/total
streamflow
Total use/total
streamflow
Flow regime
Definition
N/A
Rates of productivity, decomposition, and chemical reactions in surface
waters
Volume of the upper mixed layer in stratified lakes.
Temperature of the bottom water in stratified lakes.
N/A
Concentrations of dissolved materials in water
Areal extent of surface waters
Relative importance of various hydrological pathways (i.e., surface runoff,
shallow subsurface flow, deep groundwater flow)
Rate of soil erosion from watershed lands
N/A
Rate of water inflow to and outflow from a water body
Ratio of in-stream use to total streamflow. Method of calculation: (In-
stream flow requirements to meet the needs offish and wildlife)/(1975
streamflow + 1975 consumption - 1975 groundwater overdraft)
Method of calculation: (In-stream flow requirements to meet the needs
of fish and wildlife + 1975 consumption)/(1975 streamflow + 1975
consumption - 1975 groundwater overdraft)
This indicator is a composite of several measurable streamflow variables,
including magnitude, frequency, duration, seasonal timing, and rate of
change of flows
Literature Source
[See Appendix A for full
citation]
Murdoch etal., 2000.
Murdoch etal., 2000.
Murdoch etal., 2000.
Murdoch et al., 2000.
Murdoch et al., 2000.
Murdoch et al., 2000.
Murdoch et al., 2000.
Murdoch et al., 2000.
Murdoch et al., 2000.
Murdoch et al., 2000.
Murdoch et al., 2000.
Meyer etal., 1999.
Meyer etal., 1999.
Meyer etal., 1999.
Duplicate
153
76
Do Not Cite or Quote
PageB-47
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
354
355
356
357
358
359
360X
361
362
363
Indicator
Broad geographic region
Cyanobacteria abundance
Enterococci abundance
Peatland extent
Isolated wetland extent
Water quality criteria
Vulnerability to floods
Snowmelt reliance
Pesticide concentration
Predicted pesticide
concentration
Definition
Regions are Arctic and subarctic North America, Laurentian Great Lakes and
Precambrian Shield, Rocky Mountains, Mid Atlantic and New England,
Southeastern US, Pacific Coast and Western Great Basin, Great Plains, Arid
Southwest
Relative abundance of cyanobacteria in phytoplankton as characterized by
photopigments. High performance liquid chromatography (HPLC), coupled
to photodiode array spectrophotometry (PDAS).
Number of colony forming units of enterococcal bacteria per volume of
water.
Extent or density of peatland.
Extent or density of isolated wetlands (alpine wetlands, prairie potholes).
Water quality criteria are developed for specific chemicals to evaluate
whether a water body is supporting aquatic life uses. Such criteria describe
the minimum level of water quality necessary to allow a use to occur. EPA
has developed water quality criteria for 157 pollutants to protect a variety
of water body uses.
Percent of population that lives in floodplains.
Dependence of water uses on seasonal melting of snow and/or glaciers.
Pesticide concentration in surface water, groundwater, bed sediments, or
fish tissue relative to a health benchmark (human or aquatic life). Acute
and chronic benchmarks have been determined for most commonly used
pesticides.
Predicted pesticide concentration for selected pesticides (atrazine,
dieldrin). The indicator values are predicted by a regression model.
Literature Source
[See Appendix A for full
citation]
Meyer etal., 1999.
Paerl et al., 2003.
Paerl et al., 2003.
Burkett and Kusler, 2000.
Burkett and Kusler, 2000.
USEPA, 2002.
Intergovernmental Panel
on Climate Change, 2007.
Intergovernmental Panel
on Climate Change, 2007.
Gilliom etal., 2006.
Gilliom etal., 2006.
Duplicate
209
Do Not Cite or Quote
Page B-48
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
364**
365
366
367**
368
369**
370
Indicator
Pesticide toxicity index
Groundwater contribution
to baseflow
Phosphorus concentration
-streams (total)
Herbicide concentrations
in streams (Percent of
streams with highest
concentration)
Herbicides Use (Weight
per unit of agricultural
land)
Insecticide
concentrations in streams
(Percent of streams with
highest concentration)
Insecticide Use (Weight
per unit of agricultural
land)
Definition
The Pesticide Toxicity Index (PTI) accounts for multiple pesticides in a
sample, including pesticides without established benchmarks for aquatic
life. The PTI combines information on exposure of aquatic biota to
pesticides (measured concentrations of pesticides in stream water) with
toxicity estimates (results from laboratory toxicity studies) to produce a
relative index value for a sample or stream. The PTI value is computed for
each sample of stream water by summing the toxicity quotients for all
pesticides detected in the sample. The toxicity quotient is the measured
concentration of a pesticide divided by its toxicity concentration from
bioassays (such as an LC50 or EC50).
Proportion of baseflow in streams that is contributed by groundwater
discharge
Average annual concentration of total phosphorus (in milligrams per liter).
Average concentrations of herbicides in US streams. *
Herbicide use in pounds, per acre of agricultural land.
Average concentrations of insecticides in US streams. *
Insecticide use in pounds, per acre of agricultural land.
Literature Source
[See Appendix A for full
citation]
Gilliom et al., 2006.
Hayashi and Rosenberry,
2002.
USGS, 1999.
USGS, 1999.
USGS, 1999.
USGS, 1999.
USGS, 1999.
Duplicate
Do Not Cite or Quote
Page B-49
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
Indicator
Definition
Literature Source
[See Appendix A for full
citation]
Duplicate
371*
Organochlorines in Bed
Sediment (Percent of
streams with highest
concentration)
Average concentrations of organochlorines in bed sediments.
USGS, 1999.
372
Organochlorines Use
(Weight per unit of
agricultural land)
Organochlorines use in pounds, per acre of agricultural land.
USGS, 1999.
373*
Herbicides in
Groundwater (Percent of
aquifers with highest
concentration)
Average concentrations of herbicides in shallow groundwater and
aquifers. *
USGS, 1999.
374**
Insecticides in
Groundwater (Percent of
aquifers with highest
concentration)
Average concentrations of insecticides in shallow groundwater and
aquifers. *
USGS, 1999.
375
Occurrence of One or
More VOCs in Aquifers
(Concentration detected
by aquifer)
Detection of VOCs in aquifer samples demonstrates the vulnerability of
many of the Nation's aquifers to VOC contamination.
Zogorski et al., 2006.
376
Occurrence of One or
More VOCs in Aquifers -
by Aquifer Study (Percent
of aquifer studies with
detects)
Percentage of the 98 aquifer studies conducted as part of the Study-Unit
investigations that had detections of one or more VOCs at an assessment
level of 0.2 ug/L
Zogorski et al., 2006.
377
Occurrence of One or
More VOCs in Aquifers -
by Principal Aquifer
(Percent of aquifers with
detects)
Detection frequencies, expressed as a percentage, by principal or other
aquifer, of one or more VOCs at an assessment level of 0.2 ug/L
Zogorski et al., 2006.
Do Not Cite or Quote
Page B-50
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
Indicator
Definition
Literature Source
[See Appendix A for full
citation]
Duplicate
378
Occurrence of One or
More VOCs in Aquifers -
by Aquifer Lithology
(Percent of detects per
aquifer type)
Detection frequencies, expressed as a percentage, by aquifer lithology
category, of one or more VOCs at an assessment level of 0.2 |Jg/L
Zogorski et al., 2006.
379
Occurrence of VOC
Groups in Aquifers
(Percent of aquifers with
detects)
Detection frequencies, expressed as a percentage, for various VOC groups
at assessment levels of 0.2 |jg/Land 0.02 |jg/L
Zogorski et al., 2006.
380
Occurrence of Individual
VOCs in Aquifers
Detection frequencies, expressed as a percentage, for various VOCs at
assessment levels of 0.2 ng/Land 0.02 ng/L The 15 most frequently
detected VOCs represent most of the use groups and include 7 solvents, 4
THMs, 2 refrigerants, 1 gasoline oxygenate, and 1 gasoline hydrocarbon.
Zogorski et al., 2006.
381
Population Served by
Domestic Wells (Percent
of population relying on
domestic wells for
drinking water)
Domestic wells are privately owned, self-supplied sources for domestic
water use.
Zogorski et al., 2006.
382
Population Served by
Public Wells (Percent of
population relying on
public wells for drinking
water)
Public wells are privately or publicly owned and supply ground water for
PWSs. As defined by the USEPA, PWSs supply drinking water to at least 15
service connections or regularly serve at least 25 individuals daily at least
60 days a year.
Zogorski et al., 2006.
383
Pesticide detection
frequency
Atrazine, DEA, metolachlor, prometon, and simazine were "were detected
with sufficient frequency to perform statistical analysis of temporal
changes in both detection frequency and concentration."
Bexfield, 2008.
384
Playas (shallow ephemeral
lakes)
The existence of lakes in these valleys depends on local (watershed)
precipitation, playa surface evaporation and, except in endorheic basins
(especially prevalent in the Great Basin and north-central Mexico), alluvial
hydrological drainage.
Grimm etal., 1997.
Do Not Cite or Quote
Page B-51
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
385
386
387
388
389
390
391
392
393
394
395
Indicator
Ecosystem thermal regime
(latitude, shift in miles)
Migration corridor
restriction
Thermal - Habitat suitable
for trout
Stream baseflow
(summer)
Wetland hydroperiod
Fens (area, abundance)
Salinity intrusion (coastal
wetlands)
Heat-Related Illnesses
Incidence
Populations at Increased
Risk - Older and Younger
Age Groups
Populations at Increased
Risk - Using of Certain
Drugs
Populations at Increased
Risk - Dehydrated
Individuals
Definition
N/A
Dams and reservoirs, deforestation, diversion of water for offstream uses
such as irrigation and urban development.
N/A
Water in stream channel.
Patterns of water depth, and the duration, frequency, and seasonality of
flooding.
Groundwater-dominated wetlands
N/A
N/A
Heat-related mortality in individuals over 65 years of age, and in babies and
infants.
Certain drugs (such as stimulants, beta-blockers, anticholinergics, digitalis,
and barbiturates) interfere with the body's ability to cope with high
temperatures.
Populations of individuals that have a tendency to consume fewer non-
alcoholic fluids.
Literature Source
[See Appendix A for full
citation]
Poffetal., 2002.
Poffetal., 2002.
Poffetal., 2002.
Poffetal., 2002.
Poffetal., 2002.
Poffetal., 2002.
Poffetal., 2002.
Ebi et al., 2007.
Ebietal., 2007.
Ebi etal., 2007.
Ebietal., 2007.
Duplicate
Do Not Cite or Quote
Page B-52
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
396
397
398
399
400
401
402
403
404
405X
Indicator
Populations at Increased
Risk - Individuals with Low
Fitness
Populations at Increased
Risk- Individuals Who
Engage in Excessive
Exertion
Populations at Increased
Risk - Individuals Who Are
Overweight
Populations at Increased
Risk- Reduced
Adjustment to High
Outdoor Temperatures
Populations at Increased
Risk- Urban Populations
Populations at Increased
Risk - Individuals with
Lower Socio-economic
Status
Populations at Increased
Risk- Individuals Living
Alone
Temperature Increases
Precipitation Increases
Sea level rise
Definition
N/A
Outdoor workers and those who maintain a vigorous exercise regimen
during a heat wave are particularly at risk.
N/A
N/A
N/A
N/A
N/A
Impact of increased temperature on hypoxia in wetlands.
Impact of increased precipitation on hypoxia in wetlands.
Impact of increased sea level rise on hypoxia in wetlands.
Literature Source
[See Appendix A for full
citation]
Ebietal., 2007.
Ebietal., 2007.
Ebietal., 2007.
Ebietal., 2007.
Ebietal., 2007.
Ebietal., 2007.
Ebietal., 2007.
Ebietal., 2007.
Ebietal., 2007.
Ebi et al., 2007.
Duplicate
228, 229,
241, 412
Do Not Cite or Quote
Page B-53
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
406
407
408
409
410
411
412X
413X
414
415X
416
Indicator
Strength of Summer Wind
Hurricane Intensity
Hurricane Frequency
Frequency and Magnitude
of El Nino/Southern
Oscillation (ENSO)
Peak Flows
Intensity of Rainfall
Sea level rise
Streamflow Variability
(annual)
Days of 99th Percentile
Flow
Streamflow Variability
(annual min, med, max
value)
Seasonal Streamflow
Timing (Month of annual
min, med, max)
Definition
Impact of stronger summer winds on hypoxia in wetlands.
N/A
N/A
N/A
N/A
N/A
N/A
Annual minimum, median, and maximum daily Streamflow values for 1941-
1999 at 400 sites in the coterminous U.S. considered to have natural
Streamflow.
Number of days per year that meet or exceed 99th percentile Streamflow
values for 1941-1999 at 400 sites in the coterminous U.S. considered to
have natural Streamflow. 99th percentile flow value calculated for each site
from HCDN Streamflow data. 99th percentile value = 0.99 (highest flow
value for site during period of record).
Annual minimum, median, and maximum daily Streamflow values from
1940-1999 at 435 sites in the coterminous U.S. and southeastern Alaska
considered to have natural Streamflow.
Month of the annual minimum, median, and maximum streamflows for
1940-1999 at 435 sites in the coterminous U.S. and southeastern Alaska
considered to have natural Streamflow.
Literature Source
[See Appendix A for full
citation]
Ebietal., 2007.
Ebietal., 2007.
Ebietal., 2007.
Ebietal., 2007.
Ebietal., 2007.
Ebietal., 2007.
Ebi et al., 2007.
McCabe and Wolock,
2002.
McCabe and Wolock,
2002.
Lins and Slack, 2005.
Lins and Slack, 2005.
Duplicate
228, 229,
241, 405
279,415
279,413
421
Do Not Cite or Quote
Page B-54
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
417
418
419
420
42 lx
422
423
424
425
426
427
428
Indicator
Center of Volume Date
Peak Flow Date
Ratio of Seasonal to
Annual Flow
Magnitude of Monthly
Flows
Timing of Seasonal Peak
Flows
Median Total Seasonal
Snowfall
Lake Ice-Out Date
Monthly Average
Streamflow
Winter Average
Streamflow
Winter Average
Temperature
Winter Average
Precipitation
Surface Air Temperature
Definition
Date by which half of the total water volume for a time period has gone by
a river gauging station. Time periods were winter/spring (Jan 1 to May 31)
and fall (Oct 1 to Dec 31). Analyzed for 27 rural, unregulated river gauging
stations in New England with an average of 68 years of record.
Winter/spring center of volume date = WSCV.
Date of the highest daily mean flow within a season: winter/spring (Jan 1 to
May 31) and fall (Oct 1 to Dec 31). Analyzed at 27 rural, unregulated river
gauging stations in New England with an average of 68 years of record.
N/A
N/A
N/A
The median accumulation of snow (usually in units of height) over a
particular snow season.
Lake ice-out dates are the dates of ice break-up, i.e. the annual dates in
spring when winter ice cover leaves a lake.
Daily Streamflow data averaged into monthly time series data for 37 sites in
New England considered to have natural Streamflow.
Daily Streamflow data averaged into winter seasonal Streamflow data for
37 sites in New England considered to have natural Streamflow.
Monthly divisional temperature averaged into "winter average
temperature" values for the period 1895-1999.
Monthly divisional precipitation averaged into "winter average
precipitation" values for the period 1895-1999.
N/A
Literature Source
[See Appendix A for full
citation]
Hodgkinsetal., 2003.
Hodgkins et al., 2003.
Hodgkinsetal., 2003.
Hodgkinsetal., 2003.
Hodgkins et al., 2003.
Hodgkins et al., 2003.
Hodgkins et al., 2003.
Bradbury et al., 2002.
Bradbury et al., 2002.
Bradbury et al., 2002.
Bradbury et al., 2002.
Bradbury et al., 2002.
Duplicate
416
134, 275
Do Not Cite or Quote
Page B-55
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
429
430
431
432
433
434
435
436
437**
438
439
Indicator
Storm-Track Patterns
Snowfall Variability
Tree-Ring Chronology
Diurnal Range of Surface
Air Temperature (DTR)
Daily Mean Cloud Cover
Top 5-cm Soil Moisture
Content
Surface Specific Humidity
Dew Point
Precipitation Elasticity of
Streamflow
Snowpack Depth
Concentration of
Particulate and Colored
Dissolved Organic
Material (CDOM)
Definition
Spatial variations in the paths followed by centers of low atmospheric
pressure.
N/A
N/A
The difference between the daytime maximum temperature and the
nighttime minimum temperature for a 15 km by 15 km site on the Konza
Prairie near Manhattan, KS, in 1987-1989.
The daily average amount of the sky covered by clouds (reported in eighths
or oktas of sky covered) for a 15 km by 15 km site on the Konza Prairie near
Manhattan, KS, in 1987-1989.
The moisture content of the top 5 cm of soil for a 15 km by 15 km site on
the Konza Prairie near Manhattan, KS, in 1987-1989.
The ratio of the mass of water vapor within a given mass of air near the
Earth's surface (reported in g of water vapor per kg of air at specified
temperature) for a 15 km by 15 km site on the Konza Prairie near
Manhattan, KS, in 1987-1989.
The temperature to which a given unit of air must be cooled (at constant
pressure) in order for vapor to condense into water.
The proportional change in streamflow (Q) divided by the proportional
change in precipitation (P) for 1,291 gauged watersheds across the
continental US.
The regional average snowpack depth, in inches.
Concentration of pigmented compounds, dissolved in a waterbody that
derives from organic material.
Literature Source
[See Appendix A for full
citation]
Bradbury et al., 2002.
Bradbury et al., 2002.
Bradbury et al., 2002.
Daietal., 1999.
Daietal., 1999.
Daietal., 1999.
Daietal., 1999.
Roderick and Farquhar,
2002.
Sankarasubramanian et
al., 2001.
Sankarasubramanian et
al., 2001.
Warwick and Pienitz,
2006.
Duplicate
266, 440
Do Not Cite or Quote
Page B-56
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
Indicator
Definition
Literature Source
[See Appendix A for full
citation]
Duplicate
440
Snowpack Depth (2)
Regional snowpack for the Western United States.
Barnett, et al., 2005.
266,438
441
Aerosols
The amount of aerosol pollution (particulate matter) in a unit of air.
Barnett, et al., 2005.
442
Water withdrawals: Public
Supply (Volume of water
used per day)
Volume of water consumed in the public supply per day. This is different
from water withdrawals, as one of the largest uses of water is for cooling of
thermoelectric power plants, and much of that water is returned to the
streams from which it is withdrawn (use of water for hydroelectric power
generation, virtually none of which is consumptively used, is not included in
this category). On the other hand, a much higher fraction of the water
withdrawn for irrigation is consumptively used.
Lettenmaier et al., 2008.
443
Water withdrawals: Rural
Domestic and Livestock
(Volume of water used
per day)
Volume of water consumed by rural domestic and livestock per day. This is
different from water withdrawals, as one of the largest uses of water is for
cooling of thermoelectric power plants, and much of that water is returned
to the streams from which it is withdrawn (use of water for hydroelectric
power generation, virtually none of which is consumptively used, is not
included in this category). On the other hand, a much higher fraction of the
water withdrawn for irrigation is consumptively used.
Lettenmaier et al., 2008.
444
Water withdrawals:
Irrigation (Volume of
water used per day)
Volume of water consumed for irrigation purposes per day. This is different
from water withdrawals, as one of the largest uses of water is for cooling of
thermoelectric power plants, and much of that water is returned to the
streams from which it is withdrawn (use of water for hydroelectric power
generation, virtually none of which is consumptively used, is not included in
this category). On the other hand, a much higher fraction of the water
withdrawn for irrigation is consumptively used.
Lettenmaier et al., 2008.
150
445
Water withdrawals:
Thermoelectric Power
(Volume of water used
per day)
Volume of water consumed for thermoelectric power generation per day.
This is different from water withdrawals, as one of the largest uses of water
is for cooling of thermoelectric power plants, and much of that water is
returned to the streams from which it is withdrawn (use of water for
hydroelectric power generation, virtually none of which is consumptively
used, is not included in this category). On the other hand, a much higher
fraction of the water withdrawn for irrigation is consumptively used.
Lettenmaier et al., 2008.
Do Not Cite or Quote
PageB-57
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
Indicator
Definition
Literature Source
[See Appendix A for full
citation]
Duplicate
446
Water withdrawals: Other
Industrial Uses (Volume of
water used per day)
Volume of water consumed for other industrial uses per day. This is
different from water withdrawals, as one of the largest uses of water is for
cooling of thermoelectric power plants, and much of that water is returned
to the streams from which it is withdrawn (use of water for hydroelectric
power generation, virtually none of which is consumptively used, is not
included in this category). On the other hand, a much higher fraction of the
water withdrawn for irrigation is consumptively used.
Lettenmaier et al., 2008.
447
Water withdrawals: Total
(Volume of water used
per day)
The data compiled by the USGS are somewhat limited in that they are for
water withdrawals, rather than consumptive use. The distinction is
important, as one of the largest uses of water is for cooling of
thermoelectric power plants, and much of that water is returned to the
streams from which it is withdrawn (use of water for hydroelectric power
generation, virtually none of which is consumptively used, is not included in
this category). On the other hand, a much higher fraction of the water
withdrawn for irrigation is consumptively used.
Lettenmaier et al., 2008.
321
448
Extent of reservoir storage
(Volume of water stored
per area)
N/A
Lettenmaier et al., 2008.
449**
Ratio of reservoir storage
to mean annual flow
(Volume of water stored
per unit flow)
A small storage-to-runoff ratio usually indicates that the reservoir is
primarily used to shape within-year variations in runoff, and a large
storage-to-runoff value usually indicates that the reservoir is primarily
used to smooth interannual variations in runoff. *
Lettenmaier et al., 2008.
450
Precipitation (variability)
Variability of mean annual precipitation, expressed as the Coefficient of
Variation, C sub v (the standard deviation divided by the mean), for the
continental U.S.
Lettenmaier et al., 2008.
451
Runoff Ratio
Annual mean runoff divided by annual mean precipitation, for the
continental U.S. and Alaska.
Lettenmaier et al., 2008.
452
Snow to runoff ratio
Ratio of maximum mean snow accumulation to mean annual runoff, for the
continental U.S. and Alaska.
Lettenmaier et al., 2008.
Do Not Cite or Quote
Page B-58
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
453**
454
455
456
457
458
459
460**
Indicator
Runoff (variability)
Runoff (variability -
persistence)
Variability of April-Sept.
streamflow
Incidence of "surplus"
flow days
Length of snow season
Warm Season Surface
Water Supply
Phosphorus and nitrogen
concentration
Macroinvertebrate Index
of Biotic Condition
Definition
Variability of annual runoff, expressed as the Coefficient of Variation, C
sub v (the standard deviation divided by the mean), for the continental
U.S. and Alaska.
Variability of annual runoff, expressed as the "lag one" correlation
coefficient, for the continental U.S. This correlation coefficient reflects the
correlation between two values of the same variable (i.e. runoff) at
different points in time.
Variability of April-Sept, streamflow at 141 unregulated sites across the
Western U.S.
Number of days with flows above a station-dependent surplus threshold,
for 42 HCDN sites in the central and southern U.S.
Length of the snow season in the Ohio Valley over the second half of the
20th century.
Defined as precipitation minus potential evapotranspiration, for several
sites on the Arctic coastal plain over the past 50 years.
Concentrations of total P and N from 1975 to 1994 at 250 river sites with
drainages greater than 1000 square kilometers.
Total index score is the sum of scores for a variety of individual measures,
also called indicators or metrics. The metrics used to develop the Macro-
invertebrate Index for the WSA covered six different characteristics of
macroinvertebrate assemblages that are commonly used to evaluate
biological condition: taxonomic richness, taxonomic composition,
taxonomic diversity, feeding groups, habits, and pollution tolerance. Each
metric was scored and then combined to create an overall
Macroinvertebrate Index for each region, with values ranging from 0 to
100.*
Literature Source
[See Appendix A for full
citation]
Lettenmaier et al., 2008.
Lettenmaier et al., 2008.
Lettenmaier et al., 2008.
Lettenmaier et al., 2008.
Lettenmaier et al., 2008.
Lettenmaier et al., 2008.
Lettenmaier et al., 2008.
USEPA, 2006b.
Duplicate
186
Do Not Cite or Quote
Page B-59
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
461**
462
463X
464
465
466
467
468
Indicator
Macroinvertebrate
Observed/Expected (O/E)
Ratio of Taxa Loss
Coastal Benthic
Communities
Alien or Invasive Plant
Species
Native Plant Species &
Genera
Plant Species Diversity
Percent Similarity of Plant
Species Composition to
Reference Standard
Threatened &
Endangered Plant Species
Graminoid Taxa
Definition
The Macroinvertebrate O/E Ratio of Taxa Loss (henceforth referred to as
O/E Taxa Loss) measures a specific aspect of biological health: taxa that
have been lost at a site. The taxa expected (E) at individual sites are
predicted from a model developed from data collected at least-disturbed
reference sites; thus, the model allows a precise matching of sampled
taxa with those that should occur under specific, natural environmental
conditions. By comparing the list of taxa observed (O) at a site with those
expected to occur, the proportion of expected taxa that have been lost
can be quantified as the ratio of O/E.
This indicator is based on a multi-metric benthic communities index that
reflects overall species diversity in estuarine areas throughout the
contiguous United States (adjusted for salinity, if necessary) and, for
some regions, the presence of pollution-tolerant and pollution-sensitive
species. The benthic community condition at each sample site is given a
high score if the index exceeds a particular threshold (e.g., has high
diversity or populations of many pollution-sensitive species), a low score
if it falls below the threshold conditions, and a moderate score if it falls
within the threshold range.
N/A
N/A
N/A
N/A
N/A
N/A
Literature Source
[See Appendix A for full
citation]
USEPA, 2006b.
USEPA, 2008b.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
Duplicate
569, 32
143
22
Do Not Cite or Quote
Page B-60
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
469
470
471
472
473
474
475
476
477
478
479
Indicator
Dominant Plant Taxa
Dicot and Monocot
Species
Woody Species
Seedless Vascular Plants
(Ferns & Fern Allies)
Nonvascular
Plants/Bryophytes
Overall Floristic Quality -
Floristic Quality
Assessment Indices (FQAI)
and Coefficients of
Conservatism (CC)
Vegetation Indices of
Biotic Integrity (IBI)
Other Indices of
Vegetation Condition
PFG Composition -
Nutrient Functional
Groups
PFG Composition -
Sediment Functional
Groups
PFG Composition - Growth
Habit/Canopy
Architecture
Definition
N/A
N/A
N/A
N/A
N/A
FQAI use species-specific CCs that reflect the tendency of a species to occur
in pristine vs. disturbed habitats. Intact native plant communities in
relatively undisturbed sites will have high FQAI scores.
IBI use relationships of plant species, plant communities, plant guilds,
vegetation structure, etc. to anthropogenic disturbance and stress to
describe wetland condition. Typically, the highest IBI values represent
reference standards or least-disturbed ecological conditions.
N/A
Composition of plant functional groups (PFG) based on nutrient properties.
Composition of PFG based on sediment properties.
Composition of PFG based on growth habit (tap-rooted, stoloniferous,
rhizomatous, matrix interstitial, etc.).
Literature Source
[See Appendix A for full
citation]
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
Duplicate
Do Not Cite or Quote
Page B-61
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
480
481
482
483
484
485
486
487
488
489
490
491
Indicator
PFG Composition -
Sensitive Species
PFG Composition -
Tolerant Species
PFG Composition -
Hydrophyte Status
PFG Composition - Life
Span
PFG Composition - Aquatic
Plant Guilds
PFG Composition -
Pioneer Species or
Opportunistic Native or
Alien Species
PFG Composition - Light
Requirements/Shade
Tolerance
Community Type
Patchiness (Interspersion)
of Vegetation
Distribution of Plant
Communities
Zonation or Vegetation
Distribution Typical of
Wetland Class
Total Absolute Cover
Definition
Composition of PFG based on species with high coefficients of
conservatism.
Composition of PFG based on species with low coefficients of conservatism.
Composition of PFG based on hydrophyte status.
Composition of PFG based on life span (annual, perennial, .etc.).
Composition of PFG based on aquatic plants.
Composition of PFG based on pioneer status or opportunism.
Composition of PFG based on light requirements or shade tolerance.
N/A
N/A
N/A
N/A
N/A
Literature Source
[See Appendix A for full
citation]
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
Duplicate
Do Not Cite or Quote
Page B-62
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
492
493
494
495
496
497
498
499
500
501
502
503
504
Indicator
Distribution and Cover of
Growth Forms
Vertical Strata Height and
Volume
Local Microtopography
Generated By Structure of
Non-woody Plants
Leaf Area Index
Shrub or Small Tree Stem
Density
Evidence or Quantity of
Regeneration of Key Plant
Species
Litter Cover
Below Ground Biomass
Wrack Extent/Cover
Allochthonous Inputs
Depth of Submerged Plant
Cover
Distribution of Woody
Debris
Percent Leaf and Root
Tissue Carbon/Nitrogen
Definition
Floating/submerged plants, emergent/terrestrial herbs, shrubs, and trees.
Herbaceous ground layer, shrub layer(s), tree canopy layer(s).
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Literature Source
[See Appendix A for full
citation]
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
Duplicate
Do Not Cite or Quote
Page B-63
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
Indicator
Root and Leaf Tissue
Carbon/Phosphorus
Photosynthesis/Respiratio
n (P/R) Ratio
Plant
Health/Stress/Herbivory
Stable Nitrogen Isotopes
in Leaf Tissue
Genetic Diversity
Palynology
Paleoecology
USDA-NRCSHydricSoil
Field Indicators
Soil Profile Description
Soil Bulk Density
Thickness of Organic Soil
Layers
Soil Texture
SoilpH
Major Plant Nutrients in
Soil
Soil Organic Matter
Soil Electrical Conductivity
Soil Nitrogen
Definition
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Presence of hydric soil indicators.
N/A
The mass of soil particles divided by the total volume they occupy.
N/A
Percent sand, silt, and clay particles.
N/A
Presence of P, K, Ca, Mg.
Percent organic matter or carbon.
A measure of how well soil units conduct an electric current.
Presence of all N in soil.
Literature Source
[See Appendix A for full
citation]
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
Duplicate
Do Not Cite or Quote
Page B-64
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
522
523
524
525
526
527
528
529
530
531
532X
533
534
535
536
537
538
539
Indicator
Soil Micronutrients
Soil Enzyme Analysis
Soil Sulfur
GeomorphicSoil
Disturbance
Soil Surface Disturbance
Soil Channels
Sedimentation Rate
Bare Soil
Presence of Plow Layer
Soil Subsidence
Paleoecology
Depth Measurements
Water Sources
Hydrogeomorphic Unit
Vegetation Hydrologic
Guild
Landscape Characteristics
Water Chemistry
Season of Flooding
Definition
Presence of Cu, Mn, Zn, Fe.
N/A
Presence of S.
Presence of, e.g., aggregation/degradation.
N/A
Presence of incisions, channels, ditches, etc.
Rate of accretion or erosion.
Presence of bare soil features.
N/A
N/A
N/A
Surface and groundwater depth (depth of standing water, depth of water
in soil excavation pit, etc.).
N/A
N/A
N/A
N/A
N/A
N/A
Literature Source
[See Appendix A for full
citation]
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
Duplicate
567
Do Not Cite or Quote
Page B-65
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
540
541
542
543
544
545
546
547
548
549X
550
551
552
553
554
555
556
Indicator
Hydrologic
Complexity/Microhabitats
Water Velocity
Stream Bedforms
Hydrologic Regimes
COE Primary Indicators of
Inundation
COE Secondary Indicators
of Inundation
Extent of Inundation
Tidal Range
Spatial Pattern of Flooding
Stream Discharge
Direct Observation of
Streams
Stream Gauge Data
Ditch Spacing & Depth
Dams or Weirs
Levees
Drain Spacing and Depth
Irrigation
Definition
N/A
N/A
N/A
N/A
N/A
N/A
Percent of assessment area inundated, as well as spatial pattern and
average depth of inundation.
N/A
N/A
N/A
Information from landowner interviews.
N/A
N/A
Presence of dams or weirs.
Presence of levees.
N/A
N/A
Literature Source
[See Appendix A for full
citation]
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
Duplicate
285
Do Not Cite or Quote
Page B-66
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
557
558
559
560
561
562
563
564
565
566
567
568
569X
570
571
572
Indicator
Sediment Load
Snag Density
Fetch
Bird Species Diversity
Sensitive/Tolerant Bird
Species
Alien Bird Species
Dominant Bird Species
Bird Guild Composition -
Foraging
Bird Guild Composition -
Dietary
Bird Guild Composition -
Nesting Strategy
Bird Abundance
Bird Habitat Evaluation
Macroinvertebrate
Species Diversity
Sensitive/Tolerant
Macroinvertebrate
Species
Alien Macroinvertebrate
Species
Dominant
Definition
N/A
Snag density is a sign of impoundment.
Length of water surface over which wind blows in generating waves.
N/A
N/A
N/A
N/A
Composition of bird guild based on foraging technique.
Composition of bird guild based on dietary habits (omnivores, granivores,
insectivores, etc.).
Composition of bird guild based on nesting strategy (platform, ground,
cavity, etc.).
Number, status, and types of nests.
N/A
N/A
N/A
N/A
N/A
Literature Source
[See Appendix A for full
citation]
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
Duplicate
532
461
Do Not Cite or Quote
Page B-67
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
573
574
575
576
577
578
579X
580
581
582
583
584
585
586
587
588
Indicator
Macroinvertebrate
Species
Macroinvertebrate Guild
Compositions
Macroinvertebrate
Habitat Evaluation
Fish Species Diversity
Sensitive/Tolerant Fish
Species
Alien Fish Species
Dominant Fish Species
Fish Tissue Contaminants
Fish Guilds
Fish Habitat Evaluation
Fish Deformities
Algae Species Diversity
Sensitive/Tolerant Algae
Species
Alien Algae Species
Dominant Algae Species
Algal Productivity
Algae Habitat Evaluation
Definition
N/A
N/A
N/A
N/A
N/A
N/A
Presence of toxicants in fish tissue.
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Measure of algal chlorophyll and biovolume.
N/A
Literature Source
[See Appendix A for full
citation]
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
Duplicate
48, 58, 99
Do Not Cite or Quote
Page B-68
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603X
Indicator
Hydrologic Modification
Vegetative Alteration
Buffer Alteration/Buffer
Characteristics
Onsite Human
Disturbance
Invasive Species
Sedimentation
Landscape
Composition/Land
use/Land cover
Substrate Alterations
Local Surface Disturbance
Point Source Stormwater
Input
Eutrophication and
Nutrient Enrichment
Water Quality
Trash, Dredge, and Fill
Shoreline
Hardening/Barriers to
Landward Migration
Salinity and/or
Definition
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Literature Source
[See Appendix A for full
citation]
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
Duplicate
240, 84
Do Not Cite or Quote
Page B-69
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
604
605
606
607
608X
609
610
611
612
613
614
615
616
Indicator
Conductivity
Acidity
Alteration of Natural
Disturbance Regime
Toxic Contaminants
(Pharmaceuticals)
Habitat Modification
Recreational Use
Turbidity
Pipelines, Wells, Oil Rigs,
Sewer Lines
Contaminants
Thermal Stress
Alteration of Natural
Turbidity
Pathogens
Mercury Body Burdens in
Sentinel Species
Thermal Alterations
Definition
Change in pH.
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Literature Source
[See Appendix A for full
citation]
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
USEPA, 2008a.
Duplicate
621, 151
220
Do Not Cite or Quote
PageB-70
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
Indicator
Definition
Literature Source
[See Appendix A for full
citation]
Duplicate
617
Carbon Storage
This indicator reports the overall amount of carbon gained or lost over time
by major ecosystem types; the change in the amount of carbon stored per
unit area of land or water (carbon density), by major ecosystem type; and
the recent trends in the concentrations of major carbon-containing gases -
carbon dioxide and methane - in the atmosphere compared to average
preiundustrial concentrations.
Heinz Center, 2008.
35,36
618
Established Non-native
Species
This indicator reports for plants, animals, and plant and animal pathogens
across all ecosystem types: (a) the number of new non-native species that
become established over time, by decade; (b) the area with different
numbers of established non-native species; (c) the area with different
proportions of established non-native species, as a percentage of total
species.
Heinz Center, 2008.
144
619
Pattern in Coastal Area
This indicator describes the intermingling of "natural" and "non-natural"
landscape (or seascape) features in coastal areas. The interplay between
various types of coastal habitats (such as wetlands and open waters), as
well as between these habitat types and human development in the coastal
zone (such as built structures and dredged areas) provide a general
description of the structural pattern of coastal areas. The structural pattern
of these areas can be linked to how well they function ecologically and to
the amount and type of ecosystem services humans receive from them.
Heinz Center, 2008.
620
In-stream Connectivity
This indicator reports on the proportion of watersheds with different
levels of in-stream connectivity measured as the distance downstream
from the "pour point" (where the streams leaves the watershed) to the
nearest dam or diversion, such as a pumping station. This indicator also
reports on the proportion of watersheds that contain streams with
unobstructed flow to their natural endpoint, typically the ocean or a large
lake. This indicator focueses on the loss of connectivity over and above
any natural discontinuities in aquatic systems.
Heinz Center, 2008.
Do Not Cite or Quote
PageB-71
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
621*
622
623**
Indicator
Freshwater Acidity
Total withdrawal
information by source &
type of use
Water Availability: Net
Streamflow per Capita
(Vulnerability of
Domestic Water Uses)
Definition
This indicator reports: (a) the amount of nitrogen and sulfate deposited
from the atmosphere to watersheds each year; (b) the percentage of
stream miles and area of lakes and ponds with different levels of acid-
neutralizing capacity, a measure of sensitivity of acidification.
N/A
A measure estimating per capita water availability using net withdrawals
(QW) from streamflow. Method of calculation: (QS-QW)/pop.
Literature Source
[See Appendix A for full
citation]
Heinz Center, 2008.
Miller, personal
communication
Hurd et al., 1998
Duplicate
604, 151
Do Not Cite or Quote
PageB-72
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Do Not Cite or Quote Page C-l
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
This appendix describes the data sources obtained for 32 of the 53 vulnerability indicators. These
data sources were used, when possible, to create one or more maps for each indicator. Supporting
information for each data source include: the names of the data sets used; how to obtain the data;
website, when available; spatial and temporal resolution; coverage of data set (i.e. national, state,
or local); type of data source; format of data relevant to mapping (e.g., ArcGIS, Excel, Access,
etc.); and metadata (e.g., definitions of variables in data set, method of data collection, etc.). The
Additional Data Characteristics sections for each indicator provide additional detail not captured
in the preceding sections.
One indicator (marked with *) had an incomplete map. Five indicators (marked with **) were
not mappable given project resources and technical difficulties encountered, as noted in the main
report and in Appendix D. The remaining 25 indicators were mapped. The mapping
methodology for the 25 mapped indicators is presented in Appendix E, and maps for these
indicators are presented in Appendix F (by HUC-4 watershed) and Appendix H (by ecoregion).
#1 Acid Neutralizing Capacity (ANC)
Literature Source (see Appendix A for full citation):
USEPA, 2006b.
Data Sets Used:
USEPA - Wadeable Streams Assessment (WSA): Water Chemistry Data.
How To Obtain Data:
Download online
URL to Data (if any):
http ://www. epa. gov/o wow/stream survey/web data.html
Spatial Resolution:
Small streams
Temporal Resolution (period and frequency of collection):
2004-2005; every 5 years (first year of round of data collection was 2004-2005)
Extent/Coverage of Data Set:
National
Type of Data Source:
Survey
Format of Data:
Comma separated
Do Not Cite or Quote Page C-2
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Metadata:
• Definitions and data descriptions as . txt files.
USEPA. 2008. Wadeable Streams Assessment - Definitions of Variables. Available at:
http ://www. epa. gov/o wow/stream survey /web_data.html. Accessed July 21, 2009.
Additional Data Characteristics:
The WSA Water Chemistry data set contains water chemistry data for small streams, including
pH data. Data files are associated with companion text files (using EPA's WSA Definitions of
Variables in metadata) that list data set labels and give individual descriptions for each variable.
The original literature source, EPA's 2006 WSA report (USEPA, 2006b; see Appendix A for full
citation), provides an explanation of how wadeable streams were selected for this study and how
data were collected from various sites.
#22 At-Risk Freshwater Plant Communities
Literature Source (see Appendix A for full citation):
Heinz Center, 2008.
Data Sets Used:
NatureServe - Explorer (customized dataset)
How To Obtain Data:
Data were obtained from Jason McNees at NatureServe, 1101 Wilson Blvd., 15th Floor
Arlington, VA 22201 via email on July 31, 2009.
URL to Data (if any):
N/A
Spatial Resolution:
State
Temporal Resolution (period and frequency of collection):
2006; not specified
Extent/Coverage of Data Set:
National
Type of Data Source:
Census
Format of Data:
Excel
Do Not Cite or Quote Page C-3
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Metadata:
• Details of the National Vegetation Classification System (NVCS).
Federal Geographic Data Committee. International Classification of Ecological Communities
- Terrestrial Vegetation of the United States. Volume I: The National Vegetation
Classification System: Development, Status, and Applications. Available online at:
http://www.NatureServe.org/library/vol 1 .pdf Accessed July 21, 2009.
• Explanation of Conservation Status Ranks.
NatureServe and Natural Heritage. 2009. Conservation Status Ranks. Available online at:
http://www.NatureServe.org/explorer/ranking.htm. Accessed July 21, 2009.
• Explanation of the Terrestrial Ecological Classification System.
NatureServe. 2008. Terrestrial Ecological Classification System. Available online at:
http://www.NatureServe.org/publications/usEcologicalsystems.jsp.
Additional Data Characteristics:
NatureServe data, customized for EPA, were used to inform this indicator. Data include the
percent of plant species at risk of extinction in each state. In each state, species considered to be
at a risk of extinction were ones classified by NatureServe as critically imperiled, imperiled, or
vulnerable (using NatureServe and Natural Heritage's Conservation Status Ranks in metadata).
In addition, plant species were classified into plant community types based on their physiognomy
(using FGDC's NVCS) or based on their landscape settings, biological dynamics, and
environmental features (using NatureServe's Terrestrial Ecological Classification System).
#24 At-Risk Native Freshwater Species
Literature Source (see Appendix A for full citation):
Heinz Center, 2008.
Data Sets Used:
NatureServe - Explorer (customized dataset).
How To Obtain Data:
Data were obtained from Jason McNees at NatureServe, 1101 Wilson Blvd., 15th Floor
Arlington, VA 22201 via email on July 31, 2009.
URL to Data (if any):
N/A
Spatial Resolution:
State
Temporal Resolution (period and frequency of collection):
2006; not specified
Extent/Coverage of Data Set:
National
Do Not Cite or Quote Page C-4
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Type of Data Source:
Census
Format of Data:
Excel
Metadata:
• Explanation of Conservation Status Ranks.
NatureServe and Natural Heritage. 2009. Conservation Status Ranks. Available online at:
http://www.NatureServe.org/explorer/ranking.htm. Accessed July 21, 2009.
Additional Data Characteristics:
NatureServe data, customized for EPA, were used to inform this indicator. Data include the
percent of all species at risk of extinction in each state. In each state, species considered to be at
a risk of extinction were ones classified by NatureServe as critically imperiled, imperiled, or
vulnerable (using NatureServe and Natural Heritage's Conservation Status Ranks in metadata).
Note: The data set used to inform this indicator was the same as that used for the indicator #326
(Wetland and Freshwater Species at Risk (number of species)).
#57 Coastal Vulnerability Index (CVI)
Literature Source (see Appendix A for full citation):
Day etal., 2005.
Data Sets Used:
USGS - A Preliminary Database for the U.S. Atlantic, Pacific and Gulf of Mexico Coasts (U.S.
Geological Survey Digital Data Series - 68)
How To Obtain Data:
Download online
URL to Data (if any):
http://pubs.usgs.gov/dds/dds68/htmldocs/data.htm
Spatial Resolution:
1:2,000,000 shoreline at 3-minute resolution
Temporal Resolution (period and frequency of collection):
Time periods vary for the 3 different regions of United States: Atlantic Coast (2000); Pacific
Coast (2001); and Gulf Coast (2001). Variable frequency (as data on each of the six variables on
which the CVI depends are collected for different time periods and at a different frequencies).
Extent/Coverage of Data Set:
All U.S. coastline.
Do Not Cite or Quote Page C-5
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Type of Data Source:
Database
Format of Data:
ARC/INFO or ASCII
Metadata:
• Calculation of the CVI.
Thieler, R.R., Hammar-Klose, E.S., 2001. National Assessment of Coastal Vulnerability to
Sea-Level Rise: Preliminary Results for the U.S. Gulf of Mexico Coast. U.S. Geological
Survey Open-File Report 00-179. Available online at: http://pubs.usgs.gov/of/2000/ofOO-
179/index.html. Accessed July 21, 2009.
Additional Data Characteristics:
The calculation of the value of this indicator is based on six independent variables (described in
Thieler and Hammar-Klose, 2001 in metadata). The original data were housed in the Carbon
Dioxide Information Analysis Center's (CDAIC) Coastal Hazards Database(CHD) for 3
geographic regions - East Coast, West Coast, and Gulf Coast. The attributes in this dataset are
based on A Coastal Hazards Database for the U.S. Gulf Coast (Gornitz, V. and White, T. W.
1992. ORNL/CDIAC-60, NDP-043B. Oak Ridge National Laboratory, Oak Ridge, Tennessee)
updated with data from more recent sources.
#55 Commercially Important Fish Stocks**
Literature Source (see Appendix A for full citation):
Heinz Center, 2008.
Data Sets Used:
NOAA National Marine Fisheries Service (NMFS)
How To Obtain Data:
Request directly from NOAA NMFS Research Centers
URL to Data (if any):
N/A
Spatial Resolution:
Regional (e.g., Pacific Coast, Southeast, Northeast, etc.)
Temporal Resolution (period and frequency of collection):
1981-2005; at least 10 years of annual data per stock
Extent/Coverage of Data Set:
National (coastal)
Do Not Cite or Quote Page C-6
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Type of Data Source:
Census
Format of Data:
Unknown
Metadata:
N/A
Additional Data Characteristics:
The NOAA NMFS data set includes spawning stock biomass and total exploitable stock biomass
data on 109 fish stocks over a 10-year period. These data do not include near-shore stocks (i.e.,
those in state waters within 3 miles of the shore), many of which are under state management
jurisdiction, and anadromous salmon stocks from the Pacific Northwest. The Heinz Center
calculated stock trends for 109 stocks based on the estimated weight or biomass of the entire
stock using linear regression analysis to establish which stocks were increasing or decreasing by
more than 25%.
#95 Fish and Bottom-Dwelling Animals**
Literature Source (see Appendix A for full citation):
Heinz Center, 2008.
Data Sets Used:
(a) USEPA - Wadeable Streams Assessment (WSA)
(b) USEPA - Storage and Retrieval System (STORET)
How To Obtain Data:
(a) Download online
(b) Download online (if file size is large, can get via e-mail).
URL to Data (if any):
(a) http://www.epa.gov/owow/streamsurvey/web data.html
(b)http://iaspub.epa.gov/storpubl/DW_home
Spatial Resolution:
(a) N/A
(b) HUC-8 watershed level
Temporal Resolution (period and frequency of collection):
(a) 2004-2005, every 5 years (first year of round of data collection was 2004-2005)
(b) 1900-2009; daily
Extent/Coverage of Data Set:
(a) National
(b) National
Do Not Cite or Quote Page C- 7
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Type of Data Source:
(a) Survey
(b) Database
Format of Data:
(a) Comma separated
(b) Excel
Metadata:
• Definitions and data descriptions as txt files.
USEPA. 2008a. Wadeable Streams Assessment - Definitions of Variables. Available at:
http://www.epa.gov/owow/streamsurvey/web_data.html. Accessed July 21, 2009.
Additional Data Characteristics:
To assess the condition offish and bottom-dwelling animals, the Macroinvertebrate Index of
Biotic Condition was calculated (using methods in EPA's 2006 WSA report; USEPA 2006b).
This index is based on multiple metrics, such as: taxa richness, evenness of species across taxa,
the relative abundance of different taxa, the feeding strategy of taxa, the habitat preference of
taxa, and the tolerance of taxa to stressors. Data on these are available in EPA's WSA data set
and in EPA's STORET data set. Sites with index scores 75-95% lower than the reference streams
were identified as 'moderate,' whereas 'degraded' sites were those with index scores lower than
95% of the reference streams.
#725 Groundwater Reliance
Literature Source (see Appendix A for full citation):
Kurd et al., 1998; Kurd et al.,1999.
Data Sets Used:
United States Geological Survey (USGS) - National Water-Use Dataset.
How To Obtain Data:
Download online
URL to Data (if any):
http ://water.usgs. gov/watuse/
Spatial Resolution:
HUC-8 watershed
Temporal Resolution (period and frequency of collection):
1985-2000; every 5 years
Extent/Coverage of Data Set:
National
Do Not Cite or Quote Page C-8
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Type of Data Source:
Database
Format of Data:
ASCII
Metadata:
• Description of water use parameters.
USGS. Estimated Use of Water in the United States. Available online at:
http://water.usgs.gov/watuse/. Accessed December 15, 2010.
Additional Data Characteristics:
The USGS National Water-Use Dataset provides data on the total annual withdrawals from
groundwater and surface water and total annual withdrawals from only groundwater for the year
1990.
#165 Meteorological Drought Indices
Literature Source (see Appendix A for full citation):
National Assessment Synthesis Team, 2000a.
Data Sets Used:
NOAA National Climatic Data Center (NCDC) - Divisional Data on the Palmer Drought
Severity Index (PSDI).
How To Obtain Data:
Download online
URL to Data (if any):
http://www7.ncdc.noaa.gov/CDO/CDODivisionalSelect.jsp
Spatial Resolution:
344 "climate divisions," with varying number of divisions per state
Temporal Resolution (period and frequency of collection):
1895 to present; monthly
Extent/Coverage of Data Set:
National
Type of Data Source:
Database
Format of Data:
ASCII
Do Not Cite or Quote Page C-9
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Metadata:
• Description of the PSDI.
Karl, T.R., R.W. Knight, D.R. Easterling, and R.G. Quayle. 1996. Indices of climatic change
for the United States. Bulletin of the American Meteorological Society. 77 (2): 279-292.
• Data dictionary for PSDI data.
National Oceanic and Atmospheric Administration (NOAA). 2007. Time-bias Corrected
Divisional Temperature-Precipitation-Drought Index. Prepared by the National Climatic Data
Center (NCDC). Available online at: http://www7.ncdc.noaa.gov/CDO/DIV DESC.txt.
Accessed July 21,2009.
• Definitions of climate divisions.
National Oceanic and Atmospheric Administration (NOAA). 2009. Definitions of Climate
Divisions. Prepared by the National Climatic Data Center (NCDC). Available online at:
http://www4.ncdc.noaa.gov/cgi-win/wwcgi.dll7WWDI~getstate~US. Accessed July 21,
2009.
Additional Data Characteristics:
The NOAA NCDC Divisional Data on the PSDI includes the calculated value of the PSDI for
each of 344 climate divisions in the U.S. The description of the index (from Karl et al., 1996 in
metadata) and other supporting information (from NOAA's data dictionary for PSDI data and
definitions of climate divisions in metadata) are also available.
#190 Number of Dry Periods in Grassland/Shrubland Streams and Rivers
(Percent of streams with dry periods over time) *
Literature Source (see Appendix A for full citation):
Heinz Center, 2008.
Data Sets Used:
(a) USGS - Hydro Climatic Data Network (HDCN)
(b) USGS - Stream Gauge Data
How To Obtain Data:
Data (integrated data set of (a) and (b) above) were obtained from Anne Marsh, Ph. D., Director
of Observation and Understanding Programs, The H. John Heinz III Center for Science,
Economics, and the Environment, 900 17th Street, NW, Suite 700, Washington, D. C. 20006 via
email on April 28, 2009.
URL to Data (if any):
(a) N/A
(b) N/A
Spatial Resolution:
(a) HUC-8 watershed level
(b) HUC-8 watershed level
Do Not Cite or Quote Page C-10
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Temporal Resolution (period and frequency of collection):
(a) 1874-1988 (inclusive); daily, monthly, and annual
(b) 1899-2007; daily, monthly, annual
Extent/Coverage of Data Set:
(a) National
(b) National
Type of Data Source:
(a) Database
(b) Database
Format of Data:
(a) ASCII;
(b) .rrd or .img or .xml or .sgml
Metadata:
• Ecoregion definitions.
Bailey, R. G. 1995. Description of the Ecoregions of the United States. Available online at:
http://www.fs.fed.us/land/ecosysmgmt/. Accessed July 21, 2009.
• National Land Cover Dataset (NLCD) - Landcover categories.
USEPA. Undated. National Land Cover Dataset (NLCD) - Landcover categories. Available
online at: http://www.mrlc.gov/index.php. Accessed July 21, 2009.
Additional Data Characteristics:
Data were obtained from the two different datasets: daily streamflow data at study sites from
USGS' Stream Gauge Data data set, and daily streamflow data for reference sites in relatively
undisturbed locations from USGS' HDCN data set. Data were integrated into a single data set by
the Heinz Center; this integrated data set was ultimately used for mapping. In addition, ecoregion
definitions (from Bailey, 1995 in metadata) and land cover region definitions (from USEPA's
NLCD Landcover Categories in metadata) were used to define watersheds in grasslands and
shrublands.
#209 Population (human) Susceptible to Flood Risk**
Literature Source (see Appendix A for full citation):
Kurd et al., 1998; Kurd et al.,1999.
Data Sets Used:
(a) FEMA Q3 Flood Data
(b) ESRI ArcUSA - US Census tract data.
How To Obtain Data:
(a) 26 CDs (for all USA) available for a fee, online or on phone.
(b) Available on ArcGIS.
Do Not Cite or Quote Page C-ll
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
URL to Data (if any):
(a) http://www.fetna.gov/hazard/tnap/q3.shttntfO
(b) N/A
Spatial Resolution:
(a) Individual street addresses
(b) 2 scales - 1:2,000,000 and 1:25,000,000
Temporal Resolution (period and frequency of collection):
(a) N/A
(b) N/A
Extent/Coverage of Data Set:
(a) National
(b) National
Type of Data Source:
(a) Collection of flood maps
(b) Census
Format of Data:
(a) Digital Line Graph (DLG), ARC/INFO, Maplnfo
(b) ARC/INFO (ArcView and ArcGIS compatible)
Metadata:
N/A
Additional Data Characteristics:
500-year floodplain data from FEMA's Q3 Digital Flood Data were overlaid with U.S. Census
Tract Data from ArcGIS.
#218 Ratio of Snow to Total Precipitation
Literature Source (see Appendix A for full citation):
Lettenmaier et al., 2008.
Data Sets Used:
Monthly Climate Data and Observation Station Locations from National Climatic Data Center.
How To Obtain Data:
Download online (free for .gov and .edu domains).
URL to Data (if any):
http://gis.ncdc.noaa.gov/snowfallmo/
Do Not Cite or Quote Page C-12
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Spatial Resolution:
18,116 stations
Temporal Resolution (period and frequency of collection):
1867 to present; annual
Extent/Coverage of Data Set:
National
Type of Data Source:
Database
Format of Data:
Tabular, Delimited, Time Series, NWS Condensed B (suitable for SWMM), Worksheet, dBase,
Binary, NCDC
Metadata:
N/A
Additional Data Characteristics:
NOAA's Monthyl Climate Data and Observation Locations data set contains climate data (rain,
snow, evaporation, temperature, degree days). Based on available station-level time-series data,
the unweighted numeric average for each HUC-4 can be calculated using stations within that
HUC watershed.
#219 Ratio of Withdrawals to Stream Flow
Literature Source (see Appendix A for full citation):
Kurd etal., 1999.
Data Sets Used:
(a) Oregon State University - PRISM Climate Modeling System: Mean Annual Precipitation
data
(b) Oregon State University - PRISM Climate Modeling System: Mean Daily Maximum
Temperature data
(c) USGS - National Water-Use Dataset.
How To Obtain Data:
Download online
URL to Data (if any):
(a) http://www.prism.oregonstate.edu/products/matrix.phtml
(b) http://www.prism.oregonstate.edu/products/matrix.phtml
(c) http://water.usgs.gov/watuse/
Do Not Cite or Quote Page C-13
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Spatial Resolution:
(a) 30 arc-second (800 meters)
(b) 30 arc-second (800 meters)
(c) HUC-8 watershed
Temporal Resolution (period and frequency of collection):
(a) 1971-2000; monthly
(b) 1971-2000; monthly
(c) 1985 - 2000; every 5 years
Extent/Coverage of Data Set:
(a) National
(b) National
Type of Data Source:
(a) Interpolated grid
(b) Interpolated grid
(c) Database
Format of Data:
ASCII
Metadata:
• Calculation of stream/low.
Vogel, R.M., I. Wilson, and C. Daly. 1999. Regional Regression Models of Annual
Streamflow for the United States." Journal of Irrigation and Drainage Engineering. 125 (3):
148-157.
• Metadata for PRISM U.S. average monthly or annual precipitation data.
http://www.prism.oregonstate.edu/docs/meta/ppt 30s meta.htm
• Metadata for PRISM U.S. average monthly temperature data.
http://www.climatesource.com/us/fact_sheets/meta_tmin_us_71b.html
• Description of water use parameters.
USGS. Estimated Use of Water in the United States. Available online at:
http://water.usgs.gov/watuse/. Accessed December 15, 2010.
Additional Data Characteristics:
As described in the metadata for (b), mean annual temperature was calculated as the average of
the mean maximum and mean minum temperature for a given location.
Do Not Cite or Quote Page C-14
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
#284 Stream Habitat Quality
Literature Source (see Appendix A for full citation):
Heinz Center, 2008.
Data Sets Used:
USEPA - Wadeable Streams Assessment (WSA).
How To Obtain Data:
Download online
URL to Data (if any):
http ://www. epa. gov/o wow/stream survey/web data.html
Spatial Resolution:
Small streams
Temporal Resolution (period and frequency of collection):
2004-2005; every 5 years (first year of round of data collection was 2004-2005)
Extent/Coverage of Data Set:
National
Type of Data Source:
Survey
Format of Data:
Comma separated
Metadata:
• Definitions and data descriptions as . txt files.
USEPA. 2008. Wadeable Streams Assessment - Definitions of Variables. Available at:
http://www.epa.gov/owow/streamsurvey/web_data.html. Accessed July 21, 2009.
Additional Data Characteristics:
Data on the biological condition of small streams, water quality, and biological data on various
stream types were used to inform this indicator.
#322 WaterborneHuman Disease Outbreaks (events)**
Literature Source (see Appendix A for full citation):
Heinz Center, 2008.
Data Sets Used:
Centers for Disease Control and Prevention (CDC) - Waterborne Disease and Outbreak
Surveillance System (WBDOSS).
Do Not Cite or Quote Page C-15
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
How To Obtain Data:
Download online
URL to Data (if any):
http://www.cdc.gov/healthywater/statistics/wbdoss/surveillance.html
Spatial Resolution:
State
Temporal Resolution (period and frequency of collection):
1985-2004, yearly or every 2 to 3 years
Extent/Coverage of Data Set:
National
Type of Data Source:
Data summaries
Format of Data:
PDF and web page
Metadata:
N/A
Additional Data Characteristics:
Data from the CDC's WBDOSS on the number of waterborne human disease outbreaks in both
drinking and recreational water, listed by state, by etiologic agent and type of water system, by
deficiency/type of exposure and type of water system, are used to inform this indicator.
#325 Wetland Loss**
Literature Source (see Appendix A for full citation):
MEA, 2005b.
Data Sets Used:
(a) USFWS - National Wetlands Inventory (NWI)
(b) NatureServe Explorer
How To Obtain Data:
(a) Download online
(b) Download online
URL to Data (if any):
(a) http://www.fws.gov/wetlands/Data/DataDownload.html
(b) http://www.NatureServe. org/explorer/
Do Not Cite or Quote Page C-16
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Spatial Resolution:
(a) Lower 48 states (USGS 1:24,000 or 1:100,000 topographic quadrangle), Alaska (USGS
1:63,000 topographic quadrangle), Hawaii (County and USGS topographic quadrangle),
Puerto Rico and USVI (County and USGS topographic quadrangle), Pacific Trust Territories
(County/Island), Data are in decimal degrees on the North American Datum of 1983
(b) State or by NLCD 2001 map-zones
Temporal Resolution (period and frequency of collection):
(a) N/A
(b) N/A
Extent/Coverage of Data Set:
(a) National
(b) National
Type of Data Source:
(a) Database
(b) Census
Format of Data:
(a) NWI - digital wetlands polygon, .sgml and .xml
(b) .xls or .xml
Metadata:
• National Vegetation Classification System (NVCS).
Federal Geographic Data Committee. 2009. National Vegetation Classification System.
Available online at: http://biology.usgs.gov/npsveg/nvcs.html. Accessed on November 19,
2009.
• Terrestrial Ecological Classification System.
Nature Serve/Natural Heritage. 2009. Terrestrial Ecological Classification System. Available
online at: http://www.natureserve.org/explorer/classeco.htm. Accessed on November 19,
2009.
Additional Data Characteristics:
Data from two datasets were used to inform this indicator: data on wetland vegetation and
hydrologic properties of wetlands from NWI, and plant species vulnerability rankings and data
on their abundance in U.S. states from NatureServe Explorer. Vulnerability rankings (from
Natural Heritage) were used to estimate relative susceptibility to extinction. In addition, plant
species were classified into plant community types based on their physiognomy (using NVCS) or
based on their landscape settings, biological dynamics, and environmental features (using
Terrestrial Ecological Classification System).
#326 Wetland and Freshwater Species at Risk (number of species)
Literature Source (see Appendix A for full citation):
Hurdetal., 1998.
Do Not Cite or Quote Page C-l 7
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Data Sets Used:
NatureServe - Explorer (customized dataset).
How To Obtain Data:
Data were obtained from Jason McNees at NatureServe, 1101 Wilson Blvd., 15th Floor
Arlington, VA 22201 via email on July 31, 2009.
URL to Data (if any):
N/A
Spatial Resolution:
State
Temporal Resolution (period and frequency of collection):
2006; not specified
Extent/Coverage of Data Set:
National
Type of Data Source:
Census
Format of Data:
Excel
Metadata:
• Explanation of Conservation Status Ranks.
NatureServe and Natural Heritage. 2009. Conservation Status Ranks. Available online at:
http://www.NatureServe.org/explorer/ranking.htm. Accessed July 21, 2009.
Additional Data Characteristics:
Data on the number of at-risk water-dependent species were originally compiled by the Natural
Heritage Data Centers and The Nature Conservancy for the EPA's Index of Watershed Indicators
(IWI), a study published in 1997. The IWI project was discontinued due to lack of adequate
funding. Therefore, NatureServe data, customized for EPA, were used to inform this indicator.
Data include the percent of all species at risk of extinction in each state. In each state, species
considered to be at a risk of extinction were ones classified by NatureServe as critically
imperiled, imperiled, or vulnerable (using NatureServe and Natural Heritage's Conservation
Status Ranks in metadata).
Note: The data set used to inform this indicator was identical to that used for the indicator #24
(At-Risk Native Freshwater Species).
#348 Erosion Rate
Literature Source (see Appendix A for full citation):
Murdoch et al., 2000.
Do Not Cite or Quote Page C-18
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Data Sets Used:
Soil erosion rates estimated with the Revised Universal Soil Loss Equation (RUSLE)
How To Obtain Data:
Request directly from:
Dawen Yang, PhD, Professor
Department of Hydraulic Engineering
Tsinghua University
Beijing 100084, China
Tel: +86-10-62796976
Fax: +86-10-62796971
E-mail: yangdw@tsinghua.edu.cn
URL to Data (if any):
N/A
Spatial Resolution:
0.5 degree grid
Temporal Resolution (period and frequency of collection):
1980; not applicable (modeled data)
Extent/Coverage of Data Set:
Global
Type of Data Source:
Model output
Format of Data:
ASCII grid
Metadata:
• Description ofdataset development.
Yang, D. W., S. Kanae, T. Oki, T. Koike, and K. Musiake. 2003. Global potential soil
erosion with reference to land use and climate changes. HydrologicalProcesses. 17:2913-
2928.
Additional Data Characteristics:
Erosion Rate was estimated with the Revised Universal Soil Loss Equation (RUSLE). RUSLE
does not account for deposition of eroded soil, so this indicator is more precisely defined as soil
movement. The literature source also describes variations of this dataset that incorporate
simulations of future climatic conditions (see Yang et al. (2003) for details). The study also
includes future soil erosion predictions based on changes in precipitation, temperature, and land
cover.
Do Not Cite or Quote Page C-19
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
#351 Instream Use/Total Stream/low
Literature Source (see Appendix A for full citation):
Meyeretal.,1999.
Data Sets Used:
(a) USGS - National Water-Use Dataset
(b) USGS - Mean annual runoff data
(c) WRC (U.S. Water Resources Council) - Groundwater Recharge. From: WRC. 1978. The
Nation's Water Resources: 1975-2000 (Vol. 2). U.S. Government Printing Office, Washington
D.C.
How To Obtain Data:
(a) Download online
(b) Download online
(c) Hardcopy book (obtained from library).
URL to Data (if any):
(a) http://water.usgs.gov/watuse/
(b) http://water.usgs.gov/GIS/metadata/usgswrd/XML/runoff.xml
(c) N/A
Spatial Resolution:
(a) HUC-8 watershed
(b) 1:5,000,000 (runoff of tributary streams)
(c) HUC-2 regions
Temporal Resolution (period and frequency of collection):
(a) 1985-2000; every 5 years
(b) 1951-1980; one-time effort
(c) 1975; one-time effort
Extent/Coverage of Data Set:
(a) National
(b) National
(c) National
Type of Data Source:
(a) Database
(b) Modeled dataset
(c) Published report
Format of Data:
(a) ASCII
(b) ArcGIS file (.eOO)
(c) Tabular (hard copy)
Do Not Cite or Quote Page C-20
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Metadata:
• Description of water use parameters.
USGS. Estimated Use of Water in the United States. Available online at:
http://water.usgs.gov/watuse/. Accessed December 15, 2010.
Additional Data Characteristics:
Data for this indicator were derived from three sources: data on total groundwater withdrawals
for the year 1995 from USGS' National Water-Use Dataset, modeled mean annual runoff
estimates based on annual streamflow data from USGS' mean annual runoff dataset, and
groundwater recharge estimates from WRC, 1978. The ratio of in-stream use to total streamflow
was then calculated as described in the WRC report (1978).
#352 Total Use/Total Streamflow
Literature Source (see Appendix A for full citation):
Meyeretal.,1999.
Data Sets Used:
(a) USGS - National Water-Use Dataset
(b) USGS - Mean annual runoff data
(c) WRC (U.S. Water Resources Council) - Groundwater Recharge. From: WRC. 1978. The
Nation's Water Resources: 1975-2000 (Vol. 2). U.S. Government Printing Office,
Washington D.C.
How To Obtain Data:
(a) Download online
(b) Download online
(c) Hardcopy book (obtained from library).
URL to Data (if any):
(a) http://water.usgs.gov/watuse/
(b) http://water.usgs.gov/GIS/metadata/usgswrd/XML/runoff.xml
(c) N/A
Spatial Resolution:
(a) HUC-8 watershed
(b) 1:5,000,000 (runoff of tributary streams)
(c) HUC-2 regions
Temporal Resolution (period and frequency of collection):
(a) 1985-2000; every 5 years
(b) 1951-1980; 1 time effort
(c) 1975; one-time effort
Do Not Cite or Quote Page C-21
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Extent/Coverage of Data Set:
(a) National
(b) National
(c) National
Type of Data Source:
(a) Database
(b) Modeled dataset
(c) Published report
Format of Data:
(a) ASCII
(b) ArcGIS file (.eOO)
(c) Tabular (hard copy)
Metadata:
• Description of water use parameters.
USGS. Estimated Use of Water in the United States. Available online at:
http://water.usgs.gov/watuse/. Accessed December 15, 2010.
Additional Data Characteristics:
Data for this indicator were derived from three sources: data on total groundwater withdrawals
and consumptive use for the year 1995 from USGS' National Water-Use Dataset, modeled mean
annual runoff estimates based on annual streamflow data from USGS' mean annual runoff
dataset, and groundwater recharge estimates from WRC, 1978. The ratio of total use to total
streamflow was then calculated as described in the WRC report (1978).
#364 Pesticide Toxicity Index (PTI)
Literature Source (see Appendix A for full citation):
Gilliom et al., 2006.
Data Sets Used:
USGS - NAWQA.
How To Obtain Data:
Download online
URL to Data (if any):
http://water.usgs.gov/nawqa/pnsp/pubs/circl291/appendix6/
Spatial Resolution:
51 study units
Do Not Cite or Quote Page C-22
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Temporal Resolution (period and frequency of collection):
1991 (20 study units); 1994 (16 study units); 1997 (15 study units); Variable frequency (from
one-time collection to daily depending on purpose and collection site)
Extent/Coverage of Data Set:
National
Type of Data Source:
Database
Format of Data:
Comma separated or Excel
Metadata:
• NA WQA Study Description.
USGS. 2006. About NAWQA Study Units. Available online at:
http://water.usgs.gov/nawqa/studies/study units.html. Accessed July 21, 2009.
• Daphnia EC50 Values.
USEPA. 2009. ECOTOX Database. Available online at: http://cfpub.epa.gov/ecotox/.
Accessed September 11, 2009.
Additional Data Characteristics:
The USGS NAWQA data set, which contained data on the occurrence of 76 pesticides (including
herbicides and insecticides) and 7 pesticide by-products in streams and shallow groundwater
(100ft or less below ground level) at 20 USGS study sites in 1991, 1994, and 1997, was used to
inform this indicator. Descriptions of study sites and their year of assessment were also available
(from USGS's About NAWQA webpage in metadata). The Pesticide Toxicity Index (PTI)
accounts for multiple pesticides in a sample, including pesticides without established
benchmarks for aquatic life. The PTI combines information on measured concentrations of
pesticides in stream water with toxicity estimates (i.e. toxicity quotients calculated based on
laboratory toxicity studies, e.g. EC50 values for Daphnia from EPA's ECOTOX database) to
produce a relative index value for a sample or stream. The PTI value is computed for each
sample of stream water by summing the toxicity quotients for all pesticides detected in the
sample.
#367 Herbicide Concentrations in Streams
Literature Source (see Appendix A for full citation):
USGS, 1999.
Data Sets Used:
USGS - NAWQA.
How To Obtain Data:
Download online
Do Not Cite or Quote Page C-23
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
URL to Data (if any):
http://water.usgs.gov/nawqa/pnsp/pubs/circl291/appendix6/
Spatial Resolution:
51 study units
Temporal Resolution (period and frequency of collection):
1991 (20 study units); 1994 (16 study units); 1997 (15 study units); Variable frequency (from
one-time collection to daily depending on purpose and collection site)
Extent/Coverage of Data Set:
National
Type of Data Source:
Database
Format of Data:
Comma separated or Excel
Metadata:
• NA WQA Study Description.
USGS. 2006. About NAWQA Study Units. Available online at:
http://water.usgs.gov/nawqa/studies/study units.html. Accessed July 21, 2009.
Additional Data Characteristics:
The USGS NAWQA data set, which contained data on the occurrence of 76 pesticides (including
herbicides and insecticides) and 7 pesticide by-products in streams and shallow groundwater
(100ft or less below ground level) at 20 USGS study sites in 1991, 1994, and 1997, was used to
inform this indicator. Descriptions of study sites and their year of assessment were also available
(from USGS's About NAWQA webpage in metadata).
#369 Insecticide Concentrations in Streams
Literature Source (see Appendix A for full citation):
USGS, 1999.
Data Sets Used:
USGS - NAWQA.
How To Obtain Data:
Download online
URL to Data (if any):
http://water.usgs.gov/nawqa/pnsp/pubs/circl291/appendix6/
Spatial Resolution:
51 study units
Do Not Cite or Quote Page C-24
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Temporal Resolution (period and frequency of collection):
1991 (20 study units); 1994 (16 study units); 1997 (15 study units); Variable frequency (from
one-time collection to daily depending on purpose and collection site)
Extent/Coverage of Data Set:
National
Type of Data Source:
Database
Format of Data:
Comma separated or Excel
Metadata:
• NA WQA Study Description.
USGS. 2006. About NAWQA Study Units. Available online at:
http://water.usgs.gov/nawqa/studies/study units.html. Accessed July 21, 2009.
Additional Data Characteristics:
The USGS NAWQA data set, which contained data on the occurrence of 76 pesticides (including
herbicides and insecticides) and 7 pesticide by-products in streams and shallow groundwater
(100ft or less below ground level) at 20 USGS study sites in 1991, 1994, and 1997, was used to
inform this indicator. Descriptions of study sites and their year of assessment were also available
(from USGS's About NAWQA webpage in metadata).
#371 Organochlorines in Bed Sediment
Literature Source (see Appendix A for full citation):
USGS, 1999.
Data Sets Used:
USGS - National Water-Quality Assessment (NAWQA).
How To Obtain Data:
Download online
URL to Data (if any):
http://water.usgs.gov/nawqa/pnsp/pubs/circl291/appendix6/
Spatial Resolution:
51 study units
Temporal Resolution (period and frequency of collection):
1991 (20 study units); 1994 (16 study units); 1997 (15 study units); Variable frequency (from
one-time collection to daily depending on purpose and collection site)
Do Not Cite or Quote Page C-25
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Extent/Coverage of Data Set:
National
Type of Data Source:
Database
Format of Data:
Comma separated or Excel
Metadata:
• NA WQA Study Description.
USGS. 2006. About NAWQA Study Units. Available online at:
http://water.usgs.gov/nawqa/studies/study_units.html. Accessed July 21, 2009.
Additional Data Characteristics:
The USGS NAWQA data set, which contained data on the occurrence of 32 compounds (8
individual parent compounds, 1 individual breakdown product, and 7 groups of parent
compounds, plus related breakdown products or chemical impurities in the manufactured
product) in bed sediment at 20 USGS study sites in 1991, 1994, and 1997, was used to inform
this indicator. Descriptions of study sites and their year of assessment were also available (from
USGS's About NAWQA webpage in metadata).
#373 Herbicides in Groundwater
Literature Source (see Appendix A for full citation):
USGS, 1999.
Data Sets Used:
USGS - NAWQA.
How To Obtain Data:
Download online
URL to Data (if any):
http://water.usgs.gov/nawqa/pnsp/pubs/circl291/appendix6/
Spatial Resolution:
51 study units
Temporal Resolution (period and frequency of collection):
1991 (20 study units); 1994 (16 study units); 1997 (15 study units); Variable frequency (from
one-time collection to daily depending on purpose and collection site)
Extent/Coverage of Data Set:
National
Do Not Cite or Quote Page C-26
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Type of Data Source:
Database
Format of Data:
Comma separated or Excel
Metadata:
• NA WQA Study Description.
USGS. 2006. About NAWQA Study Units. Available online at:
http://water.usgs.gov/nawqa/studies/study_units.html. Accessed July 21, 2009.
Additional Data Characteristics:
The USGS NAWQA data set, which contained data on the occurrence of 76 pesticides (including
herbicides and insecticides) and 7 pesticide by-products in streams and shallow groundwater
(100ft or less below ground level) at 20 USGS study sites in 1991, 1994, and 1997, was used to
inform this indicator. Descriptions of study sites and their year of assessment were also available
(from USGS's About NAWQA webpage in metadata).
#374 Insecticides in Groundwater
Literature Source (see Appendix A for full citation):
USGS, 1999.
Data Sets Used:
USGS - NAWQA.
How To Obtain Data:
Download online
URL to Data (if any):
http://water.usgs.gov/nawqa/pnsp/pubs/circl291/appendix6/
Spatial Resolution:
51 study units
Temporal Resolution (period and frequency of collection):
1991 (20 study units); 1994 (16 study units); 1997 (15 study units); Variable frequency (from
one-time collection to daily depending on purpose and collection site)
Extent/Coverage of Data Set:
National
Type of Data Source:
Database
Format of Data:
Comma separated or Excel
Do Not Cite or Quote Page C-27
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Metadata:
• NA WQA Study Description.
USGS. 2006. About NAWQA Study Units. Available online at:
http://water.usgs.gov/nawqa/studies/study_units.html. Accessed July 21, 2009.
Additional Data Characteristics:
The USGS NAWQA data set, which contained data on the occurrence of 76 pesticides (including
herbicides and insecticides) and 7 pesticide by-products in streams and shallow groundwater
(100ft or less below ground level) at 20 USGS study sites in 1991, 1994, and 1997, was used to
inform this indicator. Descriptions of study sites and their year of assessment were also available
(from USGS's About NAWQA webpage in metadata).
#437Precipitation Elasticity of Stream/low
Literature Source (see Appendix A for full citation):
Sankarasubramanian et al., 2001.
Data Sets Used:
(a) USGS - Hydroclimatic Data Network (HDCN): Streamflow data
(b) Oregon State University - PRISM Climate Modeling System: Mean Annual Precipitation data
How To Obtain Data:
(a) Download online
(b) Download online
URL to Data (if any):
(a) http://pubs.usgs.gov/of/1992/ofr92-129/hcdn92/hcdn/ascii/
(b) http://www.prism.oregonstate.edu/
Spatial Resolution:
(a) HUC-8
(b) 2.5 arc-min
Temporal Resolution (period and frequency of collection):
(a) 1874- 1988; annual
(b) 1990 - present; monthly
Extent/Coverage of Data Set:
(a) National
(b) National
Type of Data Source:
(a) Database
(b) Database
Do Not Cite or Quote Page C-28
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Format of Data:
(a) ASCII
(b) ASCII
Metadata:
• Description of HDCN data.
Landwehr, J. M., and J. R. Slack, 1992. Hydro-Climatic Data Network: A U.S. Geological
Survey Streamflow Data Set for the United States for the Study of Climate Variations, 1874-
1988. U.S. Geological Survey Open-File Report 92-129. Available online at:
http://pubs.er.usgs.gov/usgspubs/ofr/ofr92129. Accessed July 21, 2009.
• Description of PRISM modeling.
Daly, C., R. P. Neilson, and D. L. Phillips. 1994. "A Statistical-Topographic Model for
Mapping Climatological Precipitation over Mountainous Terrain." Journal of Applied
Meterology. 33: 140-158.
• Calculation of precipitation elasticity of streamflow.
Sankarasubramanian, A., R. M. Vogel, and J. F. Limbrunner. 2001. Climate Elasticity of
Streamflow in the United States. Water Resources Research. 37 (6): 1771-1781.
Additional Data Characteristics:
Data on streamflow were obtained from USGS's HDCN data set. Data on precipitation were
obtained from Oregon State University's PRISM data set. Explanations for how data in HDCN
and PRISM were collected and/or modeled are also available (from Landwehr and Slack, 1992
and Daly et al., 1994 in metadata). For the purposes of mapping this indicator with relative ease,
Figure 4 in the original literature source, Sankarasubramanian et al., 2001 (see Appendix A for
full reference) was used. However, the complete original data set could also be recalculated for
mapping purposes using the data sources listed here.
#449 Ratio of Reservoir Storage to Mean Annual Runoff
Literature Source (see Appendix A for full citation):
Lettenmaier et al., 2008.
Data Sets Used:
(a) US Army Corps of Engineers (USAGE) - National Inventory of Dams (NID)
(b) USGS - Mean annual runoff data
How To Obtain Data:
(a) Download online (need to register for free)
(b) Download online
URL to Data (if any):
(a) https://nid.usace.army.mil (Note: You must accept the security certificate in order to view
database) OR http://www.nationalatlas.gov/mld/damsOOx.html
(b) http://water.usgs.gov/GIS/metadata/usgswrd/XML/runoff.xml
Do Not Cite or Quote Page C-29
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Spatial Resolution:
(a) 80,000 dams (latitude and longitude specifications included only major dams available at
National Atlas site)
(b) 1:5,000,000 (runoff of tributary streams)
Temporal Resolution (period and frequency of collection):
(a) 1972-2006; unclear how often it is updated
(b) 1951-1980; 1 time effort
Extent/Coverage of Data Set:
(a) National
(b) National
Type of Data Source:
(a) Database
(b) Modeled dataset
Format of Data:
(a) Tabular (need to copy-paste into Excel)
(b) ArcGIS file (.eOO)
Metadata:
• Calculation of ratio of reservoir storage to mean annual runoff.
Graf, W.L., 1999: Dam nation: A geographic census of American dams and their large-scale
hydrologic impacts, Water Resources Research, 35 (4): 1305-1311.
Additional Data Characteristics:
Data for this indicator were derived from two sources: dam reservoir storage data from the
USACE's NID data set and modeled mean annual runoff estimates based on annual streamflow
data from USGS' mean annual runoff dataset. The storage to runoff ratio can then be calculated
(as specified in Graf, 1999 in metadata).
#453 Runoff Variability
Literature Source (see Appendix A for full citation):
Lettenmaier et al., 2008.
Data Sets Used:
University of Washington (Land Surface Hydrology Research Group) - Variable Infiltration
Capacity (VIC) Land Surface Data Set.
How To Obtain Data:
Download online (need to register for free).
Do Not Cite or Quote Page C-30
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
URL to Data (if any):
http://www.hydro.washington.edu/SurfaceWaterGroup/Data/VIC retrospective/index.html (file
is:
ftp://ftp.hydro.washington.edu/pub/CE/HYDRO/nijssen/vic_global/calibrated/runoff.calibrated.
monthly. 1980_1993. nc.gz)
Spatial Resolution:
2 degrees
Temporal Resolution (period and frequency of collection):
1950-2000; 3-hourly
Extent/Coverage of Data Set:
National
Type of Data Source:
Modeled dataset
Format of Data:
NetCDF
Metadata:
• Explanation of VIC model.
Maurer, E.P., A.W. Wood, J.C. Adam, D.P. Lettenmaier, and B. Nijssen, 2002. A long-term
hydrologically-based data set of land surface fluxes and states for the conterminous United
States, Journal of Climate, 15: 3237-3251.
• VIC Global Hydrologic Simulations - Variable descriptions.
University of Washington (Land Surface Hydrology Research Group). Undated. VIC Global
Hydrologic Simulations. Available online at:
http://www.hydro.washington.edu/SurfaceWaterGroup/Data/vic global.html. Accessed July
21, 2009.
Additional Data Characteristics:
Data on runoff are available from the output of the University of Washington's VIC model
which simulates the global land surface hydrological cycle. The model outputs include five
model-derived variables: evapotranspiration, snow-water equivalent, soil moisture storage, total
storage, and runoff (which is the variable of interest). Detailed descriptions of input variables and
output parameters of this model are also availabled (from Maurer et al., 2002 in metadata).
Do Not Cite or Quote Page C-31
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
#460 Macroinvertebrate Index ofBiotic Condition
Literature Source (see Appendix A for full citation):
USEPA, 2006b.
Data Sets Used:
USEPA - Wadeable Streams Assesment (WSA) (Stream Water Benthic Macroinvertebrate
Metrics).
How To Obtain Data:
Download online
URL to Data (if any):
http://www.epa.gov/owow/streamsurvey/web_data.html (file is: wsa_benmet300_ts_fmal.csv)
Spatial Resolution:
Small streams
Temporal Resolution (period and frequency of collection):
2004-2005; every 5 years (first year of round of data collection was 2004-2005)
Extent/Coverage of Data Set:
National
Type of Data Source:
Survey
Format of Data:
Comma separated
Metadata:
• Definitions and data descriptions as . txt files.
USEPA. 2008. Wadeable Streams Assessment - Definitions of Variables. Available at:
http ://www. epa. gov/o wow/stream survey/web data.html. Accessed July 21, 2009.
• Site information for WSA sites.
USEPA 2008. Wadeable Streams Assessment - Post-Sampling Site Info and Survey Design.
Available at: http ://www. epa. gov/o wow/stream survey/web data.html. Accessed July 21,
2009. (file is: wsa_siteinfo_ts_fmal.csv).
Additional Data Characteristics:
Data on counts of individual benthic macroinvertebrates at various stages of their life cycle were
obtained from EPA's WSA Benthic Macroinvertebrate Metrics data set. Data files are associated
with companion text files (using EPA's WSA Definitions of Variables in metadata) that list data
set labels and give individual descriptions for each variable. The original literature source, EPA's
2006 WSA report (USEPA, 2006b; see Appendix A for full citation), provides an explanation of
how wadeable streams were selected for this study and how data were collected from various
sites.
Do Not Cite or Quote Page C-32
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
#461 Macroinvertebrate Observed/Expected (O/E) Ratio ofTaxa Loss
Literature Source (see Appendix A for full citation):
USEPA, 2006b.
Data Sets Used:
USEPA - Wadeable Streams Assessment (Stream Water Benthic Macroinvertebrate Metrics).
How To Obtain Data:
Download online
URL to Data (if any):
http ://www. epa. gov/o wow/stream survey/web data.html (file is: wsa_benmet300_ts_fmal.csv)
Spatial Resolution:
Small streams
Temporal Resolution (period and frequency of collection):
2004-2005; every 5 years (first year of round of data collection was 2004-2005)
Extent/Coverage of Data Set:
National
Type of Data Source:
Survey
Format of Data:
Comma separated
Metadata:
• Definitions and data descriptions as . txt files.
USEPA. 2008. Wadeable Streams Assessment - Definitions of Variables. Available at:
http://www.epa.gov/owow/streamsurvey/web_data.html. Accessed July 21, 2009.
Additional Data Characteristics:
Data on percent diversity of benthic macroinvertebrates were obtained from EPA's WSA
Benthic Macroinvertebrate Metrics data set. Data files are associated with companion text files
(using EPA's WSA Definitions of Variables in metadata) that list data set labels and give
individual descriptions for each variable. The original literature source, EPA's 2006 WSA report
(USEPA, 2006b; see Appendix A for full citation), provides an explanation of how wadeable
streams were selected for this study and how data were collected from various sites.
Do Not Cite or Quote Page C-33
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
#462 CoastalBenthic Communities**
Literature Source (see Appendix A for full citation):
USEPA, 2008b.
Data Sets Used:
Underlying sampling data in USEPA's National Coastal Assessment (NCA) database.
How To Obtain Data:
Download online
URL to Data (if any):
http ://www. epa. gov/emap/nca/html/data/index.html
Spatial Resolution:
Not clear (EPA's website states that there are "thousands" of sampling sites, but no specific
number; data contains latitude/longitude specs.)
Temporal Resolution (period and frequency of collection):
1990-2002; no defined frequency (1 datum/site for only year listed)
Extent/Coverage of Data Set:
National
Type of Data Source:
Database
Format of Data:
Excel
Metadata:
• Definition and calculation ofbenthic index.
USEPA. 2005. National Coastal Condition Report (NCCR). EPA-620/R-03/002. Available
online at: http://www.epa.gov/owow/oceans/nccr/2005/index.html. Accessed July 21, 2009.
Additional Data Characteristics:
The coastal benthic communities index used in this indicator is based on multiple independent
variables (described in EPA's NCCR in metadata). Data for these independent variables,
(including total count of taxa, total abundance, mean abundance, mean biomass, total biomass,
and diversity index) can be obtained from the underlying sampling data in EPA's NCA dataset.
#623 Water Availability: Net Stream Flow per Capita
Literature Source (see Appendix A for full citation):
Kurd etal., 1999.
Do Not Cite or Quote Page C-34
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Data Sets Used:
(a) Oregon State University - PRISM Climate Modeling System: Mean Annual Precipitation
data
(b) Oregon State University - PRISM Climate Modeling System: Mean Daily Maximum
Temperature data
(c) USGS - National Water-Use Dataset.
How To Obtain Data:
Download online
URL to Data (if any):
(a) http://www.prism.oregonstate.edu/products/matrix.phtml
(b) http://www.prism.oregonstate.edu/products/matrix.phtml
(c) http://water.usgs.gov/watuse/
Spatial Resolution:
(a) 30 arc-second (800 meters)
(b) 30 arc-second (800 meters)
(c) HUC-8 watershed
Temporal Resolution (period and frequency of collection):
(a) 1971-2000; monthly
(b) 1971-2000; monthly
(c) 1985 - 2000; every 5 years
Extent/Coverage of Data Set:
(a) National
(b) National
(c) National
Type of Data Source:
(a) Interpolated grid
(b) Interpolated grid
(c) Database
Format of Data:
ASCII
Metadata:
• Calculation of stream/low.
Vogel, R.M., I. Wilson, and C. Daly. 1999. "Regional Regression Models of Annual
Streamflow for the United States." Journal of Irrigation and Drainage Engineering. 125(3):
148-157.
• Metadata for PRISM U.S. average monthly or annual precipitation data.
http://www.prism.oregonstate.edu/docs/meta/ppt_30s_meta.htm
• Metadata for PRISM U.S. average monthly temperature data.
http://www.climatesource.com/us/fact sheets/meta tmin us 71b.html
Do Not Cite or Quote Page C-35
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
• Description of water use parameters.
USGS. Estimated Use of Water in the United States. Available online at:
http://water.usgs.gov/watuse/. Accessed December 15, 2010.
Additional Data Characteristics:
As described in the metadata for (b), mean annual temperature was calculated as the average of
the mean maximum and mean minum temperature for a given location.
Do Not Cite or Quote Page C-36
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Do Not Cite or Quote Page D-l
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
The following appendix is a compilation of technical notes for the 32 vulnerability indicators for
which data sources were readily available. The issues presented below may affect the
interpretation of an indicator map (if the indicator was mappable) or may provide guidance for
future mapping efforts (if the indicator was non-mappable or had an incomplete map). Data
sources for these 32 indicators are presented in Appendix C.
One indicators (marked with *) had an incomplete map. Five indicators (marked with ** were
non-mappable). The remaining 25 indicators were mapped. The mapping methodology for these
25 mappable indicators is presented in Appendix E, and maps for these indicators are presented
in Appendix F (by HUC-4 watershed) and Appendix H (by ecoregion).
#1 Acid Neutralizing Capacity (ANC) **
• Non-uniform Spatial Distribution of Data: The ANC sample point data are not distributed
homogeneously, which results in dissimilar amounts of ANC data within a HUC-4
boundary. In cases where there are few sample points within a HUC-4 boundary, individual
sites may have a large influence on the metric (percentage of "at risk" sites) that is calculated
for the reporting unit.
#22 At-Risk Freshwater Plant Communities
• State-level Data: The map for this indicator is based on data calculated on a state-by-state
basis. Transforming the data from a state-based representation to a HUC-4 representation
requires an assumption that the distribution of at-risk plants is uniform within each state. This
assumption allows for area-weighted percentages to be calculated for HUC-4 units that
intersect more than one state. Although this is an accepted transformation method because
the spatial extent of the HUC-4 units and the state boundaries are identical, the assumption of
uniform distribution may result in errors in the metric calculated for each HUC-4 unit.
#24 At-Risk Native Freshwater Species
• State-level Data. The map for this indicator is based on data calculated on a state-by-state
basis. Transforming the data from a state-based representation to a HUC-4 representation
requires an assumption that the percentage of at-risk species is uniform within each state.
This assumption allows for area-weighted percentages to be calculated for HUC-4 units that
intersect more than one state. Although this is an accepted transformation method because
the spatial extent of the HUC-4 units and the state boundaries are identical, the assumption of
uniform distribution may result in errors in the metric calculated for each HUC-4 unit.
#57 Coastal Vulnerability Index (CVI)
• Spatial Distribution of Data (special case: coastal indicator): The spatial extent of the data
for this indicator is for coastal and nearshore areas only. This is inherently disconnected from
the HUC-4 boundaries, which extend inland and do not consistently include coastal and
nearshore regions. This makes aggregation for the purposes of reporting at the HUC-4 scale
problematic. To address this issue, a special reporting unit was developed for this indicator.
Do Not Cite or Quote Page D-2
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Each unit extends approximately 20 miles inland and includes approximately 150 miles of
coastline.
• Local Variation: CVI values vary considerably at local scales. By averaging the CVI values
within the special coastal reporting units, localized vulnerability is masked.
#55 Commercially Important Fish Stocks**
• Spatial Distribution of Data (special case: marine indicator): This indicator does not provide
vulnerability information for any land or nearshore areas. This is a marine indicator that only
evaluates fish stocks located at least 3 miles from the coast and is inherently disconnected
from the HUC-4 boundaries which extend inland. No other data source for commercially
important fish stocks in inland waters or nearshore areas was identified.
• Low-Resolution Data: The data set for this indicator only provides three data values at a very
coarse spatial scale: NE Atlantic, SE Atlantic, and Pacific. These three data points do not
provide the variation necessary to assess vulnerability.
#95 Fish and Bottom-Dwelling Animals**
• Duplicate indicator. Upon further analysis, this indicator was determined to be a duplicate of
the indicator Macroinvertebrate Index of Biotic Condition (#460). Although these indicators
were cited by two distinct literature sources (Heinz Center, 2008 and USEPA, 2006b
respectively), their underlying data source i.e. EPA's Wadeable Streams Assessment is the
same and the same calculation is involved in obtaining the value of the indicator.
#725 Groundwater Reliance
There are no technical notes for this indicator.
#165 Meteorological Drought Indices
There are no technical notes for this indicator.
#190 Number of Dry Periods in Grassland/Shrubland Streams and Rivers*
• Data Gaps in National Coverage: This indicator is based on a 2009 study by the H. John
Heinz III Center for Science, Economics, and the Environment (Heinz Center) that evaluated
the number of "zero-flow" periods in streams that flow through watersheds dominated by
grassland and/or shrubland land cover types. The authors of the study delineated watersheds
for USGS stream gages, calculated the land use within those watersheds, and then analyzed
stream flow records for watersheds dominated by grassland and shrubland cover. They
determined the average annual percent of streams in Western ecoregions that experienced dry
periods during a rolling 5-year period. The Heinz Center did not conduct this analysis for
Eastern ecoregions. Although the reasons for this are unclear, it is likely that this is because
Do Not Cite or Quote Page D-3
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
few, if any, stream sites in Eastern ecoregions satisfied their definition of a "grassland" or
"shrubland" stream.
Non-uniform Spatial Distribution of Data: In the Western ecoregions, the number of
grassland sites within each HUC-4 unit varies widely. In cases where there are few grassland
sites within a HUC-4 boundary, individual sites may have a large influence on the average
number of dry periods that is calculated for that areas. The resulting spatial heterogeneity
may be a reflection of this sensitivity to data availability.
#209 Population Susceptible to Flood Risk**
• Data Gaps in National Coverage: This indicator evaluates the human population currently
residing within the 500-year floodplain. This would be calculated by overlaying estimates of
the 500-year floodplain from the Federal Emergency Management Agency (FEMA) with
population data from the U.S. Census Bureau. According to FEMA's Map Service Center,
GIS-compatible digital flood plain data are only available for certain parts of the country, as
shown in the map below. Without the digital flood plain data, significant effort is required to
create a national map for this indicator.
Availability of Digital Flood Insurance Rate Maps (DFIRMs). The Federal Emergency Management Agency
provides GIS-compatible flood plain maps for all areas in purple. Source: Federal Emergency Management Agency
#218 Ratio of Snow to Total Precipitation
There are no technical notes for this indicator.
#219 Ratio of Withdrawals to Stream Flow
There are no technical notes for this indicator.
#284 Stream Habitat Quality
• Non-uniform Spatial Distribution of Data: The sampling effort for EPA's Wadeable Streams
Assessment (WSA) varies across HUC-4 units. In cases where there are few sample points
Do Not Cite or Quote
Page D-4
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
within a HUC-4 boundary, individual sites may have a large influence on the average rapid
assessment score that is calculated for that area. The map for this indicator shows a
heterogeneous distribution of vulnerability, which may be a reflection of this sensitivity to
sampling effort.
• Local Variation: Habitat quality varies at local scales and is affected by local factors. By
calculating an average assessment score for each HUC-4 unit, localized vulnerability within
the HUC-4 unit is masked.
#322 Waterborne Human Disease Outbreaks**
• Data Collection and Reporting Discontinued: The most recent data from the Waterborne
Disease and Outbreak Surveillance System (WBDOSS) appear to be from 2006. According
to the Heinz Center (2008), data are no longer reported.
• Quality of Source Data: The WBDOSS relies on voluntary reporting from public health
departments within the Unites States. The data collection methods used for the WBDOSS
raise data quality concerns due to:
o Variable Resources for Reporting: Some states, territories, or local public health
departments may have more available resources for reporting WBDOs and may be more
able or inclined to report such cases than other states. Therefore, a map of nationwide
WBDOs reported may represent the distribution of actual outbreaks.
o Voluntary Data Reporting: The Heinz Center (2008) study, which defined this indicator,
states that inconsistencies in reporting or an altogether lack of reporting may occur as it is
not mandatory for states, territories, and local public health departments to report
WBDOs.
o Skewed Representation of Large vs. Small Public Water Supplies: There may be under-
reporting of WBDOs from public water supplies serving small communities and over-
reporting due to larger outbreaks from larger PWSs as the latter is more likely to receive
widespread attention from authorities and the media.
#325 Wetland Loss**
• Significant Data Processing Required: National Wetlands Inventory (NWI) data can only be
downloaded at the 7.5 minute (1:24K) or 15 minute (1:100K) scales. In the lower 48 states,
the USGS has designated approximately 56,500 l:24K-scale quadrangles. It would require
substantial effort to download data for each of these quadrangles and assemble them into a
national dataset. It may be possible to acquire a national wetlands dataset from the U. S. Fish
& Wildlife Service, although the FWS web site currently states that "Due to limited
resources, custom wetland data extractions by the Wetlands Team are no longer available."
Do Not Cite or Quote Page D-5
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
• Data Gaps in National Coverage: There are data gaps in the national coverage that would
diminish the usefulness of this national map. According to the FWS, the wetlands in 13 states
are either unmapped or are recorded on hardcopy maps only.
• Inadequate Historical Data: The indicator describes wetland loss, which suggests that a
second data source must be used to calculate a change in the area of wetlands over time. A
data source that delineates historical wetland extent at the national scale has not been
identified.
#326 Number of Wetland Species at Risk
• State-level Data: The map for this indicator is based on data calculated on a state-by-state
basis. Transforming the data from a state-based representation to a HUC-4 representation
requires an assumption that the distribution of at-risk species is uniform within each state.
This assumption allows for area-weighted totals to be calculated for HUC-4 units that
intersect more than one state. Although this is an accepted transformation method because
the spatial extent of the HUC-4 units and the state boundaries are identical, the assumption of
uniform distribution may result in errors in the metric calculated for each HUC-4 unit.
#348 Erosion Rate
• Limited Temporal Resolution of Source Data: The source data for this map are based on land
cover patterns in 1980. Changes in land cover since then may affect the spatial distribution of
soil erosion.
#351 In-stream Use / Total Streamflow
• Missing Data for Input Variables: Estimates of runoff (streamflow) and groundwater
overdraft (the amount of groundwater withdrawals that exceeds the long-term average annual
recharge rate) are used to calculate this indicator. For some regions, groundwater recharge
estimates were not available from the original source based on 1975 data. In these areas,
withdrawal (as measured in 1975) was assumed to equal recharge. These recharge estimates
were then compared to 1995 groundwater withdrawals to calculate updated estimates of
groundwater overdraft.
• Low-Re solution Data: Estimates of groundwater recharge were developed at the 2-digit HUC
scale and applied evenly to all sub-regions (4-digit HUCs) within the region.
#352 Total Use / Total Streamflow
• Missing Data for Input Variables: Estimates of runoff (streamflow) and groundwater
overdraft (the amount of groundwater withdrawals that exceeds the long-term average annual
recharge rate) are used to calculate this indicator. For some regions, groundwater recharge
estimates were not available from the original source based on 1975 data. In these areas,
withdrawal (as measured in 1975) was assumed to equal recharge. These recharge estimates
Do Not Cite or Quote Page D-6
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
were then compared to 1995 groundwater withdrawals to calculate updated estimates of
groundwater overdraft. In addition, this indicator assesses water consumption as a
component of total use. The consumptive use data from USGS does not distinguish between
water consumed from groundwater and surface water sources, and it is unclear in the original
data source whether this distinction is relevant. Additional investigation into the role of
consumptive use in this indicator and the original source of water consumed is
recommended.
• Low-Re solution Data: Estimates of groundwater recharge were developed at the 2-digit HUC
scale and applied evenly to all sub-regions (4-digit HUCs) within the region.
#364 Pesticide Toxicity Index (PTI)
• Non-uniform Spatial Distribution of Data: The sampling effort for the USGS National Water
Quality Assessment (NAWQA) Program varies across HUC-4 units. In cases where there
are few sample points within a HUC-4 boundary, individual sites may have a large influence
on the average pesticide concentration that is calculated for that area. The map for this
indicator shows a heterogeneous distribution of vulnerability, which may be a reflection of
this sensitivity to sampling effort. In addition, there are numerous HUC-4 units where no data
are available.
• Local Variation: Pesticide concentrations in streams vary at local scales and are affected by
local factors. By calculating an average concentration for each HUC-4 unit, localized
vulnerability within the HUC-4 unit is masked.
• Relative Rather than Absolute Toxicity: The Pesticide Toxicity Index (PTI) accounts for
multiple pesticides in a sample, including pesticides without established benchmarks for
aquatic life. The PTI combines information on exposure of aquatic biota to pesticides
(measured concentrations of pesticides in stream water) with toxicity estimates (results from
laboratory toxicity studies) to produce a relative index value for a sample or stream. The PTI
value is computed for each sample of stream water by summing the toxicity quotients for all
pesticides detected in the sample. The toxicity quotient is the measured concentration of a
pesticide divided by its toxicity concentration from bioassays (such as an LC50 or EC50).
This approach follows the Concentration Addition Model of toxicity. Although simple
addition is unlikely to strictly apply for complex mixtures of pesticides from different classes
and with different effects and modes of action, the PTI is still useful as a relative index.
While the PTI does not indicate whether water in a sample is toxic, its value can be used to
rank or compare the relative potential toxicity of different samples or different streams.
#367 Herbicide concentrations in streams
• Non-uniform Spatial Distribution of Data: The sampling effort for the USGS National Water
Quality Assessment (NAWQA) Program varies across HUC-4 units. In cases where there
are few sample points within a HUC-4 boundary, individual sites may have a large influence
on the average herbicide concentration that is calculated for that area. The map for this
indicator shows a heterogeneous distribution of vulnerability, which may be a reflection of
Do Not Cite or Quote Page D-7
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
this sensitivity to sampling effort. In addition, there are numerous HUC-4 units where no data
are available.
• Local Variation: Herbicide concentrations in streams vary at local scales and are affected by
local factors. By calculating an average concentration for each HUC-4 unit, localized
vulnerability within the HUC-4 unit is masked.
#369 Insecticide concentrations in streams
• Non-uniform Spatial Distribution of Data: The sampling effort for the USGS National Water
Quality Assessment (NAWQA) Program varies across HUC-4 units. In cases where there are
few sample points within a HUC-4 boundary, individual sites may have a large influence on
the average insecticide concentration that is calculated for that area. The map for this
indicator shows a heterogeneous distribution of vulnerability, which may be a reflection of
this sensitivity to sampling effort. In addition, there are numerous HUC-4 units where no data
are available.
• Local Variation: Insecticide concentrations vary at local scales and are affected by local
factors. By calculating an average concentration for each HUC-4 unit, localized vulnerability
within the HUC-4 unit is masked.
#371 Organochlorines in Bed Sediment
• Non-uniform Spatial Distribution of Data: The sampling effort for the USGS National Water
Quality Assessment (NAWQA) Program varies across HUC-4 units. In cases where there
are few sample points within a HUC-4 boundary, individual sites may have a large influence
on the average concentration that is calculated for that area. The map for this indicator
shows a heterogeneous distribution of vulnerability, which may be a reflection of this
sensitivity to sampling effort. In addition, there are numerous HUC-4 units where no data
are available.
• Local Variation: Organochlorine concentrations vary at local scales and are affected by local
factors. By calculating an average concentration for each HUC-4 unit, localized vulnerability
within the HUC-4 unit is masked.
#373 Herbicides in Groundwater
• Non-uniform Spatial Distribution of Data: The sampling effort for the USGS National Water
Quality Assessment (NAWQA) Program varies across HUC-4 units. In cases where there are
few sample points within a HUC-4 boundary, individual sites may have a large influence on
the average concentration that is calculated for that area. The map for this indicator shows a
heterogeneous distribution of vulnerability, which may be a reflection of this sensitivity to
sampling effort. In addition, there are numerous HUC-4 units where no data are available.
Do Not Cite or Quote Page D-8
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
• Local Variation: Herbicide concentrations vary at local scales and are affected by local
factors. By calculating an average concentration for each HUC-4 unit, localized vulnerability
within the HUC-4 unit is masked.
#374 Insecticides in Groundwater
• Non-uniform Spatial Distribution of Data: The sampling effort for EPA's Wadeable Streams
Assessment (WSA) varies across HUC-4 units. In cases where there are few sample points
within a HUC-4 boundary, individual sites may have a large influence on the average
concentration that is calculated for that area. The map for this indicator shows a
heterogeneous distribution of vulnerability, which may be a reflection of this sensitivity to
sampling effort. In addition, there are numerous HUC-4 units where no data are available.
• Local Variation: Insecticide concentrations vary at local scales and are affected by local
factors. By calculating an average concentration for each HUC-4 unit, localized vulnerability
within the HUC-4 unit is masked.
#437Precipitation Elasticity of Stream/low
• Map Derived from Figure in Source Literature: The original data used to calculate this
indicator were not available and suitable alternatives would require significant effort to
assemble. Therefore, the map for this indicator was derived from isopleths that are presented
in the source literature (Figure 4 in Sankarasubramanian et al. 2001). The original map was
based on data collected from 1951-1988. Streamflow and precipitation data available from
the U.S. Geological Survey and Oregon State University's PRISM Group, respectively, could
be used to reproduce an updated version of this map.
#449 Ratio of Storage to Runoff
• Quality of Source Data: This indicator uses the National Inventory of Dams to evaluate
reservoir storage. This database contains a wide variety of water control structures, some of
which may not be relevant for this indicator. An effort to remove irrelevant water control
structures (e.g. coastal flood gates, navigational locks) was made, although the accuracy of
the dam record attributes used to apply these filters is unclear.
#453 Runoff Variability
• Low-Re solution Data: This indicator uses simulations of the global land surface hydrological
cycle from the Variable Infiltration Capacity (VIC) Model to predict runoff. The native
spatial resolution of the model output is a 2° x 2° grid and the model output is stored in
NetCDF format. A more recent series of VIC model predictions at a finer spatial resolution
may be available from the same source, but the NetCDF file available online appears to be
corrupt.
Do Not Cite or Quote Page D-9
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
#460 Macroinvertebrate Index ofBiotic Condition
• Non-uniform Spatial Distribution of Data: The sampling effort for EPA's Wadeable Streams
Assessment (WSA) varies across HUC-4 units. In cases where there are few sample points
within a HUC-4 boundary, individual sites may have a large influence on the average
Macroinvertebrate Index ofBiotic Condition value that is calculated for that area. The map
for this indicator shows a heterogeneous distribution of vulnerability, which may be a
reflection of this sensitivity to sampling effort.
• Local Variation: Macroinvertebrate community condition varies at local scales and is
affected by local factors. By calculating an average value for the Macroinvertebrate Index of
Biotic Condition for each HUC-4 unit, localized vulnerability within the HUC-4 unit is
masked.
#461 Macroinvertebrate Observed/Expected (O/E) Ratio ofTaxa Loss
• Non-uniform Spatial Distribution of Data: The sampling effort for EPA's Wadeable Streams
Assessment (WSA) varies across HUC-4 units. In cases where there are few sample points
within a HUC-4 boundary, individual sites may have a large influence on the average
Macroinvertebrate O/E value that is calculated for that area. The map for this indicator
shows a heterogeneous distribution of vulnerability, which may be a reflection of this
sensitivity to sampling effort.
• Local Variation: Macroinvertebrate community condition varies at local scales and is
affected by local factors. By calculating an average value for the Macroinvertebrate O/E
Ratio, localized vulnerability is masked.
#462 CoastalBenthic Communities**
• Inconsistencies in Reporting Data: Data provided on the National Coastal Assessment
(NCA) web site were collected by multiple agencies. The methods used to calculate an index
to describe the status of benthic communities vary between agencies. These differences make
comparisons between states and regions problematic.
• Data Gaps in National Coverage: Data are not currently available for many of the nation's
coastal areas.
#623 Water Availability: Net Streamflow per Capita
There are no technical notes for this indicator.
Do Not Cite or Quote Page D-l 0
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Appendix E. Mapping Methodology
Do Not Cite or Quote Page E-l
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
This appendix provides details of the procedures used for mapping each of 26 indicators
(including the indicator marked with a * that had an incomplete map.) Maps were created using
ArcMap 9.2. Prior to mapping, data were prepared (including aggregation to the appropriate
scale, when necessary) using Microsoft Excel or Microsoft Access.
Data sources listed in this appendix are described in greater detail in Appendix C. Technical
issues related to creating or interpreting maps for these indicators are described in Appendix D.
Maps of the 25 indicators are presented in Appendix F (by HUC-4 watershed).
#1 Add Neutralizing Capacity (ANC)
Data Sources (see Appendix C for more details):
• Stream Water Chemistry (Filename: waterchemistry.csv): EPA Wadeable Streams
Assessment. Available online at: http://www.epa.gov/owow/streamsurvey/web data.html
• Stream Site Info: EPA Wadeable Streams Assessment (Filename:
wsa_siteinfo_ts_final.csv). Available online at:
http ://www. epa. gov/owow/streamsurvey/web_data.html
• Hydrologic Units - http://water.usgs.gov/GIS/metadata/usgswrd/XML/huc250k.xml
Processing Steps:
(1) Water chemistry data were downloaded and opened as a comma-delimited text file in
Microsoft Excel. The columns containing "SitelD" and "ANC" were saved as a new .dbf
file.
(2) The .dbf file with water chemistry data, the hydrologic units, and Site Info were opened
in ArcMap 9.2. The water chemistry table was joined to the Site Info table using the
SitelD field.
(3) An event theme was mapped for all sites with corresponding water chemistry data, using
the LON_DD and LAT_DD fields (North American Datum of 1983).This event theme
was exported as a shapefile.
(4) The "AtRisk" field was added to the exported shapefile. If the value in the "ANC" field
was < 100, then the AtRisk field value was calculated as 1. All other records were
assigned a value of 0.
(5) The sites were aggregated with the 4-digit HUCs using a spatial join. If a HUC contained
more than one site, the total (sum) for the numeric fields was calculated. As a result of
the spatial join, a new shapefile was created. The "Pct_AtRisk" field was added to the
new shapefile, and field values were calculated by using:
Percent at Risk = At Risk Sites within HUC
Total Sites within HUC
(6) Data were mapped using the Pct_AtRisk field to indicate low, medium, and high
vulnerability categories.
Do Not Cite or Quote Page E-2
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
#22 Percent ofAt-Risk Freshwater Plant Communities
Data Sources (see Appendix C for more details):
• Freshwater Plants Status: The H. John Heinz III Center for Science, Economics, and the
Environment (Heinz Center). 2009. Email message to Cadmus. April 17, 2009.
(Filename: G1-G5 wetlands by state.xls).
• State Boundaries: Environmental Systems Research Institute (ESRI) Data & Maps. -
Projected to Albers map projection)
• Hydrologic Units - http://water.usgs.gov/GIS/metadata/usgswrd/XML/huc250k.xml
Processing Steps:
(1) The Microsoft Excel spreadsheet containing the percentages of at-risk plants was opened
in ArcMap. These percentages were reported by state.
(2) State boundaries and hydrologic units were also opened in ArcMap.
(3) The percentages were joined to the state boundaries shapefile using a table join and the
State Name attribute. The resulting shapefile was intersected with the 4-digit hydrologic
units.
(4) Using the area of each hydrologic unit, the area of the intersected shapes, and the
percentage of at-risk plant species, an area-weighted percentage value was calculated for
each intersected area.
(5) The shapefile was dissolved by the 4-digit HUC code to re-aggregate the 4-digit HUCs.
The area-weighted percentages were summed.
(6) The final map was created using the summed area-weighted percentages in the
HUC4_AtRiskFWPlants.shp shapefile to indicate low, medium, and high vulnerability
categories.
#24 Percent ofAt-Risk Freshwater Species
Data Sources (see Appendix C for more details):
• Freshwater Species Status: The H. John Heinz III Center for Science, Economics, and
the Environment (Heinz Center). 2009. Email message to Cadmus. April 17, 2009.
(Filename: AtRiskFWanimalSPby state.xls)
• State Boundaries: Environmental Systems Research Institute (ESRI) Data & Maps. -
Projected to Albers map projection)
• Hydrologic Units - http://water.usgs.gov/GIS/metadata/usgswrd/XML/huc250k.xml
Processing Steps:
(1) The Microsoft Excel spreadsheet containing the percentages of at-risk species was
opened in ArcMap. These percentages were reported by state.
(2) State boundaries and hydrologic units were also opened in ArcMap.
(3) The percentages were joined to the state boundaries shapefile using a table join and the
State Name attribute. The resulting shapefile was intersected with the 4-digit hydrologic
units.
(4) Using the area of each hydrologic unit, the area of the intersected shapes, and the
percentage of at-risk species, an area-weighted percentage value was calculated for each
intersected area.
Do Not Cite or Quote Page E-3
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
(5) The shapefile was dissolved by the 4-digit HUC code to reaggregate the 4-digit HUCs.
The area-weighted percentages were summed.
(6) The final map was created using the summed area-weighted percentages in the
HUC4_AtRiskFWSpecies.shp shapefile to indicate low, medium, and high vulnerability
categories.
#57 Coastal Vulnerability Index
Data Sources (see Appendix C for more details):
• USGS - A Preliminary Database for the U.S. Atlantic, Pacific and Gulf of Mexico Coasts
(U.S. Geological Survey Digital Data Series - 68; Three data sets: Gulf Coast, East
Coast, and West Coast).) Available online at:
http://pubs.usgs.gov/dds/dds68/htmldocs/data.htm
Processing Steps:
(1) ArcGIS shapefiles, containing attributes for the raw CVI variables, CVI, and risk
categories associated with CVI values, were opened in ArcMap 9.2.
(2) The Coastal Vulnerability Index (CVI) uses 6 variables, which appear as attributes in the
shapefiles:
• Mean Wave Height - - mean elevation of all nonnegative 5' by 5' grid cells within a
given 0.25° grid cell; values in meters. WIS hindcast nearshore mean wave height
1976-1995.
• Mean Tide Range - average of the mean tide range for all the gauge stations that
occur within a given 0.25° grid cell (mean tide range is the difference in height
between mean high water and mean low water in 1988); values in meters.
• Regional Coastal Slope (%) - Acquired from ETOPO5 and NGDC elevation data.
• Erosion and Accretion rates (m/yr) - the local subsidence trend.
• Relative Sea-Level Rise (mm/yr) - Acquired from NOS tide stations.
• Geomorphology Risk - ordinal value indicative of the type and susceptibility of the
landforms within a given 0.25° grid cell to inundation and erosion.
(3) The values for each variables are grouped into risk categories, and these risk categories
are used to calculate the CVI value using the following formula:
CVI = (a*b*c*d*e*f*g)/6] A V*
The calculated CVI values are then grouped into risk categories (4 = very high risk, 3 =
high risk, 2 = medium risk, 1 = low risk).
(4) Hydrologic units are not a suitable reporting unit for this indicator, so a new coastal unit
was produced by creating a 20 mile inland buffer of the shoreline. The buffer was
divided into 150-mile long segments to show variation along the coast.
(5) The vertices within the linear geometry of each shapefile feature were converted to
points. Each point inherited the attributes, including the CVI risk categories (4 = very
high risk, 3 = high risk, 2 = medium risk, 1 = low risk), of the original linear feature. All
points within the new coastal units were averaged using a spatial join.
(6) The final map of coastal units was created using the average CVI risk values to indicate
low, medium, and high vulnerability categories.
Do Not Cite or Quote Page E-4
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
#125 Groundwater Reliance
Data Sources (see Appendix C for more details):
• United States Geological Survey (USGS). Water Usage, 1995
http://water.usgs.gov/watuse/spread95/ush895.txt
• Hydrologic Units - http://water.usgs.gov/GIS/metadata/usgswrd/XML/huc250k.xml
Processing Steps:
(1) Data from USGS were downloaded and imported into ArcMap 9.2.
(2) The water usage data table was joined to the attribute table for the hydrologic units
shapefile, using the "HUC_CODE" and "HUCSCode" fields.
(3) The 8-digit HUC regions were aggregated into 4-digit HUCs with the 'dissolve' function
in ArcMap. The "SUB" attribute was used as the basis for the dissolve process. During
the dissolve operation, summary statistics were calculated for Total Groundwater
Withdrawals (TO_WGWTo) and Total Withdrawals (TO_WTotl) attributes.
(4) After the 4-digit HUC shapefile was produced with the dissolve operation, a new field for
Groundwater Reliance (GWRel_95) was added to store groundwater reliance values.
(5) Next, the indicator values were calculated using:
Groundwater Reliance (GWRel_95) = Total Groundwater Withdrawals (TO WGWTo)
Total Withdrawals (TO_WTotl)
(6) Finally, the Groundwater Reliance data were displayed on the map with symbology to
indicate low, medium, and high vulnerability categories.
#165 Meteorological Drought Indices
Data Sources (see Appendix C for more details):
• Palmer Drought Severity Index: National Climatic Data Center. Available online at:
http://www7.ncdc. noaa.gov/CDO/CDODivisi onalSelect.jsp#
• NCDC Climate Division Boundaries. Available online at:
ftp: //ftp. ncdc. noaa.gov/pub/data/divb oundari es/gi s/
• Hydrologic Units - http://water.usgs.gov/GIS/metadata/usgswrd/XML/huc250k.xml
Processing Steps:
(1) The drought index data were downloaded for the climate divisions within each of the
continental U.S. states as comma-delimited text for the years 2003-2007. The data for
each state were imported into a Microsoft Access database and compiled into a single
table.
(2) A select query was used to compute average Palmer Drought Severity Index values for
each climate division over the 2003-2007 time period using values in the PDSI column.
The query results were exported as a .dbf.
(3) The NCDC Climate Division Boundaries shapefile was opened in ArcMap 9.2, along
with the hydrologic units and the .dbf containing drought severity values.
(4) The DIVISION_ID attribute was used to join the climate boundaries to the drought index
data.
(5) The joined shapefile was intersected with the hydrologic units.
Do Not Cite or Quote Page E-5
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
(6) Using the area of the original climate boundaries, the area of the intersected shapes, and
the PDSI values, an area-weighted Palmer Drought Severity Index value was calculated
for each intersected area.
(7) The shapefile was dissolved by the 4-digit HUC code to reaggregate the 4-digit HUCs.
The area-weighted percentages were summed during the 'dissolve' operation.
(8) The final map was created using summed area-weighted percentages in the dissolved
shapefile to indicate low, medium, and high vulnerability categories.
#190 Number of Dry Periods in Grassland/Shrubland Streams and Rivers*
Data Sources (see Appendix C for more details):
• Grassland Stream Sites and Flow Data, 2002-2006: The H. John Heinz III Center for
Science, Economics, and the Environment (Heinz Center). 2009. Email message to
Cadmus. April 28, 2009. (Filename: GSdry periods_Cadmus.xls).
• Hydrologic Units - http://water.usgs.gov/GIS/metadata/usgswrd/XML/huc250k.xml
Processing Steps:
(1) The Microsoft Excel spreadsheet containing site and zero-flow data from the Heinz
Center was opened in ArcMap, along with the hydrologic units, and Site data.
(2) Grassland sites within the zero-flow table were joined to the site information table using
the Site Number field.
(3) An event theme was created for the joined records, using the LON_DD and LAT_DD
fields (North American Datum 1983). The event theme was exported as a shapefile.
(4) The sites were aggregated within the 4-digit HUCs using a spatial join. If a HUC
contained more than one site, the numeric attributes within the site data were summarized
with the "SUM" option. As a result of the spatial join, a new shapefile was created
(Ind 190_HUC4_Sites. shp).
(5) To determine the average annual percentage of streams with zero-flow period, the
proportion of streams with zero flow within each 4-digit HUC was computed for each
year, and then the mean of five years (2002-2006) was computed for each HUC.
(6) The final map was created using the Mean_Pct attribute to indicate low, medium, and
high vulnerability categories. In 4-digit HUCs where no site data were available, the
HUC was assigned to a 'No Data' category.
#218 Ratio of Snow to Total Precipitation
Data Sources (see Appendix C for more details):
• Monthly Climate Data and Observation Station Locations from National Climatic Data
Center. Available online at: http://gis.ncdc.noaa.gov/snowfallmo/ (The web site provides
access to multiple parameters, including snowfall totals. Data can be downloaded for
free from .edu and .gov domains.)
• Hydrologic Units - http://water.usgs.gov/GIS/metadata/usgswrd/XML/huc250k.xml
Do Not Cite or Quote Page E-6
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Processing Steps:
(1) Climate data for all U.S. observation stations were downloaded and imported into
Microsoft Access. (Data were downloaded in two batches due to NCDC file size limits).
(2) The ratio of average annual snowfall (Element Code = 'TSNW') to average annual
precipitation (Element Code = 'TPCP') was calculated using a series of queries that
summed the monthly snow and precipitation totals for each year, calculated an annual
ratio, then averaged the annual ratios across the 1998-2007 time period. The output of the
queries was saved as a .dbf and imported into ArcMap 9.2.
(3) Observation station site data were also downloaded and opened as a fixed-width file in
Excel. Within this file, latitude and longitude coordinates for the observation stations
were in the format of Degrees: Minutes. These values were converted in Microsoft Excel
to Decimal Degrees using the formula:
Decimal Degrees = Degrees + (Minutes/60)
(4) An event theme was created in ArcMap for the Observation Stations using the adjusted
Lat/Long coordinates. This theme was joined to the .dbf containing heat sensitivity data
using the "COOPID" attribute. Not all Observation Stations had corresponding heat
sensitivity data; sites with heat sensitivity data were exported to shapefile and mapped.
(5) The stations (points) were aggregated within each 4-digit HUC (polygon) using a spatial
join. If a HUC contained more than one site, the average value across sites was
calculated.
(6) The final map was produced using the Avg_RATIO field to indicate low, medium, and
high vulnerability categories.
#219 Ratio of Withdrawals to Stream Flow
Data Sources (see Appendix C for more details):
• Mean Annual Precipitation 1971-2000: PRISM Climate Group (Oregon State
University). Available online at:
http://www.prism.oregonstate.edu/products/matrix.phtml
• Mean Daily Maximum Temperature 1971-2000: PRISM Climate Group (Oregon State
University). Available online at:
http://www.prism.oregonstate.edu/products/matrix.phtml
• United States Geological Survey (USGS). Water Usage, 1995
http://water.usgs.gov/watuse/spread95/ush895.txt
• Hydrologic Units - http://water.usgs.gov/GIS/metadata/usgswrd/XML/huc250k.xml
Processing Steps:
(1) PRISM data were downloaded and converted to ESRI© GRID format using the ASCII
Text to Grid tool in ArcMap 9.2.
(2) Mean daily temperature was calculated with the raster calculator using the following
formula as described by Vogel, 1999: (Max Temp + Min Temp) / 2.
(3) Mean precipitation and mean daily temperature within each 4-digit HUC were calculated
using the Zonal Statistics function.
Do Not Cite or Quote Page E-7
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
(4) A new attribute for mean annual streamflow was calculated with the field calculator in
ArcMap for each HUC-4 unit using the regression equations in Vogel, 1999 (Table 2).
The HUC-2 code was used to associate each HUC-4 with the appropriate regional
regression equation.
(5) The water usage data table was joined to the attribute table for the hydrologic units
shapefile, using the "HUC_CODE" and "HUCSCode" fields.
(6) The 8-digit HUC regions were aggregated into 4-digit HUCs with the 'dissolve' function
in ArcMap. The "SUB" attribute was used as the basis for the dissolve process. During
the dissolve operation, summary statistics were calculated for Total Withdrawals.
(7) Streamflow and withdrawals were adjusted for some HUC-4 units to account for
withdrawals and streamflow that occur upstream.
(8) A new attribute for the ratio of withdrawals to streamflow was calculated with the field
calculator. The units for streamflow and withdrawals were converted as needed.
(9) The final map was produced indicating low, medium, and high vulnerability categories.
#284 Stream Habitat Quality
Data Sources (see Appendix C for more details):
• Stream Rapid Assessment Metrics: EPA Wadeable Streams Assessment. (Filename:
rapidhabmetrics.csv). Available online at:
http ://www. epa. gov/owow/streamsurvey/web_data.html
• Stream Site Info: EPA Wadeable Streams Assessment. (Filename:
wsa_siteinfo_ts_final.csv). Available online at:
http ://www. epa. gov/o wow/stream survey/web data.html
• Hydrologic Units - http://water.usgs.gov/GIS/metadata/usgswrd/XML/huc250k.xml
Processing Steps:
(1) The rapid assessment data were downloaded and opened as a comma-delimited text file
in ArcMap, along with the stream site data, and the hydrologic units.
(2) This Rapid Assessment and Site Info tables were joined using the SitelD field.
(3) An event theme was mapped for the joined records, using the LON_DD and LAT_DD
fields (North American Datum 1983). Only sites with corresponding Rapid Assessment
scores were mapped. This event theme was exported to shapefile.
(4) The sites were aggregated within the 4-digit HUCs using a spatial join. If a HUC
contained more than one site, the average value for the Rapid Assessment score
(RH_SUM) was calculated. As a result of the spatial join, a new shapefile was created
(HUC4_RapidAssessment.shp).
(5) The final map was created using the Avg_RH_SUM field to indicate low, medium, and
high vulnerability categories. In 4-digit HUCs where no sampling occurred, the HUC
was assigned a 'No Data' category.
Do Not Cite or Quote Page E-8
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
#348 Erosion Rate
Data Sources (see Appendix C for more details):
• Revised Universal Soil Loss Equation (RUSLE): RUSLE_1980.asc grid file of soil
erosion rates estimated for entire globe at 0.5 deg resolution with the RUSLE. Data were
obtained from: Dawen YANG, PhD, Professor, Department of Hydraulic Engineering
Tsinghua University, Beijing 100084, China; Tel: +86-10-62796976; Fax: +86-10-
62796971; E-mail: yangdw@tsinghua.edu.cn
• State Boundaries: Environmental Systems Research Institute (ESRI) Data & Maps. -
Projected to Albers map projection)
• Hydrologic Units - http://water.usgs.gov/GIS/metadata/usgswrd/XML/huc250k.xml
Processing Steps:
(1) The ASCII file containing RUSLE data was converted to raster grid using ASCII to
Raster tool in ArcMap.
(2) Using the raster calculator, the grid values were multiplied by 1,000,000 to facilitate
conversion to polygons. The raster type was changed to integer.
(3) The raster was converted to a polygon layer using the Raster to Polygon tool in ArcMap.
(4) A new field called RUSLE = grid/1,000,000 was created in the polygon layer.
(5) The Intersect tool was used to combine the HUC4 layer with the RUSLE polygon layer to
create a new layer called HUC4_RUSLE_Intersect.
(6) A new field called AREAXRUSLE = AREA • RUSLE was created.
(7) Summarized HUC4_RUSLE_Intersect layer on the HUC ID field (SUB), and calculating
the sum of AREAXRUSLE
(8) The summarized data table was joined to the HUC4 polygon layer and exported as a new
layer called HUC4_RUSLE.
(9) In HUC4_RUSLE, a new field RUSLE = AREAXRUSLE / AREA was created.
(10) The final map was created using the "RUSLE" attribute in the HUC4_RUSLE.shp
shapefile to indicate low, medium, and high vulnerability categories.
#351 In-Strearn Use / Total Streamflow
Data Sources (see Appendix C for more details):
• United States Geological Survey (USGS). Water Usage, 1995
http://water.usgs.gov/watuse/spread95/ush895.txt
• Hydrologic Units - http://water.usgs.gov/GIS/metadata/usgswrd/XML/huc250k.xml
• Mean Annual Runoff: U.S. Geological Survey. Available online at:
http://water.usgs.gov/GIS/metadata/usgswrd/XML/runoff.xml
• Groundwater Recharge - WRC (U.S. Water Resources Council), 1978. The Nation's
Water Resources: 1975-2000 (Vol. 2). U.S. Government Printing Office, Washington
D.C.
Processing Steps:
(1) The water usage data table was joined to the attribute table for the hydrologic units
shapefile, using the "HUC_CODE" and "HUCSCode" fields.
(2) The 8-digit HUC regions were aggregated into 4-digit HUCs with the 'dissolve' function
in ArcMap. The "SUB" attribute was used as the basis for the dissolve process. During
Do Not Cite or Quote Page E-9
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
the dissolve operation, summary statistics were calculated for the Total Groundwater
Withdrawals attribute.
(3) After the 4-digit HUC shapefile was produced with the dissolve operation, surface water
withdrawal values were converted from megagallons per day to gallons per year.
(4) The isopleths in the mean annual runoff dataset were opened in ArcMap. The Spatial
Analyst extension was used to interpolate continuous runoff values across the country.
(5) Mean runoff values within each 4-digit HUC were calculated using the Zonal Statistics
function.
(6) An attribute for groundwater recharge rates was added and groundwater overdraft values
were calculated based on the definition in the WRC (1978) report: (Groundwater
Recharge - Groundwater Withdrawals)
(7) An attribute for in-stream use was added and calculated based on the definition in the
WRC (1978) report: (Streamflow * 0.6)
(8) Indicator values were calculated using the formula described in WRC (1978).
Streamflow is assumed to be equal to runoff:
In-stream use / (Streamflow - Groundwater overdraft)
(9) Finally, the indicator values were displayed on the map with symbology to indicate low,
medium, and high vulnerability categories.
#352 Total Use / Total Streamflow
Data Sources (see Appendix C for more details):
• United States Geological Survey (USGS). Water Usage, 1995
http://water.usgs.gov/watuse/spread95/ush895.txt
• Hydrologic Units - http://water.usgs.gov/GIS/metadata/usgswrd/XML/huc250k.xml
• Mean Annual Runoff: U.S. Geological Survey. Available online at:
http://water.usgs.gov/GIS/metadata/usgswrd/XML/runoff.xml
• Groundwater Recharge - WRC (U.S. Water Resources Council), 1978. The Nation's
Water Resources: 1975-2000 (Vol. 2). U.S. Government Printing Office, Washington
D.C.
Processing Steps:
(1) The water usage data table was joined to the attribute table for the hydrologic units
shapefile, using the "HUC_CODE" and "HUCSCode" fields.
(2) The 8-digit HUC regions were aggregated into 4-digit HUCs with the 'dissolve' function
in ArcMap. The "SUB" attribute was used as the basis for the dissolve process. During
the dissolve operation, summary statistics were calculated for the Total Consumptive Use
and Total Groundwater Withdrawals attributes.
(3) After the 4-digit HUC shapefile was produced with the dissolve operation, surface water
withdrawal values were converted from megagallons per day to gallons per year.
(4) The isopleths in the mean annual runoff dataset were opened in ArcMap. The Spatial
Analyst extension was used to interpolate continuous runoff values across the country.
(5) Mean runoff values within each 4-digit HUC were calculated using the Zonal Statistics
function.
Do Not Cite or Quote Page E-10
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
(6) An attribute for groundwater recharge rates was added and groundwater overdraft values
were calculated based on the definition in the WRC (1978) report: (Groundwater
Recharge - Groundwater Withdrawals)
(7) An attribute for in-stream use was added and calculated based on the definition in the
WRC (1978) report: (Streamflow * 0.6)
(8) Indicator values were calculated using the formula described in WRC (1978).
Streamflow is assumed to be equal to runoff:
(In-stream use + Total Consumptive Use) / (Streamflow - Groundwater overdraft)
(9) Finally, the indicator values were displayed on the map with symbology to indicate low,
medium, and high vulnerability categories.
#364 Pesticide Toxicity Index (PTI)
Data Sources (see Appendix C for more details):
• Pesticide (herbicides and insecticide) Concentrations: USGS NAWQA Program. "The
Quality of Our Nation's Waters Pesticides in the Nation's Streams and Ground Water,
1992-2001" (USGS Circular 1291). Available online at:
http://water.usgs.gov/nawqa/pnsp/pubs/circl291/appendix6/appendix6a.txt
• USEPA. 2009. ECOTOX Database. Available online at: http://cfpub.epa.gov/ecotox/.
Accessed September 11, 2009.
• NAWQA Sites: USGS. NAWQA Data Warehouse: SiteFile Master.
shttp ://infotrek. er.usgs. gov/nawqa_queries/sitemaster/index.j sp
• Hydrologic Units: http://water.usgs.gov/GIS/metadata/usgswrd/XML/huc250k.xml
Processing Steps:
(1) EC50 values for all Daphnia species for pesticides measured by NAWQA were
downloaded from EPA's ECOTOX database. For each pesticide, the mean Daphnia
EC50 value was calculated.
(2) Pesticide concentration data were downloaded and imported into Microsoft Excel.
(3) In Excel, the concentration of each pesticide at each sample event was divided by its
mean Daphnia EC50 value to calculate its toxicity quotient.
(4) The toxicity quotients for all pesticides were summed for each sampling event to
calculate that event's PTI. The toxicity quotients for constituents that were not measured
or were below detection levels were assumed to be zero.
(5) The PTIs for all sampling events at each site were summed to calculate a site PTI. This
table was imported into ArcMap.
(6) The PTI table was joined to the NAWQA Sites using the "STAID" attribute. In some
cases, the STAID value required minor edits to correctly join the tables. A total of 187
sites with pesticide concentration data were successfully joined to the NAWQA spatial
data.
(7) NAWQA spatial data were displayed as an event theme, using the Latitude and
Longitude variables (North American Datum of 1983).
(8) Next, the hydrologic units (4-digit HUCs) were added to the ArcMap project and joined
using a spatial join to the NAWQA points. If more than one site with pesticide data was
located within a 4-digit HUC, the PTI values were averaged across sites. In some cases,
there were no sites with pesticide data within a 4-digit HUC.
Do Not Cite or Quote Page E-ll
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
(9) The final map was created using symbology to indicate low, medium, and high
vulnerability categories for each 4-digit HUC. In 4-digit HUCs where no sampling
occurred, the HUC was assigned a 'No Data' category.
Note: Pesticide concentrations in agricultural areas, urban areas, and mixed land use areas
were combined for this indicator, although the USGS reports these land use types separately.
#367 Herbicide Concentrations in Streams
Data Sources (see Appendix C for more details):
• Herbicide Concentrations: USGS NAWQA Program. "The Quality of Our Nation's
Waters Pesticides in the Nation's Streams and Ground Water, 1992-2001" (USGS
Circular 1291). Available online at:
http://water.usgs.gov/nawqa/pnsp/pubs/circl291/appendix6/appendix6a.txt
• NAWQA Sites: USGS. NAWQA Data Warehouse: SiteFile Master.
shttp://infotrek.er.usgs.gov/nawqa_queries/sitemaster/index.jsp
• Hydrologic Units: http://water.usgs.gov/GIS/metadata/usgswrd/XML/huc250k.xml
Processing Steps:
(1) Herbicide data were downloaded and imported into Microsoft Excel.
(2) In Excel, measured values for herbicides, herbicide degradates, and fungicides were
identified and summed for each sampling event. Constituents that were not measured or
were below detection levels were assumed to be zero.
(3) The herbicide concentration table was imported into ArcMap. For each sampling site, the
average of these total concentrations was calculated for all sampling events that occurred
at that site using the "Summarize" function.
(4) The summarized herbicide table and was joined to the NAWQA Sites using the "STAID"
attribute. In some cases, the STAID value required minor edits to correctly join the
tables. A total of 182 sites with herbicide concentration data were successfully joined to
the NAWQA spatial data.
(5) NAWQA spatial data were displayed as an event theme, using the Latitude and
Longitude variables (North American Datum of 1983).
(6) Next, the hydrologic units (4-digit HUCs) were added to the ArcMap project and joined
using a spatial join to the NAWQA points. If more than one site with herbicide data was
located within a 4-digitt HUC, the concentration values were averaged across sites. In
some cases, there were no sites with herbicide data within a 4-digit HUC.
(7) The final map was created using symbology to indicate low, medium, and high
vulnerability categories for each 4-digit HUC. In 4-digit HUCs where no sampling
occurred, the HUC was assigned a 'No Data' category.
Note: Herbicide concentrations in agricultural areas, urban areas, and mixed land use areas
were combined for this indicator, although the USGS reports these land use types separately.
Do Not Cite or Quote Page E-12
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
#369 Insecticide Concentrations in Streams
Data Sources (see Appendix C for more details):
• Insecticide Concentrations: USGS NAWQA Program. "The Quality of Our Nation's
Waters Pesticides in the Nation's Streams and Ground Water, 1992-2001" (USGS
Circular 1291). Available online at:
http://water.usgs.gov/nawqa/pnsp/pubs/circl291/appendix6/appendix6a.txt
• NAWQA Sites: USGS. NAWQA Data Warehouse: SiteFile Master. Available online at:
shttp://infotrek.er.usgs.gov/nawqa queries/sitemaster/index.j sp
• Hydrologic Units: http://water.usgs.gov/GIS/metadata/usgswrd/XML/huc250k.xml
Processing Steps:
(1) Data were downloaded (see data sources) and imported into Microsoft Excel.
(2) In Excel, measured values for insecticides, insecticide degradates, and acaricides were
identified and summed for each sampling event. Constituents that were not measured or
were below detection levels were assumed to be zero.
(3) The insecticide concentration table was imported into ArcMap. For each sampling site,
the average of these total concentrations was calculated for all sampling events that
occurred at that site using the "Summarize" function.
(4) The summarized insecticide table and was joined to the NAWQA Sites using the
"STAID" attribute. In some cases, the STAID value required minor edits to correctly
join the tables. A total of 182 sites with insecticide concentration data were successfully
joined to the NAWQA spatial data.
(5) NAWQA spatial data were displayed as an event theme, using the Latitude and
Longitude variables (North American Datum of 1983).
(6) Next, the hydrologic units (4-digit HUCs) were added to the ArcMap project and joined
using a spatial join to the NAWQA points. If more than one site with insecticide data
was located within a 4-digitt HUC, the concentration values were averaged across sites.
In some cases, there were no sites with insecticide data within a 4-digit HUC.
(7) The final map was created using symbology to indicate low, medium, and high
vulnerability categories for each 4-digit HUC. In 4-digit HUCs where no sampling
occurred, the HUC was assigned a 'No Data' category.
Note: Insecticide concentrations in agricultural areas, urban areas, and mixed land use areas
were combined for this indicator, although the USGS reports these land use types separately.
#371 Organochlorines in Bed Sediment
Data Sources (see Appendix C for more details):
• Organochlorine Concentrations: USGS NAWQA Program. "The Quality of Our Nation's
Waters Pesticides in the Nation's Streams and Ground Water, 1992-2001" (USGS
Circular 1291). Available online at:
http://water.usgs.gov/nawqa/pnsp/pubs/circl291/appendix6/appendix6c.txt
• NAWQA Sites: USGS. NAWQA Data Warehouse: SiteFile Master. Available online at:
shttp ://infotrek. er.usgs. gov/nawqa queries/sitemaster/index.j sp
• Hydrologic Units: http://water.usgs.gov/GIS/metadata/usgswrd/XML/huc250k.xml
Do Not Cite or Quote Page E-13
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Processing Steps:
(1) Organochlorine concentration data were downloaded and imported into Microsoft Excel.
(2) In Excel, measured values for all parameters except P49271 (organic carbon in sediment)
were summed for each site. Only one sampling event occurred at each site, so
aggregation at the site level was not conducted. Constituents that were not measured or
were below detection levels were assumed to be zero.
(3) The organochlorine occurrence data were imported into ArcMap, and joined to the
NAWQA Sites using the "STAID" attribute. A total of 1,015 sites with organochlorine
concentration data were successfully joined to the NAWQA spatial data.
(4) NAWQA spatial data were displayed as an event theme, using the Latitude and
Longitude variables (North American Datum of 1983).
(5) The hydrologic units (4-digit HUCs) were opened in ArcMap and joined to the NAWQA
points using a spatial join. If more than one site with organochlorine data was located
within a 4-digit HUC, the total concentration values were averaged across sites. In some
cases, there were no sites with organochlorine data within a 4-digit HUC.
(6) The final map was created using symbology to indicate low, medium, and high
vulnerability categories for each 4-digit HUC. In 4-digit HUCs where no sampling
occurred, the HUC was assigned a 'No Data' category.
Note: Organochlorine concentrations in agricultural areas, urban areas, and mixed land use
areas were combined for this indicator, although the USGS reports these land use types
separately.
#373 Herbicide Concentrations in Groundwater
Data Sources (see Appendix C for more details):
• Herbicide Concentrations: USGS NAWQA Program. "The Quality of Our Nation's
Waters Pesticides in the Nation's Streams and Ground Water, 1992-2001" (USGS
Circular 1291). Available online at:
http://water.usgs.gov/nawqa/pnsp/pubs/circl291/appendix6/appendix6b.txt
• Hydrologic Units: http://water.usgs.gov/GIS/metadata/usgswrd/XML/huc250k.xml
Processing Steps:
(1) Herbicide concentration data were downloaded and imported into Microsoft Excel.
(2) In Excel, measured values for herbicides, herbicide degradates, and acaricides were
identified and summed for each sampling event. Constituents that were not measured or
were below detection levels were assumed to be zero.
(3) These herbicide occurrence data were imported into ArcMap and displayed as an event
theme, using the Latitude and Longitude variables (North American Datum of 1983).
(4) Next, the hydrologic units (4-digit HUCs) were added to the ArcMap project and joined
using a spatial join to the sampling event points. If more than one sampling event with
herbicide data occurred within a 4-digit HUC, the concentration values were averaged
across events. In some cases, no herbicide collection events occurred within a 4-digit
HUC.
Do Not Cite or Quote Page E-14
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
(5) The final map was created using symbology to indicate low, medium, and high
vulnerability categories for each 4-digit HUC. In 4-digit HUCs where no sampling
occurred, the HUC was assigned a 'No Data' category.
Note: Herbicide concentrations in agricultural and urban areas were combined for this
indicator, although the USGS reports these land use types separately.
#374 Insecticide Concentrations in Groundwater
Data Sources (see Appendix C for more details):
• Insecticide Concentrations: USGS NAWQA Program. "The Quality of Our Nation's
Waters Pesticides in the Nation's Streams and Ground Water, 1992-2001" (USGS
Circular 1291). Available online at:
http://water.usgs.gov/nawqa/pnsp/pubs/circl291/appendix6/appendix6b.txt
• Hydrologic Units: http://water.usgs.gov/GIS/metadata/usgswrd/XML/huc250k.xml
Processing Steps:
(1) Insecticide concentration data were downloaded and imported into Microsoft Excel.
(2) In Excel, measured values for insecticides, insecticide degradates, and acaricides were
identified and summed for each sampling event. Constituents that were not measured or
were below detection levels were assumed to be zero.
(3) These insecticide concentration data were imported into ArcMap and displayed as an
event theme, using the Latitude and Longitude variables (North American Datum of
1983).
(4) Next, the hydrologic units (4-digit HUCs) were added to the ArcMap project and joined
using a spatial join to the sampling event points. If more than one sampling event with
insecticide data occurred within a 4-digit HUC, the concentration values were averaged
across events. In some cases, no insecticide collection events occurred within a 4-digit
HUC.
(5) The final map was created using symbology to indicate low, medium, and high
vulnerability categories for each 4-digit HUC. In 4-digit HUCs where no sampling
occurred, the HUC was assigned a 'No Data' category.
Note: Insecticide concentrations in agricultural and urban areas were combined for this
indicator, although the USGS reports these land use types separately.
#437 Precipitation Elasticity of Stream/low
Data Sources (see Appendix C for more details):
• Precipitation Elasticity of Streamflow: Adapted from Figure 4 in Sankarasubramanian et
al. (2001). Water Resources Research 37(6).
• Hydrologic Units - http://water.usgs.gov/GIS/metadata/usgswrd/XML/huc250k.xml
Processing Steps:
(1) An image of Figure 4 was imported into ArcMap and georeferenced.
Do Not Cite or Quote Page E-15
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
(2) Using the Spatial Analyst extension, the isopleths in Figure 4 were digitized and used to
interpolate continuous elasticity values across the country.
(3) Mean elasticity values within each 4-digit HUC were calculated using the Zonal Statistics
function.
(4) The final map was created using symbology to indicate low, medium, and high
vulnerability categories for each 4-digit HUC.
Note: For the purposes of mapping this indicator with relative ease, Figure 4 in the original
literature source, Sankarasubramanian et al., 2001 (see Appendix A for full reference) was used.
However, the complete original data set could also be recalculated for mapping purposes using
the data sources listed in Appendix C.
#449 Ratio of Storage to Runoff
Data Sources (see Appendix C for more details):
• Reservoir Storage: National Inventory of Dams (from the National Atlas). Available
online at: http://www.nationalatlas.gov/mld/damsOOx.html
• Mean Annual Runoff: U.S. Geological Survey. Available online at:
http://water.usgs.gov/GIS/metadata/usgswrd/XML/runoff.xml
• Hydrologic Units - http://water.usgs.gov/GIS/metadata/usgswrd/XML/huc250k.xml
Processing Steps:
(1) The isopleths in the mean annual runoff dataset were opened in ArcMap. The Spatial
Analyst extension was used to interpolate continuous runoff values across the country.
(2) Mean runoff values within each 4-digit HUC were calculated using the Zonal Statistics
function.
(3) The shapefile containing data for dams was opened in ArcMap. The total maximum
reservoir storage capacity was calculated by joining the 4-digit HUCs to the dam data
using a spatial join. If more than one reservoir occurred within a 4-digit HUC, the
storage capacity was summed.
(4) The ratio of storage capacity to mean annual runoff was calculated within each 4-digit
HUC and saved within a new attribute.
(5) The final map was created using symbology to indicate low, medium, and high
vulnerability categories for each 4-digit HUC.
#453 Runoff Variability
Data Sources (see Appendix C for more details):
• Global Runoff, 1980-1993. Available online at:
http ://www.hydro. Washington. edu/SurfaceWaterGroup/Data/vic_global .html
Specific data file:
ftp://ftp.hy dro.washington.edu/pub/CE/HYDRO/nijssen/vic_global/calibrated/runoff.cali
brated.monthly.l980_1993.nc.gz
• Hydrologic Units - http://water.usgs.gov/GIS/metadata/usgswrd/XML/huc250k.xml
Do Not Cite or Quote Page E-16
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Processing Steps:
(1) The modeled runoff data were downloaded and decompressed.
(2) The NetCDF file, containing monthly runoff data, was imported into ArcMap. Monthly
grids were exported from the NetCDF format for all months during the 1984-1993 time
period (120 months).
(3) Annual runoff was calculated by aggregating the monthly values for each year.
(4) The mean and standard deviation of the annual runoff within each 2° x 2° grid cell was
calculated using the 10 annual values.
(5) The coefficient of variance was calculated by dividing mean annual runoff into the
standard deviation for each grid cell.
(6) The mean coefficient of variation within the 4-digit HUCs was calculated using the Zonal
Statistics function.
(7) The final map was created using symbology to indicate low, medium, and high
vulnerability categories for each 4-digit HUC.
#460 Macroinvertebrate Index ofBiotic Condition
Data Sources (see Appendix C for more details):
• Stream Water Benthic Macroinvertebrate Metrics: EPA Wadeable Streams Assessment.
(Filename: wsa_benmet300_ts_final.csv). Available online at:
http://www.epa.gov/owow/streamsurvey/web_data.html
• Stream Site Info: EPA Wadeable Streams Assessment (Filename:
wsa_siteinfo_ts_final.csv). Available online at:
http ://www. epa. gov/o wow/stream survey/web data.html
• Hydrologic Units - http://water.usgs.gov/GIS/metadata/usgswrd/XML/huc250k.xml
Processing Steps:
(1) Benthic macroinvertebrate data were downloaded and opened as a comma-delimited text
file in Microsoft Excel. The columns containing the SitelD and Macroinvertebrate Index
(MMI_WSA) attributes were saved as a new comma-delimited text file.
(2) The text file with macroinvertebrate data, the hydrologic units, and Site Info were opened
in ArcMap 9.2. The macroinvertebrate and Site Info tables were joined using the SitelD
attribute.
(3) An event theme was mapped for the joined records with corresponding macroinvertebrate
data, using the LON_DD and LAT_DD fields (North American Datum 1983). This event
theme was exported as a shapefile.
(4) The sites were aggregated within the 4-digit HUCs using a spatial join. If a HUC
contained more than one site, the average value for the Macroinvertebrate Index was
calculated.
(5) The final map was created using the Avg_MMI_WS field to indicate low, medium, and
high vulnerability categories. In 4-digit HUCs where no sampling occurred, the HUC was
mapped assigned a 'No Data' category.
Do Not Cite or Quote Page E-l 7
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
#461 Macroinvertebrate Observed/Expected Ratio ofTaxa Loss
Data Sources (see Appendix C for more details):
• Stream Water Benthic Macroinvertebrate Metrics: EPA Wadeable Streams Assessment.
(Filename: wsa_benmet300_ts_fmal.csv). Available online at:
http://www.epa.gov/owow/streamsurvey/web data.html
• Stream Site Info: EPA Wadeable Streams Assessment. (Filename:
wsa_siteinfo_ts_final.csv). Available online at:
http ://www. epa. gov/o wow/stream survey/web data.html
• Hydrologic Units - http://water.usgs.gov/GIS/metadata/usgswrd/XML/huc250k.xml
Processing Steps:
(1) Benthic macroinvertebrate data were downloaded and opened as a comma-delimited text
file in Microsoft Excel. The columns containing the SitelD and Macroinvertebrate O/E
Ratio (OE_5_3REG) attributes were saved as a new comma-delimited text file.
(2) The text file with macroinvertebrate data, the hydrologic units, and Site Info were opened
in ArcMap 9.2. The macroinvertebrate and Site Info tables were joined using the SitelD
attribute.
(3) An event theme was mapped for the joined records with corresponding macroinvertebrate
data, using the LON_DD and LAT_DD fields (North American Datum 1983). This event
theme was exported as a shapefile.
(4) The sites were aggregated within the 4-digit HUCs using a spatial join. If a HUC
contained more than one site, the average value for the O/E Ratio was calculated.
(5) The final map was created using the Avg_OE_5_3 field to indicate low, medium, and
high vulnerability categories. In 4-digit HUCs where no sampling occurred, the HUC was
mapped assigned a 'No Data' category.
#623 Water Availability: Net Stream/low per Capita
Data Sources (see Appendix C for more details):
• Mean Annual Precipitation 1971-2000: PRISM Climate Group (Oregon State
University). Available online at:
http://www.prism.oregonstate.edu/products/matrix.phtml
• Mean Daily Maximum Temperature 1971-2000: PRISM Climate Group (Oregon State
University). Available online at:
http://www.prism.oregonstate.edu/products/matrix.phtml
• United States Geological Survey (USGS). Water Usage, 1995
http://water.usgs.gov/watuse/spread95/ush895.txt
This data set includes population within each HUC-8 unit.
• Hydrologic Units - http://water.usgs.gov/GIS/metadata/usgswrd/XML/huc250k.xml
Processing Steps:
(1) PRISM data were downloaded and converted to ESRI© GRID format using the ASCII
Text to Grid tool in ArcMap 9.2.
(2) Mean daily temperature was calculated with the raster calculator using the following
formula as described by Vogel, 1999: (Max Temp + Min Temp) / 2.
Do Not Cite or Quote Page E-18
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
(3) Mean precipitation and mean daily temperature within each 4-digit HUC were calculated
using the Zonal Statistics function.
(4) A new attribute for mean annual streamflow was calculated with the field calculator in
ArcMap for each HUC-4 unit using the regression equations in Vogel, 1999 (Table 2).
The HUC-2 code was used to associate each HUC-4 with the appropriate regional
regression equation.
(5) The water usage data table was joined to the attribute table for the hydrologic units
shapefile, using the "HUC_CODE" and "HUCSCode" fields.
(6) The 8-digit HUC regions were aggregated into 4-digit HUCs with the 'dissolve' function
in ArcMap. The "SUB" attribute was used as the basis for the dissolve process. During
the dissolve operation, summary statistics were calculated for total population and total
withdrawals.
(7) Streamflow and withdrawals were adjusted for some HUC-4 units to account for
streamflow and withdrawals upstream of the HUC-4 unit.
(8) A new attribute for net per capita streamflow was calculated with the field calculator,
using the following equation:
(Streamflow - Withdrawals) / Population
The units for streamflow and population were converted as needed.
(9) The final map was produced indicating low, medium, and high vulnerability categories.
Do Not Cite or Quote Page E-19
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Do Not Cite or Quote Page E-20
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Appendix F. Example Maps for Indicators of Water Quality and Aquatic Ecosystem Vulnerability
byHUC-4 Watershed
Do Not Cite or Quote Page F-l
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Do Not Cite or Quote Page F-2
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#1 Acid Neutralizing Capacity, 2000-2004
Percent of Sites with ANC < 100 millieq/L
o%
0.0001%- 10.00%
10.01%-17.39%
17.40%-25.00%
25.01%-100.0%
No Data
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
Page F-3
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#22 At-Risk Freshwater Plant Communities, 2006
Percent of At-Risk Freshwater Plant
Communities
8.708-38.91
38.92-48.02
48.03-52.24
52.25-56.50
56.51 - 100.0
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
Page F-4
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#24 At-Risk Native Freshwater Species, 2009
Percent of At-Risk Native Freshwater
Species
2.135%-4.032%
4.033%-6.314%
6.315%-9.839%
9.840%-12.22%
12.23%-25.25%
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
Page F-5
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#51 Coastal Vulnerability Index, 2001
Coastal Vulnerability Index
1.00-1.77
1.78-2.41
2.42-2.83
2.84-3.18
3.19-3.97
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
Page F-6
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#125 Groundwater Reliance, 1995
Percent of Water Withdrawals from
Groundwater
0.080% - 4.284%
4.285%-12.53%
12.54%-26.80%
26.81%-54.94%
54.95% - 99.94%
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageF-7
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#165 Meteorological Drought Indices, 2003-2007
Average Palmer Drought Severity Index
1.39-15.0
0.308-1.38
-0.214-0.307
-0.931 --0.215
-7.33--0.932
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
Page F-8
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#218 Ratio of Snow to Total Precipitation, 1998-2007
Ratio of Total Snowfall to
Total Precipitation
0 - 0.0040
0.0041 - 0.036
0.037-0.11
0.12-0.19
0.20 - 0.47
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
Page F-9
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#219 Ratio of Water Withdrawls to Annual Streamflow,
1995
Total Withdrawls /
Annual Streamflow
0.00068-0.055
0.056-0.16
0.17-0.53
0.54-1.5
1.6-59
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageF-10
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#284 Stream Habitat Quality, 2000-2004
Average Rapid
Bioassessment Protocol Score
40.0-109.2
109.3- 125.0
125.1 -135.6
135.7- 147.0
147.1 -190.0
No Data
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageF-11
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#326 Wetland and Freshwater Species At-Risk, 2006
Number of Wetland and Freshwater
Species At-Risk
0-5
6-10
11 -15
16-28
29-161
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageF-12
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#348 Erosion Rate, 1980
Soil Loss (tons/ha/year)
0.5391 -2.888
2.889-4.205
4.206-5.861
5.862-9.594
9.595-25.57
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageF-13
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#351 In-Stream Use / Total Streamflow
In-Stream Use / Total Streamflow
0.60-1.00
1.01 -1.09
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageF-14
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#352 Total Use / Total Streamflow
Total Use / Total Streamflow
0.60-1.00
1.01 -17.82
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageF-15
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#364 Pesticide Toxicity Index, 1992-2001
Pesticide Toxicity Index
for Daphnia species
0.0000020 - 0.00028
0.00029-0.0017
0.0018-0.0072
0.0073-0.017
0.018-0.16
No Data
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageF-16
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#367 Herbicide Concentrations in Streams, 1993-2001
Herbicide Concentrations (ug/L)
0.00975-0.0837
0.0838 - 0.243
0.244-0.857
0.858-2.66
2.67-18.4
No Data
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageF-17
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#369 Insecticide Concentrations in Streams, 1993-2001
Insecticide Concentrations (ug/L)
0.000-0.007
0.008-0.020
0.021 - 0.048
0.049-0.115
0.116-0.681
No Data
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageF-18
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#371 Organochlorines in Bed Sediment, 1991-1997
Organochlorines in
Streambed Sediment
(ug/kg dry weight)
0.000- 7.834
7.835-32.00
32.01 - 75.82
75.83- 175.2
175.3-453.0
No Data
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageF-19
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#373 Herbicides in Groundwater, 1992-2003
Herbicides in Groundwater (ug/L)
0.00 - 0.0029
0.0030-0.016
0.017-0.063
0.064-0.20
0.21 -2.3
No Data
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageF-20
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#374 Insecticides in Groundwater, 1992-2003
Insecticides in Groundwater (ug/L)
0.00
0.0000010-0.00026
0.00027 - 0.00074
0.00075-0.0031
0.0032 - 0.28
No Data
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageF-21
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#437 Precipitation Elasticity of Streamflow, 1951-1988
Precipitation Elasticity of Streamflow
0.0-1.0
1.1 -3.0
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageF-22
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#449 Ratio of Reservoir Storage to Mean Annual Runoff
Reservoir Storage (acre-feet) /
Mean Annual Runoff (in.)
1,408,421 -73,371,814
394,810-1,408,420
133,419-394,809
53,513-133,418
0-53,512
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageF-23
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#453 Runoff
', 1984-1993
Coefficient of Variation
of Annual Runoff
0.170-0.250
0.251 - 0.293
0.294-0.335
0.336 - 0.426
0.427- 1.111
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageF-24
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#460 Macroinvertebrate Index of Biotic Condition,
2000-2004
Average Macroinvertebrate Index
51.80-77.94
44.98-51.79
37.54-44.97
29.50-37.53
4.20-29.49
No Data
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageF-25
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#461 Macro!nvertebrate O/E Ratio of
2000-2004
Average Observed/
Expected Ratio of Taxa Loss
^B 71.12%-80.95%
^^| 80.96% - 87.45%
87.46% - 96.87%
96.88%-127%
No Data
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageF-26
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#623 Water Availability: Net Streamflow per Capita, 1995
[Flow (gpd) - Withdrawals (gpd)]/
Population
24,220- 1,779,536
7,465-24,219
2,438 - 7,464
1 - 2,437
0
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageF-27
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Appendix G. Descriptions of Example Indicator Maps
by HUC-4 Watershed
Do Not Cite or Quote Page G-l
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
This appendix describes the 25 example maps of vulnerability indicators presented in Appendix
F. Descriptions of U.S. geographical regions and are based on the definitions provided by the
U.S. Census Bureau. Subregions were based on U.S. Census definitions, but modified slightly
for clarity.
Maryland
1. Northeast iv. North Carolina
South Carlolina
Connecticut
Maine
Massachussetts
New Hampshire
Rhode Island
Vermont
Middle Atlantic
New Jersey
Northeast
a. New England
i.
ii.
iii.
iv.
v.
vi.
in.
iv.
v.
vi.
vii.
Virgina
West Virginia
Southeast
Florida
i.
ii. New York
iii. Pennsylvania
2. Midwest
a. Great Lakes
i. Indiana
ii. Illinois
iii. Michigan
iv. Ohio
v. Wisconsin
b. Western Midwest
i. Iowa
ii. Kansas
iii. Minnesota
iv. Missouri
v. Nebraska
vi. North Dakota
vii. South Dakota
3. South
a. South Atlantic
i. Delaware
ii. Distric of Columbia
i.
ii.
iii.
iv.
v.
vi.
Georgia
Kentucky
Alabama
Mississippi
Tennessee
c. Central South
i.
ii.
iii.
iv.
Texas
Oklahoma
Arkansas
Louisiana
4.
West
a. Mountain West
i. Arizona
ii. Colorado
iii. Idaho
iv. New Mexico
v. Montana
vi. Utah
vii. Nevada
viii. Wyoming
b. Pacific West
i. California
ii. Oregon
iii. Washington
#1 Acid Neutralizing Capacity
This continental U.S. indicator map shows the percentage of sites with Acid Neutralizing
Capacity (ANC) less than 100 millieq/L in each HUC-4 area. Data were available for the vast
majority of lower-48 watersheds. The majority of watersheds are at 0%. Most of the watersheds
with less ANC are a narrow band which spans from the Southeast to the Northeast. Only six
watersheds are in the lowest category of ANC (25.01 - 100% of sites <100 millieq/L).
Do Not Cite or Quote
Page G-2
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
#22 At-Risk Freshwater Plant Communities
The continental U.S. map for this indicator shows the percentage of freshwater plant
communities that are considered at-risk in each HUC-4 area. Data were available for all lower-48
watersheds. The regions with the highest percentages of freshwater plant communities at risk
(56.51 - 100.0% at risk) for this indicator occur in the South Atlantic, Southeast, Northwest,
large parts of Kansas, Missouri, Indiana, Ohio, and Louisiana. Relatively high percentages of
plant communities at risk (52.25 - 56.50%) occur in Texas and in parts of the Mountain West and
Midwest, extending eastward from Wyoming to Ohio.
Moderate percentages of plant communities at risk (48.03 - 52.24%) occur in a
contiguous band in the Southwest, and in large parts of Montana, South Dakota, and Arkansas.
Relatively lower percentages of communities at risk (38.92 - 48.02%) occur in two vertical bands
in the northern Midwest region, and a horizontal band in the Southwest, including parts of
Arizona, New Mexico, and small parts of Texas and Oklahoma. The Northeast and parts of
Minnesota, Iowa, and Arizona have the lowest percentages (8.708 - 38.91%) of at-risk
freshwater plant communities.
#24 At-Risk Native Freshwater Species
This continental U.S. indicator map shows the percentage of at-risk native freshwater
species in each HUC-4 area. Data were available for all lower-48 watersheds. This map displays
a very clear pattern. Homogenous blocks of high percentages of risk (12.23 - 25.25%) are found
in the Southwest, East Texas, and the Southeast. With very few exceptions, risk is a steady
gradation from these areas to New England and the central U.S., which are at very low
percentages of risk (2.135 - 4.032%). The Chesapeake Bay is also an area with low percentage of
species at risk.
#57 Coastal Vulnerability Index- CVI
This continental U.S. indicator map shows the Coastal Vulnerability Index (CVI) for
coastline areas. Data were available for all lower-48 coastlines. Areas of high vulnerability (3.19
- 3.97) include parts of the California, Texas, and North Carolina, as well as the entire
Mississippi Delta coastline and the Chesapeake Bay. Areas of moderate (2.84 - 3.18), medium
(2.42 - 2.83), and low (1.78 - 2.41) vulnerability are interspersed along the coastline. Very low
(1.00 - 1.77) vulnerability occurs mostly on the Northeast coastline.
#725 Groundwater Reliance
This continental U.S. indicator map shows the percentage of groundwater reliance in each
HUC-4 area. Data were available for all lower-48 watersheds. A high level (54.95 - 99.94%) of
groundwater reliance is mainly observed in a vertical band in the Midwest, stretching from parts
of North Dakota to much of West Texas, as well as in two clusters in the Southwest and along
the Mississippi River. Moderate to low (4.285 - 54.94%) groundwater reliance is observed
scattered across the nation. The main area with almost no groundwater reliance (0.080 - 4.284%)
is in the Mountain West, and stretches from Montana to the Four Corners. Other watersheds with
almost no groundwater reliance are scattered across the nation, most notably in central Texas;
which is in direct contrast with adjacent watersheds with high groundwater reliance.
Do Not Cite or Quote Page G-3
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
#165 Meteorological Drought Indices
This continental U.S. indicator map shows the average Palmer Drought Severity Index
(PDSI) in each HUC-4 area. Data were available for all lower-48 watersheds. A high PDSI
values (1.39 to 15) are observed mainly in the Northeast. Moderate and low PDSI values (0.308
to 1.38, and -0.214 to 0.307) are observed in the central Midwestern states. Very low and
extremely low PDSI values (-0.931 to -0.215, and -7.33 to -0.932) are observed mainly in the
Northwest, southern Mountain West, Central South, and the Southeast.
#218 Ratio of Snow to Total Precipitation
This continental U.S. indicator map shows the ratio of total snowfall to total precipitation
in each HUC-4 area. Data were available for all lower-48 watersheds. Unsurprisingly, this map
shows a strong north-south trend, with the highest ratios (0.20 - 0.47) in the northen and
mountainous regions, including the West, Great Lakes region, and parts of New England. These
high ratios are surrounded by graded bands of moderate (0.12 - 0.19), low (0.037 -0.11), and
very low (0.0041 - 0.036) snowfall to total precipitation ratios. Parts of California and Arizona
have a ratio of zero, as does the Gulf Region.
#219 Ratio of Water Withdrawals to Annual Stream/low
This continental U.S. indicator map shows the ratio of water withdrawals to annual
streamflow in each HUC-4 area. Data were available for all lower-48 watersheds. High ratios
(1.6 - 59) are almost exclusively found in the West (with small exceptions in the Lower
Peninsula of Michigan and the Buffalo region). Moderate ratios (0.54 - 1.5) are found largely in
the West, along the Middle Atlantic Corridor, in Florida, and in the Great Lakes region. Low
(0.17 - 0.53) and very low (0.056 - 0.16) ratios are scattered throughout the country, but with
higher prevalence in the East. Ratios of almost zero (0.00068 - 0.055) are found largely near the
Mississippi River or tributaries, in the Pacific Northwest, and in New England.
#284 Stream Habitat Quality
This continental U.S. indicator map shows stream habitat quality, as defined by the
average rapid bioassessment protocol score, in each HUC-4 area. Data were available for the vast
majority of lower-48 watersheds. The highest scores (147.1 - 190.0 and 135-.7 - 147.0) are
scattered throughout the country, with clusters found in the South Atlantic and Northeast, the
northern Mountain West, and the Pacific Northwest. Moderate scores (125.1 - 135.6 and 109.3 -
125.0) are also found throughout the country, with clusters in the Northwest and Great Lakes
Region. The lowest scores (40.0 - 109.2) are found in Georgia and a vertical band in the
Midwest.
#326 Wetland and Freshwater Species at Risk
This continental U.S. indicator map shows the number of wetland and freshwater species
that are at risk in each HUC-4 area. Data were available for all lower-48 watersheds. A large
number (29 - 161) of species are at risk in most of the watersheds in the Southeast, and in a few
watersheds in the Northeast and West. Watersheds with a moderate (16 - 28) number of species
at risk are largely found near watersheds with a high number of species at risk. Watersheds with
a low (11 - 15) and very low (6 - 10) number of species at risk are found everywhere but the
Do Not Cite or Quote Page G-4
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Southeast. Watersheds with almost no (0 - 5) species at risk are mostly found in the northern
Mountain West and Western Midwest
#348 Erosion Rate
This continental U.S. indicator map shows the average erosion rate in each HUC-4 area.
Data were available for all lower-48 watersheds. High (9.595 - 25.57 tons/ha/year) and moderate
(5.862 - 9.594 tons/ha/year) soil loss is found principally in the West, Middle Atlantic, and parts
of the Southeast. Lower (0.5391 - 5.861 tons/ha/year) soil loss rates are found scattered about the
country and in a vertical band from Montana to Texas.
#351 Instream Use/Total Stream/low
This continental U.S. indicator map shows the ratio of in stream use to total streamflow
in each HUC-4 area. Data were available for all lower-48 watersheds. All watersheds but one fall
in the 0.60 to 1.00 category, with one watershed in Oklahoma and Kansas in the 1.01 to 1.09
category.
#352 Total Use / Total Streamflow
This continental U.S. indicator map shows the ratio of total use to total streamflow in
each HUC-4 area. Data were available for all lower-48 watersheds. Most watersheds fall in the
0.60 to 1.00 category. However there are a few in the Southwest and southern Midwest in the
1.01 to 17.82 category.
#364 Pesticide Toxicity Index
This continental U.S. indicator map shows the Daphnia species pesticide toxicity index in
each HUC-4 area. Data were not available for many of the lower-48 watersheds. Available data
is insufficient to infer geographic patterns.
#367 Herbicide Concentrations in Streams
This continental U.S. indicator map shows herbicide concentrations in streams in each
HUC-4 area. Data were not available for many of the lower-48 watersheds. Available data is
insufficient to infer geographic patterns.
#369 Insecticide Concentrations in Streams
This continental U.S. indicator map shows insecticide concentrations in streams in each
HUC-4 area. Data were not available for many of the lower-48 watersheds. Available data is
insufficient to infer geographic patterns.
#371 Organochlorines in Bed Sediment
This continental U.S. indicator map shows organochlorine concentrations in streambed
sediment in each HUC-4 area. Data were not available for many of the lower-48 watersheds.
Available data is insufficient to infer geographic patterns.
Do Not Cite or Quote Page G-5
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
#373 Herbicides in Groundwater
This continental U.S. indicator map shows herbicide concentrations in groundwater in
each HUC-4 area. Data were available for many of the lower-48 watersheds. Available data is
insufficient to infer geographic patterns.
#374 Insecticides in Groundwater
This continental U.S. indicator map shows insecticide concentrations in groundwater in
each HUC-4 area. Data were not available for many of the lower-48 watersheds. Available data
is insufficient to infer geographic patterns, but does indicate the possibility of higher
concentrations in the Middle Atlantic Corridor.
#437Precipitation Elasticity of Stream/low
This continental U.S. indicator map shows the precipitation elasticity of streamflow in
each HUC-4 area. Data were available for all lower-48 watersheds. Every watershed has
elasticity in the higher range (1.1 - 3.0) except for a few scattered throughout the Midwest and
one in Texas, which have elasticity in the lower range (0.0 - 1.0).
#449 Ratio of Reservoir Storage to Mean Annual Runoff
This continental U.S. indicator map shows the ratio of reservoir storage to mean annual
runoff in each HUC-4 area. Data were available for all lower-48 watersheds. High (1,408,421 -
73,371,814 acre-feet/inch) ratios are largely found in the vertical band between North Dakota
and Texas. Moderate ratios (394,810 - 1,408,420 acre-feet/inch) are largely found in the Midwest
and the West. Low (133,419 - 394,809 acre-feet/inch), very low (53,513 - 133,418 acre-
feet/inch), and extremely low (0-53,512 acre-feet/inch) ratios are largely found in coastal and
Great Lakes watersheds.
#453 Runoff Variability
This continental U.S. indicator map shows the coefficient of variation of annual runoff in
each HUC-4 area. Data were available for all lower-48 watersheds. A high (0.427 - 1.111)
coefficient is observed in clusters covering much of the West, an area in the Midwest centered on
Iowa, and part of Texas. Watersheds with a moderate (0.336 - 0.426) ratio are observed adjacent
to those clusters as well as in Maine and the Chesapeake region. Watersheds with a low (0.294 -
0.335) and very low (0.251 - 0.293) ratio are observed across the country. The watersheds with
the lowest ratio (0.170 - 0.250) are south of the Great Lakes, in New England, and the lower
Mississippi Basin.
#460 Macroinvertebrate Index ofBiotic Condition
This continental U.S. indicator map shows the macroinvertebrate index of biotic
condition in each HUC-4 area. Data were available for the vast majority of lower-48 watersheds.
There is no discernable geographic pattern to the distribution of categories of watersheds.
Do Not Cite or Quote Page G-6
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
#461 Macroinvertebrate Observed/Expected (O/E) Ratio ofTaxa Loss
This continental U.S. indicator map shows the observed taxa as a percentage of the
expected macroinvertebrate taxa in each HUC-4 area. Data were available for the vast majority
of lower-48 watersheds. This map shows a certain amount of spatial heterogeneity. The highest
ratios (96.88% - 127%) occur mostly along the West Coast, in the Pacific Northwest, the
Midwest, and New England. Moderate ratios (87.46% - 96.87%) are found in large parts of
California, parts of the Northwest, Great Lakes, and South and Middle Atlantic regions. The
remaining ratio categories (20.19% - 71.11%, 71.12% - 80.95%, and 80.96% - 87.45%) have no
discernable geographic distribution.
#623 Water Availability: Net Stream/low per Capita
This continental U.S. indicator map shows the net streamflow per capita in each HUC-4
area. Data were available for all lower-48 watersheds. High flow per capita (24,220 - 1,779,536
gpd/capita) watersheds are found in the Pacific Northwest, Colorado and Utah, the Mississippi
Basin, and Maine. Moderate streamflow per capita (2,438 - 7,464 gpd/capita) watersheds are
found mostly around high streamflow per capita watersheds. Very low streamflow per capita (1 -
2,437 gpd/capita) watersheds are found in the Great Lakes region, the Middle Atlantic Corridor,
Florida, and the West. Zero net streamflow per capita watersheds are found in the Great Lakes
Region, and throughout the West.
Do Not Cite or Quote Page G-7
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Do Not Cite or Quote Page G-8
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Do Not Cite or Quote Page H-l
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Do Not Cite or Quote Page H-2
-------
#1 Acid Neutralizing Capacity, 2000-2004
Percent of Sites with ANC < 100 millieq/L
0%
0.01%-4.17%
4.18%-11.11%
11.12%-27.27%
27.28% - 66.67%
No Data
States
0 100 200 300 400 500 Miles
I I I I I I
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#22 At-Risk Freshwater Plant Communities, 2006
Percent of At-Risk Freshwater Plant Communities
0.0% - 31.3%
31.4% - 45.7%
45.8% - 51.8%
51.9% - 55.5%
55.6%-71.1%
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageH-4
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#24 At-Risk Native Freshwater Species, 2006
Percentage of Freshwater Species At-Risk
0.5% - 4.6%
4.7% - 9.3%
9.4%-10.6%
10.7%-15.3%
15.4%-22.7%
States
0 100 200 300 400 500 Miles , s>''
I I I I I I
Do Not Cite or Quote
PageH-5
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#125 Groundwater Reliance, 1995
Percent of Water Withdrawals from Groundwater
1.7%-4.8%
4.9%- 13.4%
13.5%-23.7%
23.8% - 42.2%
42.3% - 78.7%
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageH-6
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#165 Meteorological Drought Indices, 2003-2007
Average Palmer Drought Severity Index
1.29-3.07
0.52- 1.28
-0.16-0.51
-0.41 --0.17
-2.79--0.42
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageH-7
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#218 Ratio of Snow to Total Precipitation, 1998-2007
Ratio of Total Snow to Total Precipitation
0.000-0.003
0.004-0.031
0.032-0.112
0.113-0.174
0.175-0.821
No Data
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageH-8
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#219 Ratio of Water Withdrawals to
Annual Streamflow, 1995
Total Withdrawals /Annual Streamflow
0.00-0.03
0.04-0.18
0.19-0.43
0.44 - 0.48
0.49 - 4.25
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageH-9
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#284 Stream Habitat Quality, 2000-2004
Average Rapid Bioassessment Protocol Score
78.5-114.4
114.5-126.8
126.9-135.8
135.9-146.0
146.1 -182.8
No Data
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageH-10
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#326 Wetland and Freshwater Species At-Risk, 2006
Number of Wetland and Freshwater Species At-Risk
| | 0-12
| | 13-19
I I States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageH-11
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
Soil Loss (tons/ha/year)
0.00-2.02
2.03-3.18
3.19-5.44
5.45 - 9.38
9.39-38.41
States
#348 Erosion Rate, 1980
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageH-12
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#351 Instream Use / Total Stream/low
Instream Use / Total Streamflow
0.6- 1.0
> 1.0
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageH-13
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#352 Total Use / Total Streamflow
Total Use / Total Streamflow
0.601 -1.000
1.001 -4.187
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageH-14
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#364 Pesticide Toxicity Index, 1992-2001
Pesticide Toxicity Index for Daphnia
0.0000-0.0001
0.0002 - 0.0022
0.0023 - 0.0078
0.0079-0.0142
0.0143-0.0926
No Data
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageH-15
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#367 Herbicide Concentrations in Streams, 1993-2001
Herbicide Concentration (ug/L)
0.00-0.08
0.09-0.30
0.31 -1.22
1.23-2.42
2.43-16.06
No Data
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageH-16
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#369 Insecticide Concentrations in Streams, 1993-2001
Insecticide Concentration (ug/L)
0.000-0.006
0.007-0.020
0.021 -0.044
0.045-0.107
0.108-0.751
No Data
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageH-17
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#371 Organochlorines in Streambed Sediment, 1991-1997
Organochlorines in Streambed Sediment (ug/kg dry weight)
0.14-1.33
1.34-3.45
3.46 - 8.20
8.21 -17.77
17.78-136.22
No Data
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageH-18
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
Herbicides in Groundwater (ug/L)
0.000-0.003
0.004-0.015
0.016-0.059
0.060-0.168
0.169-2.162
No Data
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageH-19
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#374 Insecticides in Groundwater, 1992-2003
Insecticides in Groundwater (ug/L)
0.0000-0.0002
0.0003-0.0006
0.0007-0.0010
0.0011 -0.0058
0.0059-0.1183
No Data
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageH-20
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#437 Precipitation Elasticity of Streamflow, 1951-1988
Precipitation Elasticity of Streamflow
0.72- 1.00
1.01 -3.15
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageH-21
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#449 Ratio of Reservoir Storage to Mean Annual Runoff
Reservoir Storage (acre-feet) /Annual Runoff (in.)
8,330,000- 161,000,000
850,000 - 8,320,000
202,000 - 849,000
109,000-201,000
0- 108,000
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageH-22
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#453 Runoff Variability, 1984-1993
Coefficient of Variation of Annual Runoff
0.183-0.267
0.268 - 0.276
0.277 - 0.323
0.324 - 0.352
0.353- 1.193
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageH-23
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#460 Macroinvertebrate Index of Biotic Condition,
2000-2004
Average Macroinvertebrate Index
55.6 - 66.3
44.0 - 55.5
40.4 - 43.9
30.5-40.3
2.4-30.4
No Data
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageH-24
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#461 Macroinvertebrate O/E Ratio of Taxa Loss,
2000-2004
Average Observed/Expected Ratio
94.76%-149.07%
88.66% - 94.75%
80.71%-88.65%
68.31%-80.70%
20.19%-68.30%
No Data
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageH-25
-------
Aquatic Ecosystems, Water Quality, and Global Change: Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft, February 2011
#623 Water Availability: Net Stream Flow per capita
[Flow (gpd) - Withdrawals (gpd)] / Population
24,000 - 275,000
5,800 - 24,000
2,800 - 5,700
1,000-2,700
0.0- 1,000
States
0 100 200 300 400 500 Miles
I I I I I I
Do Not Cite or Quote
PageH-26
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Appendix I. Descriptions of Example Indicator Maps
by Ecoregion
Do Not Cite or Quote Page 1-1
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
This appendix describes the 25 example maps of vulnerability indicators by ecoregion presented
in Appendix H. Descriptions of U.S. geographical regions and are based on the definitions
provided by the U.S. Census Bureau. Subregions were based on U.S. Census definitions, but
modified slightly for clarity.
1. Northeast
a. New England
i. Connecticut
ii. Maine
iii. Massachussetts
iv. New Hampshire
v. Rhode Island
vi. Vermont
b. Middle Atlantic
i. New Jersey
ii. New York
iii. Pennsylvania
2. Midwest
a. Great Lakes
i. Indiana
ii. Illinois
iii. Michigan
iv. Ohio
v. Wisconsin
b. Western Midwest
i. Iowa
ii. Kansas
iii. Minnesota
iv. Missouri
v. Nebraska
vi. North Dakota
vii. South Dakota
3. South
a. South Atlantic
i. Delaware
ii. Distric of Columbia
4.
iii. Maryland
iv. North Carolina
v. South Carlolina
vi. Virgina
vii. West Virginia
b. Southeast
i. Florida
ii. Georgia
iii. Kentucky
iv. Alabama
v. Mississippi
vi. Tennessee
c. Central South
i. Texas
ii. Oklahoma
iii. Arkansas
iv. Louisiana
West
a. Mountain West
i. Arizona
ii. Colorado
iii. Idaho
iv. New Mexico
v. Montana
vi. Utah
vii. Nevada
viii. Wyoming
b. Pacific West
i. California
ii. Oregon
iii. Washington
#1 Acid Neutralizing Capacity
This continental U.S. indicator map shows the percentage of sites with Acid Neutralizing
Capacity (ANC) less than 100 millieq/L in each ecoregion. Data were available for the vast
majority of lower-48 ecoregions. The greater part of ecoregions in the West, Midwest, and
Central South have 0% of sites with ANC less than 100 millieq/L. Several ecoregions in the
South East and South Atlantic have a moderate (4.18-11.11%) or high (11.12- 27.27%)
percentage of sites with ANC less than 100 millieq/L. Ecoregions with the highest percentage of
Do Not Cite or Quote
Page 1-2
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
sites (27.28 - 66.67%) with ANC less than 100 millieq/L cover a majority of Florida, parts of the
Northeast, and a smaller ecoregion spanning parts of Arkansas and Oklahoma.
#22 At-Risk Freshwater Plant Communities
This continental U.S. indicator map shows the percentage of freshwater plant
communities that are considered at-risk in each ecoregion. Data were available for all lower-48
ecoregions. The ecoregions with the highest percentages of freshwater plant communities at risk
(55.6 - 71.1%) for this indicator occur largely in the Southeast and the Pacific Northwest.
Relatively high percentages of plant communities at risk (51.9 - 55.5%) occur in ecoregions
extending northward from Texas to the Midwest, as well as in ecoregions scattered in the
Mountain West.
Moderate percentages of plant communities at risk (45.8 - 51.8%) occur predominantly in
the Southwest and Western Midwest regions. Relatively lower percentages of communities at
risk (31.4 - 45.7%) occur in New Mexico, the Great Lakes region, and parts of the Middle-
Atlantic. The northern Middle Atlantic and Northeast ecoregions have the lowest percentages (0
- 31.3%) of at-risk freshwater plant communities.
#24 At-Risk Native Freshwater Species
This continental U.S. indicator map shows the percentage of at-risk native freshwater
species in each ecoregion. Data were available for all lower-48 ecoregions. Ecoregions with high
percentages of freshwater species at risk (15.4 - 22.7%) are found in the Southwest and in the
Southeast. These high risk areas are surrounded by bands of moderate (10.7 - 15.3%)
percentages of freshwater species at risk. With very few exceptions, risk is a steady gradation
from high risk areas in the southern U.S. to low risk areas (0.5 - 4.6%) in New England and the
north-central U.S.
#725 Groundwater Reliance
This continental U.S. indicator map shows the percentage of groundwater reliance in each
ecoregion. Data were available for all lower-48 ecoregions. A high level (42.3 - 78.7%) of
groundwater reliance is mainly observed in the ecoregions that stretch from parts of North
Dakota through parts of the western Midwest and Mountain West to western Texas. Other areas
with high groundwater reliance are found in the ecoregions along the U.S.-Mexico border, in
ecoregions along the lower Mississippi River, and in parts of California and Nevada. Moderate to
low (4.9 - 42.2%) groundwater reliance is observed scattered across the nation. The main
ecoregions with almost no groundwater reliance (1.7 - 4.8%) stretch from Montana to the Four
Corners and also occur in parts of the Southeast, South, and Middle Atlantic regions.
#165 Meteorological Drought Indices
This continental U.S. indicator map shows the average Palmer Drought Severity Index
(PDSI) in each ecoregion. Data were available for all lower-48 ecoregions. Negative values of
the PDSI indicate drought, positive values indicate excess rainfall, while 0 represents normal
conditions for a given region. A very distinctive pattern emerges on this map. Ecoregions having
the lowest PDSI values (-2.79 to -0.42) occur predominantly in the West, but also in along the
Great Lakes, and parts of the Central South. Low (-0.41 to -0.17) and moderate (-0.16 to 0.51)
Do Not Cite or Quote Page 1-3
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
PDSI values are observed in the central states. High (0.52 to 1.28) and very high (1.29 to 3.07)
PDSI values observed mainly in parts of the Western Midwest, Great Lakes, South Atlantic, and
Northeast.
#218 Ratio of Snow to Total Precipitation
This continental U.S. indicator map shows the ratio of total snowfall to total precipitation
in each ecoregion. Data were available for all lower-48 ecoregions. Unsurprisingly, this map
shows a strong north-south trend, with the ecoregions with the highest ratios (0.175 - 0.821)
occurring in the northern and mountainous regions including the northern West, Great Lakes,
and parts of the Northeast. These high ratios are surrounded by graded bands of moderate (0.113
- 0.174), low (0.032 - 0.112) and very low (0.004 - 0.031) snowfall to total precipitation ratios.
Parts of California, Arizona, the Central South, and the Southeast have a ratio of zero, indicating
no snowfall.
#219 Ratio of Water Withdrawals to Annual Stream/low
This continental U.S. indicator map shows the ratio of water withdrawals to annual
streamflow in each ecoregion. Data were available for all lower-48 ecoregions. High ratios (0.49
- 4.25) are almost exclusively found in the West. Ecoregions with moderate ratios (0.44 - 0.48)
cover parts of Oregon, Nevada, Idaho, Montana, Wyoming, Nebraska, Kansas, Oklahoma, and
Texas. Low (0.19 - 0.43) and very low (0.04 - 0.18) ratios are scattered throughout the country,
but with higher prevalence in the Midwest and the East. Ratios near zero (0.00 - 0.03) are found
largely near the Mississippi River or tributaries, in the Pacific Northwest, southern Wisconsin,
and Maine.
#284 Stream Habitat Quality
This continental U.S. indicator map shows stream habitat quality based on the rapid
bioassessment protocol score in each ecoregion. Data were available for the vast majority of
lower-48 ecoregions. The ecoregions with the highest bioassessment protocol scores (146.1 -
182.8 and 135.9 - 146.0) include the northern Great Lakes region, an area which extends from
parts of the Southeast to the Northeast and other areas scattered throughout the country.
Moderate scores (126.9 - 135.8) are found primarily in the upper Midwest and Mountain West.
The lowest scores (114.5 - 126.8 and 78.5 - 114.4) are found scattered in parts of the Southeast,
Midwest, Mountain West, and Pacific Northwest.
#326 Wetland and Freshwater Species at Risk
This continental U.S. indicator map shows the number of wetland and freshwater species
that are at risk in each ecoregion. Data were available for all lower-48 ecoregions. Ecoregions
with the largest number (64 - 572) of species at risk are found primarily in the Southeast and
South Atlantic regions. Ecoregions with a moderate (34 - 63) number of species at risk are
largely found near ecoregions with a high number of species at risk in the Southeast, as well in
ecoregions in the West and Midwest. Ecoregions with a low (20 - 33) and very low (13 - 19)
number of species at risk are found everywhere but the Southeast. Ecoregions with almost no (0 -
12) species at risk are mostly found primarily in the northern Midwest and Mountain West.
Do Not Cite or Quote Page 1-4
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
#348 Erosion Rate
This continental U.S. indicator map shows the average erosion rate in each ecoregion.
Data were available for all lower-48 ecoregions. Ecoregions with high (9.39 - 38.41 tons/ha/year)
and moderate (5.45 - 9.38 tons/ha/year) soil loss are found principally in the West but also in the
central Midwest and an area extending from parts of the Southeast to the Northeast. Lower (0.00
- 2.02 tons/ha/year) soil loss rates are found scattered throughout the country, including in a
vertical band from North Dakota to Texas, the Great Lakes region, and the Eastern seaboard.
#351 Instream Use/Total Stream/low
This continental U.S. indicator map shows the ratio of instream use to total streamflow in
each ecoregion. Data were available for all lower-48 ecoregions. All ecoregions have values
within the range of 0.6 - 1.0.
#352 Total Use / Total Streamflow
This continental U.S. indicator map shows the ratio of total use to total streamflow in
each ecoregion. Data were available for all lower-48 ecoregions. Most ecoregions have values
within the 0.601 - 1.000 range. However, there are a few ecoregions in the West and in an area
extending from Texas into the Central Midwest, that fall in the 1.001 to 4.187 category.
#364 Pesticide Toxicity Index
This continental U.S. indicator map shows the Pesticide Toxicity Index (PTI) for Daphnia
species in each ecoregion. Data were not available for many of the lower-48 ecoregions.
Available data is insufficient to infer geographic patterns, but does indicate the possibility of
higher pesticide toxicity index values in the central states, Southwest, and Southeast.
#367 Herbicide Concentrations in Streams
This continental U.S. indicator map shows herbicide concentrations in streams in each
ecoregion. Data were not available for many of the lower-48 ecoregions. Ecoregions with the
highest (2.43 - 16.06 |ig/L) herbicide concentrations in streams occur predominantly in the
Midwest and the lower Mississippi Basin. Ecoregions with high (1.23 - 2.42 |ig/L) and moderate
(0.31 - 1.22 |ig/L) herbicide concentrations in streams are scattered in the East, while ecoregions
with low (0.09 - 0.30 |ig/L) and very low (0.00 - 0.08 |ig/L) herbicide concentrations in streams
cover most of the Mountain West and parts of the Northeast.
#369 Insecticide Concentrations in Streams
This continental U.S. indicator map shows insecticide concentrations in streams in each
ecoregion. Data were not available for many of the lower-48 ecoregions. Available data is
insufficient to infer geographic patterns, but does indicate the possibility of higher concentrations
of insecticides in streams in the central states, Southwest, and the lower Mississippi basin.
#371 Organochlorines in Bed Sediment
This continental U.S. indicator map shows organochlorine concentrations in streambed
sediment in each ecoregion. Data were not available for many of the lower-48 ecoregions.
Do Not Cite or Quote Page 1-5
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Available data is insufficient to infer geographic patterns, but does indicate the possibility of
higher concentrations of organochlorines in streams in the East and the Southwest.
#373 Herbicides in Groundwater
This continental U.S. indicator map shows herbicide concentrations in groundwater in
each ecoregion. Data were not available for many of the lower-48 ecoregions. Ecoregions with
the highest (0.169 - 2.162 |ig/L and 0.060 - 0.168 |ig/L) concentrations of herbicides occur
predominantly in the Southeast and in the central Midwestern states. Ecoregions with moderate
(0.016 - 0.059 |ig/L) and low (0.004 - 0.015 |ig/L) concentrations of herbicides in groundwater
are scattered throughout the country, while ecoregions with the lowest (0.000 - 0.003 |ig/L)
concentrations occur primarily in the Northwest.
#374 Insecticides in Groundwater
This continental U.S. indicator map shows insecticide concentrations in groundwater in
each ecoregion. Data were available for only a fraction of lower-48 ecoregions. Ecoregions with
the highest (0.0059 - 0.1183 |ig/L and 0.0011 - 0.0058 |ig/L) concentrations of insecticides in
groundwater occur predominantly in the Southeast and South Atlantic regions. Ecoregions with
moderate (0.0007 - 0.0010 |ig/L) and low (0.0003 - 0.0006 |ig/L) concentrations of insecticides
in groundwater are scattered throughout the country, while ecoregions with the lowest (0.0000 -
0.0002 |ig/L) concentrations occur mostly in the West.
#437Precipitation Elasticity of Stream/low
This continental U.S. indicator map shows the precipitation elasticity of streamflow in
each ecoregion. Data were available for all lower-48 ecoregions. Every ecoregion has elasticity
in the higher range (1.01 - 3.15) except for one ecoregion in Nebraska, which has elasticity in the
0.72- 1.00 range.
#449 Ratio of Reservoir Storage to Mean Annual Runoff
This continental U.S. indicator map shows the ratio of reservoir storage to mean annual
runoff in each ecoregion. Data were available for all lower-48 ecoregions. Ecoregions with the
lowest (0 - 108,000 acre-feet/inch) ratios occur in the Pacific Northwest, the lower Mississippi
basin, and in Michigan. Ecoregions with low (109,000 - 201,000 acre-feet/inch) and moderate
ratios (202,000 - 849,000 acre-feet/inch) are found primarily in the Northeast, but are also
scattered throughout the country. Ecoregions with high (850,000 - 8,320,000 acre-feet/inch) and
very high (8,330,000 - 161,000,000 acre-feet/inch) ratios cover most of the country with the
highest ratios occurring in Western Midwest and Mountain West regions.
#453 Runoff Variability
This continental U.S. indicator map shows the coefficient of variation of annual runoff in
each ecoregion. Data were available for all lower-48 ecoregions. A very distinctive pattern
emerges on this map, with ecoregions with the highest (0.353 - 1.193 and 0.324 - 0.352)
coefficients covering almost the entire western half of the country. Ecoregions with moderate
(0.277 - 0.323) ratios are observed largely in the Southeast, while ecoregions with low (0.268 -
Do Not Cite or Quote Page 1-6
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
0.276) and very low (0.183 - 0.267) ratios are observed in the lower Mississippi basin, parts of
the Great Lakes region, the Northeast, and Florida.
#460 Macroinvertebrate Index ofBiotic Condition
This continental U.S. indicator map shows the macroinvertebrate index of biotic
condition in each ecoregion. Data were available for the vast majority of lower-48 ecoregions.
Ecoregions with the lowest (2.4 - 30.4) macroinvertebrate index values occur in the central
Midwest and along the Gulf Coast, while ecoregions with the highest (44.0 - 55.5 and 55.6 -
66.3) index values occur primarily in the Northwest. The remaining ecoregions with
macroinvertebrate index values in the moderate range (30.5 - 40.3 and 40.4 - 43.9) are scattered
throughout the country.
#461 Macroinvertebrate Observed/Expected (O/E) Ratio ofTaxa Loss
This continental U.S. indicator map shows the observed taxa as a percentage of the
expected macroinvertebrate taxa in each ecoregion. Data were available for the vast majority of
lower-48 ecoregions. This map shows no discernable pattern, but does indicate the possibility of
higher ratio of taxa loss in the lower Mississippi basin and parts of the Southwest.
#623 Water Availability: Net Stream/low per Capita
This continental U.S. indicator map shows the net streamflow per capita in each
ecoregion. Data were available for all lower-48 ecoregions. Ecoregions with the lowest (0.0 -
1,000 gallons/day/capita and 1,000 - 2,700 gallons/day/capita) water availability occur
predominantly in the Southwest and parts of the central Mountain West and Midwest.
Ecoregions with moderate (2,800-5,700) water availability occur in the northern Mountain West,
the Northeast, and in Florida. Ecoregions with the highest (5,800 - 24,000 gallons/day/capita and
24,000 - 275,000 gallons/day/capita) water availability occur in parts of the Midwest, Southeast,
the Northwest, and Main
Do Not Cite or Quote Page 1-7
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Do Not Cite or Quote Page 1-8
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Appendix J. Vulnerability Category Matrix
Do Not Cite or Quote Page J-l
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
The following matrix displays the data ranges for 24 mapped indicators for each of the 204 HUC-4 watersheds in the continental
United States. (Note that one mapped indicator, the Coastal Vulnerability Index (#51), is not included here because a different spatial
unit was used to map it). Values for each indicator are represented both by colors and numbers: No data (white, 0); Lowest (light gray,
1); Low (medium gray, 2); Medium (dark gray, 3); High (darker gray, 4), and Highest (black, 5). The shades of black, white, and gray
in this matrix match those on the maps in Appendix F.
Do Not Cite or Quote Page J-2
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Appendix K. Evaluation and Potential Modification of Vulnerability Indicators
Do Not Cite or Quote Page K-l
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
This appendix provides an evaluation of each of the 25 mappable indicators within the framework of the five questions presented in
the flowchart in Figure 13 (Indicator Evaluation Process) of the report. Each indicator was evaluated to determine how well it
represents vulnerability of water quality or aquatic ecosystems, and, when appropriate, how it might be modified to improve its
representation of vulnerability (Table 1). In addition, the indicators were also evaluated to examine the extent to which objective
functional thresholds may apply to them.
Table 1. Indicator Selection
Indicators are evaluated for the extent to which they represent vulnerability. The indicators can be further evaluated to determine how
to modify them to improve their representation of vulnerability. An indicator that accounts for or could account for is then sifted
through the Indicator Display (Table 2). An indicator that neither accounts for vulnerability nor can be modified to represent
vulnerability is considered inappropriate for mapping with objective breakpoints.
Indicator
ID#
Indicator
Does the indicator describe
vulnerability?
Can the indicator be modified to describe vulnerability?
Acid Neutralizing
Capacity (ANC)
Does not directly account for exposure to
acidification.
If possible, develop model to predict changes in acidity of precipitation, and the
resulting change in stream pH, given ANC.
22
At-Risk
Freshwater Plant
Communities
Does not directly account for exposure to
additional stress from climate change.
Identify plant communities that would be most susceptible to changes in
temperature or precipitation. Overlay with predicted climate changes.
24
At-Risk Native
Freshwater
Species
Does not directly account for exposure to
additional stress from climate change.
Identify species that would be most susceptible to changes in temperature or
precipitation. Overlay with predicted climate changes.
51
Coastal
Vulnerability
Index (CVI)
Yes
N/A
125
Groundwater
Reliance
Does not put groundwater reliance into
context of groundwater availability or
availability of other water sources.
Changes in groundwater availability per capita could be simulated by coupling
population projections with a groundwater model. However, these estimates would
be more meaningful if they were integrated into a model of overall water availability
that also included surface water.
Do Not Cite or Quote
Page K-2
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
Indicator
Does the indicator describe
vulnerability?
Can the indicator be modified to describe vulnerability?
165
Meteorological
Drought Indices
Does not directly account for exposure to
additional stress from climate change.
A stochastic climate model could be used to predict change in drought frequency.
218
Ratio of Snow to
Total
Precipitation
(S/P)
Does not directly account for exposure to
additional stress from climate change.
This indicator could be improved by identifying areas where the ratio of snow to
precipitation is most sensitive to a unit change in temperature. It could also be
improved by accounting for the reliance of streamflow and human water use on
snowmelt.
219
Ratio of
Withdrawals to
Stream Flow
Does not account for water shortage risk
associated with temporal variability in
streamflow and does not directly
account for exposure to additional stress
from climate change or growth in water
demand.
This indicator could be considered one factor in an integrated climatic-hydrologic
model (e.g., Wilby, R. L, P. G. Whitehead, A. J. Wade, D. Butterfield, R. J. Davis, and
G. Watts. 2006. Integrated modelling of climate change impacts on water resources
and quality in a lowland catchment: River Kennet, UK. Journal of Hydrology 330:204-
220.)
284
Stream Habitat
Quality
Does not directly account for exposure to
additional stress from climate change.
Predictions from a climate model could be used to forecast changes in streamflow,
which could be linked to stream channel stability (one component of habitat quality)
with a hydraulic model.
326
Wetland and
Freshwater
Species At Risk
Does not directly account for exposure to
additional stress from climate change.
Identify species that would be most susceptible to changes in temperature or
precipitation. Overlay with predicted climate changes.
348
Erosion Rate
Does not account for exposure to
precipitation changes.
Yang et al. (2003) provide projections of the change in erosion rate that would result
from climate change. These projections would account for both sensitivity and
exposure. However, the model for this indicator does not account for deposition of
eroded sediment and therefore cannot be solely relied upon to estimate sediment
delivery to aquatic ecosystems.
351
Instream
Use/Total
Streamflow
Does not directly account for exposure to
additional stress from climate change.
The USGS used the information in this indicator and other information to calculate
the ratio of consumptive use to renewable water supply. This indicator is a more
holistic view of water sustainability. Forecasts of the effects of climate change and
population growth on this indicator would integrate sensitivity and exposure.
Do Not Cite or Quote
Page K-3
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
Indicator
Does the indicator describe
vulnerability?
Can the indicator be modified to describe vulnerability?
352
Total Use/Total
Streamflow
Does not directly account for exposure to
additional stress from climate change.
The USGS used the information in this indicator and other information to calculate
the ratio of consumptive use to renewable water supply. This indicator is a more
holistic view of water sustainability. Forecasts of the effects of climate change and
population growth on this indicator would integrate sensitivity and exposure.
364
Pesticide
Toxicity Index
(PTI)
Does not directly account for exposure to
additional stress from climate change.
USGS is developing predictive models for individual pesticides (e.g., atrazine;
http://infotrek.er.usgs.gov/warp/). Some of these models contain precipitation
variables whose values could be adjusted to simulate the effect of climate change on
pesticide concentrations. These individual predictions could be combined to calculate
the change in PTI that would be caused by climate change.
367
Herbicide
Concentrations
in Streams
Does not directly account for exposure to
additional stress from climate change.
The pesticide indicators could be improved by: 1. comparing the concentration of an
individual pesticide to its health-based regulatory threshold (e.g., atrazine), or 2.
calculating the pesticide toxicity index.
369
Insecticide
Concentrations
in Streams
Does not directly account for exposure to
additional stress from climate change.
The pesticide indicators could be improved by: 1. comparing the concentration of an
individual pesticide to its health-based regulatory threshold (e.g., atrazine), or 2.
calculating the pesticide toxicity index.
371
Organochlorines
in Bed Sediment
Does not directly account for exposure to
additional stress from climate change.
EPA's National Sediment Quality Survey reports and maps human health and aquatic
life risk due to contaminated sediment
(http://www.epa.gov/waterscience/cs/report/1997/). Risk is based on all sediment
contaminants, so this would be a different indicator.
373
Herbicides in
Groundwater
Does not directly account for exposure to
additional stress from climate change.
The pesticide indicators could be improved by: 1. comparing the concentration of an
individual pesticide to its health-based regulatory threshold (e.g., atrazine), or 2.
calculating the pesticide toxicity index.
Do Not Cite or Quote
Page K-4
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
Indicator
Does the indicator describe
vulnerability?
Can the indicator be modified to describe vulnerability?
374
Insecticides in
Groundwater
Does not directly account for exposure to
additional stress from climate change.
The pesticide indicators could be improved by: 1. comparing the concentration of an
individual pesticide to its health-based regulatory threshold (e.g., atrazine), or 2.
calculating the pesticide toxicity index.
437
Precipitation
Elasticity of
Streamflow
Does not account for exposure to
precipitation changes.
This indicator could be combined with predicted changes in precipitation to predict
changes in streamflow.
449
Ratio of
Reservoir
Storage to Mean
Annual Runoff
Does not directly account for exposure to
additional stress from climate change.
This indicator could be considered one factor in an integrated climatic-hydrologic
model (e.g., Wilby, R. L, P. G. Whitehead, A. J. Wade, D. Butterfield, R. J. Davis, and
G. Watts. 2006. Integrated modelling of climate change impacts on water resources
and quality in a lowland catchment: River Kennet, UK. Journal of Hydrology 330:204-
220.)
453
Runoff
Variability
Does not directly account for exposure to
additional stress from climate change.
This indicator could be considered one factor in an integrated climatic-hydrologic
model (e.g., Wilby, R. L, P. G. Whitehead, A. J. Wade, D. Butterfield, R. J. Davis, and
G. Watts. 2006. Integrated modelling of climate change impacts on water resources
and quality in a lowland catchment: River Kennet, UK. Journal of Hydrology 330:204-
220.)
460
Macroinvertebra
te Index of Biotic
Condition
The stress-response curve may be
improperly characterized, and spatial
variation in exposure to future stress is
not accounted for.
Indexes of biotic condition respond linearly to stress, so vulnerability to further
degradation should be relatively constant.
461
Macroinvertebra
te
Observed/Expec
ted (O/E) Ratio
of Taxa Loss
The stress-response curve may be
improperly characterized, and spatial
variation in exposure to future stress is
not accounted for.
The scale of vulnerability for this indicator should be reversed. The first taxa that are
lost are sensitive to small amounts of stress.
Do Not Cite or Quote
Page K-5
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
623
Indicator
Water
Availability: Net
Streamflow per
Capita
Does the indicator describe
vulnerability?
Does not account for water shortage risk
associated with temporal variability in
streamflow and does not directly
account for exposure to additional stress
from climate change or growth in water
demand.
Can the indicator be modified to describe vulnerability?
It may be more appropriate to consider net streamflow in the context of instream
flow requirements. This indicator could be considered one factor in an integrated
climatic-hydrologic model (e.g., Wilby, R. L, P. G. Whitehead, A. J. Wade, D.
Butterfield, R. J. Davis, and G. Watts. 2006. Integrated modelling of climate change
impacts on water resources and quality in a lowland catchment: River Kennet, UK.
Journal of Hydrology 330:204-220.)
Do Not Cite or Quote
Page K-6
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Table 2. Indicator Display
The numerical thresholds used for indicator example maps were determined based on the information available in the literature or by
using a continuous grayscale color ramp. Indicators that already reflect vulnerability or could be modified to do so (based on Table
1) can be further evaluated to determine whether objective, functional breakpoints can be used in displaying their values. If so,
attributes necessary for determining such functional breakpoints can be identified through a review of relevant literature or through
new data collection and analysis efforts. Finally, the validity of the breakpoints when data are aggregated to the appropriate spatial
unit can be analyzed to assess the accuracy of the resultant map.
An indicator for which objective breakpoints exist, or for which objective breakpoints can be identified, and for which breakpoints
remain valid even when data are aggregated, is considered mappable with objective thresholds. An indicator for which objective
breakpoints cannot be identified or for which breakpoints are not valid after data are aggregated is considered mappable along a
continuous gradient.
Indicator
ID#
1
22
24
Indicator
Acid Neutralizing
Capacity (ANC)
At-Risk Freshwater
Plant Communities
At-Risk Native
Freshwater Species
Are objective breakpoints in the range of
vulnerability documented?
Yes; when ANC values fall below zero, the
water is considered acidic and can be either
directly or indirectly toxic to biota (i.e., by
mobilizing toxic metals, such as aluminum).
When ANC is between 0 and 25
milliequilivents, the water is considered
sensitive to episodic acidification during
rainfall events. These threshold values were
determined based on values derived from
the National Acid Precipitation Assessment
Program (USEPA 2006).
Yes; risk levels for individual communities
are semi-quantitatively defined.
Yes; risk levels for individual species are
semi-quantitatively defined.
Can objective breakpoints be identified?
N/A
N/A
N/A
Are the breakpoints valid
when the data are
aggregated?
No; indicator mapped as
percentage of sites.
No; indicator mapped as
percentage of sites.
No; indicator mapped as
percentage of sites.
Do Not Cite or Quote
PageK-7
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
Indicator
Are objective breakpoints in the range of
vulnerability documented?
Can objective breakpoints be identified?
Are the breakpoints valid
when the data are
aggregated?
51
Coastal Vulnerability
Index (CVI)
No; it indicates relative risk.
Ideally, the CVI would be calibrated to the
occurrence of actual physical effects. This
stress response relationship could then
be divided into vulnerability categories
with natural breaks, or with subjective
evaluations of acceptable risk.
Yes
125
Groundwater Reliance
No, because this indicator is not a good
measure of vulnerability.
The ratio of per capita water availability
to water use has a natural threshold: 1.
Other thresholds would be somewhat
arbitrary.
Yes, but only with suggested
modifications.
165
Meteorological
Drought Indices
The thresholds that were used are
somewhat objective because a PDSI of 0
indicates neutral conditions, and negative
numbers indicate drought. Because the
medium category is centered around zero,
it appropriately separates areas that have
experienced recent drought from those
that have not.
While there is an objective breakpoint for
separating drought from non-drought, an
objective measure of what constitutes a
critical drought frequency was not
identified.
Yes
218
Ratio of Snow to Total
Precipitation (S/P)
No
A general model of water availability that
included snowmelt could be used to
simulate changes in water availability
relative to water demand. A ratio of 1
would be an objective threshold.
Yes, but only with suggested
modifications.
219
Ratio of Withdrawals
to Annual StreamFlow
Yes, a value of 1 indicates that there is no
room for further water withdrawals.
N/A
Yes
284
Stream Habitat Quality
No, breakpoints are arbitrary.
No
N/A
Do Not Cite or Quote
Page K-&
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
Indicator
Are objective breakpoints in the range of
vulnerability documented?
Can objective breakpoints be identified?
Are the breakpoints valid
when the data are
aggregated?
326
Wetland and
Freshwater Species At
Risk
Yes; risk levels for individual species are
semi-quantitatively defined.
N/A
No; indicator mapped as
percentage of sites.
348
Erosion Rate
No, tolerable erosion rates vary among
ecosystems and are not documented at the
national scale.
With the suggested modification, three
objective categories of vulnerability
would be less erosion, no change, and
more erosion with predicted climate
change.
Yes, but only with suggested
modifications.
351
Instream Use/Total
Streamflow
Yes, a value of 1 indicates that there is no
room for further water withdrawals,
assuming that there is no consumptive use.
The same breakpoint also applies to the
suggested modification of the indicator.
Yes, because the HUC is the
original scale of
measurement.
352
Total Use/Total
Streamflow
Yes, a value of 1 indicates that there is no
room for further water withdrawals.
The same breakpoint also applies to the
suggested modification of the indicator.
Yes, because the HUC is the
original scale of
measurement.
364
Pesticide Toxicity
Index (PTI)
Pesticide toxicity index values have a built-
in threshold (1) that indicates probable
cumulative effects equivalent to an LC50 or
EC50 assuming that the additive toxicity
model is appropriate. However, even these
standard measures of toxicity are based on
a somewhat arbitrary standard of what
constitutes a serious health effect (affects
50% of test organisms).
What constitutes a critical level of risk is a
subjective choice.
There is no basis for
identifying a critical value for
the average PTI in a HUC.
Averages also obscure
variance among the values at
individual sites.
Do Not Cite or Quote
Page K-9
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
Indicator
Are objective breakpoints in the range of
vulnerability documented?
Can objective breakpoints be identified?
Are the breakpoints valid
when the data are
aggregated?
367
Herbicide
Concentrations in
Streams
No, not for mixtures of pesticides.
What constitutes a critical level of risk is a
subjective choice. Pesticide toxicity index
values have a built-in threshold that
indicates probable cumulative effects
equivalent to an LC50 or EC50 assuming
that the additive toxicity model is
appropriate. However, even these
standard measures of toxicity are based
on a somewhat arbitrary standard of
what constitutes a serious health effect
(affects 50% of test organisms).
There is no basis for
identifying a critical value for
the average probability of
exceeding a health-based
threshold or the average PTI
in a HUC. Averages also
obscure variance among the
values at individual sites.
369
Insecticide
Concentrations in
Streams
No, not for mixtures of pesticides.
What constitutes a critical level of risk is a
subjective choice. Pesticide toxicity index
values have a built-in threshold that
indicates probable cumulative effects
equivalent to an LC50 or EC50 assuming
that the additive toxicity model is
appropriate. However, even these
standard measures of toxicity are based
on a somewhat arbitrary standard of
what constitutes a serious health effect
(affects 50% of test organisms).
There is no basis for
identifying a critical value for
the average probability of
exceeding a health-based
threshold or the average PTI
in a HUC. Averages also
obscure variance among the
values at individual sites.
371
Organochlorines in
Bed Sediment
Yes, but only with the suggested
modification.
N/A
No, data would have to be
mapped as percentages for
which thresholds are
arbitrary, or averages, which
obscure variance.
Do Not Cite or Quote
PageK-10
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
Indicator
Are objective breakpoints in the range of
vulnerability documented?
Can objective breakpoints be identified?
Are the breakpoints valid
when the data are
aggregated?
373
Herbicides in
Groundwater
No, not for mixtures of pesticides.
What constitutes a critical level of risk is a
subjective choice. Pesticide toxicity index
values have a built-in threshold that
indicates probable cumulative effects
equivalent to an LC50 or EC50 assuming
that the additive toxicity model is
appropriate. However, even these
standard measures of toxicity are based
on a somewhat arbitrary standard of
what constitutes a serious health effect
(affects 50% of test organisms).
There is no basis for
identifying a critical value for
the average probability of
exceeding a health-based
threshold or the average PTI
in a HUC. Averages also
obscure variance among the
values at individual sites.
374
Insecticides in
Groundwater
No, not for mixtures of pesticides.
What constitutes a critical level of risk is a
subjective choice. Pesticide toxicity index
values have a built-in threshold that
indicates probable cumulative effects
equivalent to an LC50 or EC50 assuming
that the additive toxicity model is
appropriate. However, even these
standard measures of toxicity are based
on a somewhat arbitrary standard of
what constitutes a serious health effect
(affects 50% of test organisms).
There is no basis for
identifying a critical value for
the average probability of
exceeding a health-based
threshold or the average PTI
in a HUC. Averages also
obscure variance among the
values at individual sites.
437
Precipitation Elasticity
of Streamflow
The thresholds that were used are
somewhat objective because a value of 1
separates areas where a given percentage
change in precipitation results in a lower
percentage change in streamflow from
areas where a that same percentage
change in precipitation results in a higher
percentage change in streamflow.
Changes in streamflow could be
evaluated against in-stream flow
requirements for aquatic life.
In-stream flow requirements
tend to stream-specific, and
therefore, cannot be
generalized to all streams in
a HUC.
Do Not Cite or Quote
Page K-11
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Indicator
ID#
449
453
460
461
623
Indicator
Ratio of Reservoir
Storage to Mean
Annual Runoff
Runoff Variability
Macroinvertebrate
Index of Biotic
Condition
Macroinvertebrate
Observed/Expected
(O/E) Ratio of Taxa
Loss
Water Availability: Net
Streamflow per Capita
Are objective breakpoints in the range of
vulnerability documented?
No
No
No, breakpoints are arbitrary.
No, breakpoints are arbitrary.
Regional differences in water-using
activities mean that the sufficiency of
available water supplies varies
geographically. No documented thresholds
were found.
Can objective breakpoints be identified?
Stochastic model output from an
integrated climatic-hydrologic model
could be evaluated to identify areas
where reservoir storage is expected to
drop to zero more often than a specified
frequency.
Stochastic model output from an
integrated climatic-hydrologic model
could be evaluated to identify areas
where reservoir storage is expected to
drop to zero more often than a specified
frequency.
No
No
Possibly, although Indicator #351
(Instream Use/Total Streamflow)
describes the same concept and has an
objective threshold (1).
Are the breakpoints valid
when the data are
aggregated?
Yes, but only with suggested
modifications.
Yes, but only with suggested
modifications.
N/A
N/A
Yes
Do Not Cite or Quote
Page K-ll
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Appendix L. Contact Information
Do Not Cite or Quote Page L-l
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
The Cadmus Group, Inc.
Julie Blue, Ph.D.
The Cadmus Group, Inc.
57 Water St,
Watertown, MA 02472
E-mail: julie.blue@cadmusgroup.com
Phone: 617-673-7154
EPA
Chris Weaver, Ph.D.
Global Change Research Program (GCRP),
U.S. Environmental Protection Agency
Office of Research and Development,
National Center for Environmental Assessment
8601-P, 1200 Pennsylvania Ave.,
Washington, DC 20460
E-mail: weaver.chris@epa.gov
Phone: 703-347-8621
Technical Advisors
Thomas Meixner, Ph.D.
Associate Professor
University of Arizona,
Room 202 Harshbarger Building,
Tucson, AZ 85721
E-mail: tmeixner@hwr.arizona.edu
Phone: 520-626-1532
David Allan, Ph.D.
Professor and Acting Dean
University of Michigan,
School of Natural Resources and Environment,
2064 Dana Building, 440 Church Street,
Ann Arbor, MI 48109-1041
E-mail: dallan@umich.edu
Phone: 734-764-6553
John Day, Ph.D.
Distinguished Professor
Louisiana State University,
2237 Energy Coast and Environment Building,
LSU-Coastal Ecology Institute,
Baton Rouge, LA 70803
E-mail: johnday@lsu.edu
Phone: 225-578-6508
Kathleen Miller, Ph.D.
Scientist III
Institute for the Study of Society and
Environment (ISSE),
National Center for Atmospheric Research
(NCAR),
P.O. Box 3000, Boulder, CO 80307
E-mail: kathleen@ucar.edu
Phone:303-497-8115
Do Not Cite or Quote
Page L-2
-------
Aquatic Ecosystems, Water Quality, and Global Change:
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments
External Review Draft
February 2011
Technical Advisors (continued)
David Yates, Ph.D.
Project Scientist
Institute for the Study of Society and
Environment (ISSE) & Research Applications
Laboratory (RAL),
National Center for Atmospheric Research
(NCAR),
62 Pennsylvania St.,
Denver, CO 80203
E-mail: yates@ucar.edu
Phone: 303-497-8394
David Gochis, Ph.D.
Scientist I
Research Applications Laboratory (RAL) &
The Institute for Integrative and
Multidisciplinary Earth Studies (TIEVIES),
National Center for Atmospheric Research
(NCAR),
3450 Mitchell Lane, Boulder, CO 80307
E-mail: gochis@rap.ucar.edu
Phone: 303-497-2809
Do Not Cite or Quote
Page L-3
-------
Aquatic Ecosystems, Water Quality, and Global Change: External Review Draft
Challenges of Conducting Multi-Stressor Global Change Vulnerability Assessments February 2011
Do Not Cite or Quote Page L-4
------- |