UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON D.C. 20460
*i PRO^ OFFICE OF THE ADMINISTRATOR
SCIENCE ADVISORY BOARD
May 19, 2008
EPA-CASAC-08-011
The Honorable Stephen L. Johnson
Administrator
U.S. Environmental Protection Agency
1200 Pennsylvania Avenue, N.W.
Washington, DC 20460
Subj ect: Consultation on EPA' s Draft Scope and Methods Plan for Risk/Exposure
Assessment: Secondary NAAQS Review for Oxides of Nitrogen and Oxides
of Sulfur (March 2008 Draft)
Dear Administrator Johnson:
The Clean Air Scientific Advisory Committee (CASAC) NOX & SOX Secondary NAAQS
Review Panel met on April 3, 2008, and has completed its consultation on EPA's Draft Scope and
Methods Plan for Risk/Exposure Assessment: Secondary NAAQS Review for Oxides of Nitrogen
and Oxides of Sulfur (March 2008 Draft). The CASAC uses a consultation as a mechanism for
individual technical experts to provide comments on the Agency's draft plan for developing
technical assessments as the basis of the review of the secondary NAAQS for NOX and SOX.
Written comments provided by the individual Panelists in response to the Agency's charge
questions are enclosed with this letter.
As this is a consultation, we do not expect a formal response from the Agency. We thank
the Agency for the opportunity to provide advice early in the NAAQS review process, and look
forward to the review of the First Draft Risk and Exposure Assessment in October 2008.
Sincerely,
/Signed/
Armistead (Ted) Russell, Ph.D.
Chair, CASAC NOX & SOX Secondary Review Panel
Enclosures
-------
Appendix A: Roster of CASAC NOX & SOX Secondary NAAQS Review Panel
U.S. Environmental Protection Agency
Clean Air Scientific Advisory Committee
NOX & SOX Secondary NAAQS Review Panel
CASAC MEMBERS
Dr. Armistead (Ted) Russell (Chair), Professor, Department of Civil and Environmental
Engineering, Georgia Institute of Technology, Atlanta, GA
Dr. Ellis B. Cowling, University Distinguished Professor At-Large Emeritus, Colleges of
Natural Resources and Agriculture and Life Sciences, North Carolina State University, Raleigh,
NC
Dr. Douglas Crawford-Brown, Professor Emeritus and Director Emeritus, Department of
Environmental Sciences and Engineering and UNC Institute for the Environment, University of
North Carolina at Chapel Hill, Chapel Hill, NC
Dr. Donna Kenski, Data Analysis Director, Lake Michigan Air Directors Consortium,
Rosemont, IL
PANEL MEMBERS
Dr. Praveen Amar, Director, Science and Policy, NESCAUM, Boston, MA
Dr. Andrzej Bytnerowicz, Senior Scientist, Pacific Southwest Research Station, USDA Forest
Service, Riverside, CA
Ms. Lauraine Chestnut, Managing Economist, Stratus Consulting Inc., Boulder, CO
Dr. Charles T. Driscoll, Jr., Professor, Environmental Systems Engineering, College of
Engineering and Computer Science, Syracuse University, Syracuse, NY
Dr. Paul J. Hanson, Distinguished R&D Staff Member, Environmental Sciences Division, Oak
Ridge National Laboratory, Oak Ridge, TN
Dr. Rudolf Husar, Professor and Director, Mechanical Engineering, Engineering and Applied
Science, Center for Air Pollution Impact & Trend Analysis (CAPITA), Washington University,
St. Louis, MO
Dr. Dale Johnson, Professor, Department of Environmental and Resource Sciences, College of
Agriculture, University of Nevada, Reno, NV
Dr. Naresh Kumar, Senior Program Manager, Environment Division, Electric Power Research
Institute, Palo Alto, CA
Dr. Myron J. Mitchell, Distinguished Professor and Director of Council on Hydrologic Systems
Science, College of Environmental and Forestry, State University of New York, Syracuse, NY
-------
Mr. Richard L. Poirot, Environmental Analyst, Air Pollution Control Division, Department of
Environmental Conservation, Vermont Agency of Natural Resources, Waterbury, VT
Mr. David J. Shaw, Director, Division of Air Resources, New York State Department of
Environmental Conservation, Albany, NY
Dr. Kathleen Weathers, Senior Scientist, Institute of Ecosystem Studies, Millbrook, NY
SCIENCE ADVISORY BOARD STAFF
Ms. Kyndall Barry, Designated Federal Officer, 1200 Pennsylvania Avenue, NW
1400F, Washington, DC, Phone: 202-343-9868, Fax: 202-233-0643, (barry.kyndall@epa.gov)
-------
Appendix B: Agency Scope & Methods Plan Charge Questions
1. In outlining the scope of this risk/exposure assessment, we have created a flow diagram that
represents how nitrogen and sulfur compounds move from 'source to dose' in the environment (see
Figure 2-1). How adequately does this conceptual model for evaluating risks due to deposition-
related ecological effects characterize what should be covered in the scope of this assessment?
2. The main ecosystem effects areas we anticipate evaluating in this risk/exposure assessment are (1)
risks to terrestrial ecosystems from nitrogen enrichment effects, (2) risks to aquatic ecosystems from
nitrogen enrichment effects (eutrophication), (3) risks to terrestrial ecosystems from acidification
effects (nitrogen and sulfur), and (4) risks to aquatic ecosystems from acidification effects (nitrogen
and sulfur). We also plan to qualitatively discuss the role of sulfur enrichment on methylmercury
production and the role of nitrous oxide in climate change. What key effects areas, if any, have been
overlooked by this approach? Should the assessment plan be modified to include other effects?
3. Due to the complexity of conducting a nationwide risk/exposure assessment for welfare effects due to
NOx and SOx, we have outlined a strategy designed to identify sensitive ecosystems and a range of
harmful/adverse effects (see Figure 3-1). The seven steps are to (1) identify documented biological,
chemical and ecological effects and potential ecosystem services, (2) define sensitive areas using GIS
mapping, (3) select risk/exposure case study assessment areas, (4) evaluate current loads and effects
in case study assessment areas, (5) scale up the case study assessment areas to larger sensitive areas
where feasible, (6) assess current ecological conditions in those areas, and (7) assess alternative levels
of protection under different ambient scenarios. Does the Panel agree with this general approach?
Should it be improved or modified?
4. In the seven-step approach to the current conditions risk/exposure assessment, Step 1 (Section 3.1)
describes an approach to identify the documented effects, biological, chemical and ecological
indicators, and potential ecosystem services related to acidification and nutrient enrichment. Does
the Panel agree with this approach or can they suggest alternative approaches we should consider?
5. In the seven-step approach to the current conditions risk/exposure assessment, Step 2 (Section 3.2)
outlines a path to define areas sensitive to total reactive nitrogen and sulfur inputs. Do the Panel
members agree with this approach or are there better alternatives that should be considered?
a. We are attempting to characterize the risks to ecosystems from sulfur and nitrogen deposition
nationwide by clustering sensitive ecosystems where possible and by using the linkages
between these areas at different scales. Please comment on the adequacy of this approach.
b. How appropriate are the datasets and GIS maps listed in Table 3-4 for identifying ecosystems
sensitive to nitrogen and sulfur and/or are there others that have been overlooked?
6. In the seven-step approach to the current conditions risk/exposure assessment, Step 3 (Section 3.3)
outlines a path to identifying risk/exposure case study assessment areas.
a. Table 3-5 provides an initial list of indicators, mapping layers and multimedia models that
may be used to assess ecosystem risk and exposure. Please comment on the appropriateness
of these and suggest alternatives that may be better suited for this analysis.
b. Please comment on the list of potential case study assessment areas in Table 3-6 and Table 3-
7 and make recommendations or suggest any alternatives.
-------
7. In the seven-step approach to the current conditions risk/exposure assessment, Step 4 (Section 3.4)
outlines a path to assess current nitrogen and sulfur loads and their effects on case study assessment
areas. Does the Panel agree with how we have described our approach to identifying datasets, gaps,
and uncertainties?
a. We have initially identified the primary chemical indicator that is most suitable for assessing
ecosystem acidification effects as acid neutralizing capacity (ANC), with alternatives
depending on data availability (see section 3.4.1 and Appendix B). Does the Panel agree with
this selection, or can they suggest alternative/additional key indicators?
b. We have described the models being considered for this analysis (see section 3.4.2 and
Appendix C). Does the Panel agree with the choice of these models, and can they help
prioritize them for modeling the responses of the indicators recommended in Step 1 (Section
3.1)?
8. In the seven-step approach to the current conditions risk/exposure assessment, Step 5 (Section 3.5)
discusses how to scale up case study areas to more spatially extensive sensitive areas, where
appropriate. Does the Panel agree with this approach or can they suggest alternatives?
9. In the seven-step approach to the current conditions risk/exposure assessment, Step 6 (Section 3.6)
outlines a path to assess the current conditions of sensitive ecosystems. How well does the Panel
agree with the approach outlined for calculating response curves and utilizing mapping and
ecosystem services to characterize current conditions or can the Panel recommend alternative
approaches?
a. How well does the Panel agree with using ecosystem services to provide a common metric
for comparing ecological risks due to nitrogen and sulfur deposition effects?
b. How well does the Panel agree with collecting current valuation studies to understand the
value of bundled ecosystem services? Can the Panel recommend additional or alternative
approaches?
10. In the seven-step approach to the current conditions risk/exposure assessment, Step 7 (Section 3.7)
describes an approach to assess degrees of protection/levels of effects under alternative forms and
levels of ambient NOx and SOx standards. This approach attempts to describe how the methods,
models, and results of the current conditions risk/exposure assessment can inform our evaluation of
the appropriate form(s) and level(s) of a national standard. How well does the Panel agree with the
approach outlined in this section, the issues presented, and the 9 steps outlined to assess potential
forms and levels of the standard? Please suggest any additional or alternative steps we should take
into consideration.
11. Additional ecological/welfare effects due to NOx and SOx emissions that we do not currently
anticipate evaluating in detail in this review include the following:
Nitrogen saturation,
Maple decline,
Ammonia air deposition and toxicity to native mussels,
Relationships between acidity/nutrient enrichment and mercury methylation,
Sensitive areas for acidity/nutrient enrichment impacts, identified from biogeochemical
characteristics, and
Climate change effects due to N2O.
Does the Panel agree that these represent lower priority effects for the current assessment? If not,
what does the Panel recommend?
-------
Appendix C: Compilation of Individual Panel Member Comments on EPA's Draft Scope and Methods
Plan for Risk/Exposure Assessment: Secondary NAAQS Review for Oxides of Nitrogen and Oxides of
Sulfur (March 2008 Draft)
Comments received:
Dr. Praveen Amar 7
Dr. Andrzej Bytnerowicz 10
Ms. Lauraine Chestnut 17
Dr. Ellis B. Cowling 19
Dr. Douglas Crawford-Brown 28
Dr. Charles T. Driscoll 31
Dr. Paul J. Hanson 33
Dr. Rudolf Husar 37
Dr. Dale Johnson 40
Dr. Donna Kenski 44
Dr. Naresh Kumar 46
Dr. Myron J. Mitchell 47
Mr. Richard L. Poirot 58
Mr. David J. Shaw 63
Dr. Kathleen Weathers 66
-------
Dr. Praveen Amar
This write up provides comments in three areas, as requested by Dr. Ted Russell, the Panel
Chair. These three areas relate to:
1. Adequacy of how well the REA (Risk/Exposure Assessment) covers some of the
"additional" issues such as maple decline, ammonia air deposition, etc. The exact charge
question #11 related to "additional effects" is reproduced here :
Additional ecological/welfare effects due to NOx and SOx emissions that we do not
currently anticipate evaluating in detail in this review include the following:
Nitrogen saturation,
Maple decline,
Ammonia air deposition and toxiciry to native mussels,
Relationship between acidity/nutrient enrichment and mercury methylation,
Sensitive areas for acidity/nutrient enrichment impacts, identified from
biogeochemical characteristics, and
Climate change effects due to N£).
Does the Panel agree that these represent lower priority effects for the current
assessment? If not, what does the Panel recommend?
2. Sections 2.3.2 (overview of nitrogen deposition) and Section 2.3.3 (overview of sulfur
oxides and sulfur deposition), and
3. Overall general comments on the complete Scope and Methods Document
First, a general comment about the whole document. I think it is generally written better than the
ISA. However, various chapters can and should be improved for more clear communication. It
is understandable that various sections have different authorships. These sections and chapters
would therefore benefit from the services of an expert technical/scientific editor resulting in a
more readable document that more clearly communicates what the proposed Risk/Exposure
Assessment is expected to accomplish.
Charge Question #11
The Scope and Methods Document currently proposes to assess the areas mentioned above under
Charge question #11 in a qualitative manner (Page 15; Appendix A.2, etc.). My sense is that
some of these areas are of greater importance than others. For example, understanding of the
relationship between sulfur deposition and enhanced methylmercury (MeHg) production in
aquatic systems has serious policy implications as a number of mercury control programs from
many industrial sources (coal-fired power plants, municipal waste combustors, steel mills,
industrial boilers) are being put in place in the US at the same time major SO2 reduction
programs are being implemented (e.g., CAIR). To the extent sulfur deposition levels would be
-------
changing in the US at the same time Hg deposition levels would be changing, this "integrated"
assessment needs to be indeed "integrated."
I do not have the expertise on the direct toxicity of ammonia on mussels, however, this Risk
Assessment would benefit greatly from more emphasis on all science-related issues related to
ammonia: better quantification of its emissions, its atmospheric chemistry and physics, its fate
and transport, wet and dry deposition of ammonia and its products, and ecological effects. As
other members of this CAS AC Panel have noted in their comments, the ISA as well as the S and
M document simply seem to address the role of ammonia as an afterthought and do not address
the issue of ammonia at a level it deserves.
I suggest that the areas in Charge Question #11 (at least some of them) be given more attention
than what is being proposed. For example, instead of addressing them in any future
Risk/Exposure Assessment document in an Appendix, they be addressed in the body of the
report itself as short and separate chapters. It is recognized that some of treatments might be
more qualitative or empirical. That does not make them less important.
Sections 2.3.2 and 2.33 (Nitrogen and Sulfur Deposition):
Sections 2.32 and 2.3.4 are short. I do not think they provide true "overview" of nitrogen
deposition, or for that matter, sulfur deposition. Section 2.3.2 mostly covers the history of how
N2O was addressed in previous standards review and then notes that it will not be covered in a
quantitative way in the current assessment. It is important, as I note above, that all the
"additional issues" including the role of N2O as a greenhouse gas be covered in this document in
a clear and well-referenced manner (for example, the recent 2007 IPCC reports). One suggestion
is that a revised Scope and Methods document remove these sections from their present location
and integrate them more effectively with other sections.
Overall General Comments on the Document
Page 1, Line 7: "... deposition of ambient particles, GASES, AND ACIDS, that can.."
Page 3, Line 12 "THOUGH" is an awkward way to start this sentence.
Page 5, Second Para: "welfare effects identified should be PROTECTED from adverse effects"
Need to rewrite.
Page 6, Section 2.2: Please be very careful when using the words uncertainty and variability in
the same sentence unless you explain their clear and very different meanings. Many people take
them to mean the same thing and they simply are very different concepts, both in science and in
policy-setting.
Page 8, Last paragraph: "this assessment would focus on ecological quality and its effects from
acidification..." This is the first and the only time this document uses the words "ecological
quality." What exactly is meant? Also, the sentence needs to be rewritten for clarity.
-------
Page 9, Line 8, Please change to "precursors NOx and volatile organic COMPOUNDS..."
A General Comment: I would very much like to reinforce what Dr. Cowling noted in his
comments on the September 2007 Draft Plan. These words speak more clearly than anything I
have seen relative to what needs to be done in this assessment for the joint secondary standard(s)
for revising/modifying/establishing a new/ standard(s):
i) "What scientific evidence and/or scientific insights have been developed since the last
review to indicate if the current public-health based and/or the current public-welfare
based NAAQS need to be revised or if alternative levels, indicators, statistical forms,
or averaging times of these standards are needed to protect public health with an
adequate margin of safety and to protect public welfare? "
ii) "What scientific evidence and/or scientific insights have been developed since the last
review to indicate whether, and if so, what particular ecosystem components or other
air-quality-related public welfare values, are more or less sensitive than the
populations of humans for which primary standards are established and for this
reason may require a different level, indicator, statistical form, or averaging time of a
secondary standard in order to protect public welfare? "
Page 11, Second Paragraph, "... Framework to analyze alternative air quality standards ..." Why
not also include deposition standards as well as critical loads here? A quick review of this
document also makes it clear that critical loads as an organizing principle for establishing
standards need to be paid more attention (just like the role of ammonia needs to be paid more
attention).
Page 24 (also other places): I suggest that this document be clearer about ecosystem services,
valuation, valuation methods (both quantitative and non-quantitative). Since these areas are
covered in the environmental economics and decision analysis areas, it would be a very good
idea that they are described more clearly for the general audience of this assessment.
Page 25: The "Seven-Step Process" to undertake REA: I do not claim expertise, but Steps 5, 6,
and 7 ("when feasible, scale up case studies...", assess the current ecological conditions, assess
alternative levels of protection...", etc.) do not seem as rigorous as they would need to be to lead
to a defensible standard(s).
Page 28:1 am not so certain that 'organic acids are common NATURAL sources of acidity in
surface waters." I would note that many VOCs do lead to secondary organic aerosols (SOAs) and
then to organic acids that are not "natural."
-------
Dr. Andrzej Bytnerowicz
General Comments
The document describes Problem Formulation, the first phase of the Ecological Risk Assessment
(ERA) for ambient NOx and SOx, and focuses on the last component of the Problem
Formulation Phase - the Analysis Plan. The document is logically structured and follows seven
clearly described steps. The plan presents interesting ideas for development of the secondary
standard for NOx and SOx. End points and conceptual models are also discussed but with much
less detail.
In the beginning of the documents a clear statement is made that for evaluating ecological
effects, total reactive N (both in the oxidized and reduced forms) will be considered. This is
because reduced N compounds may be responsible for many effects caused by oxidized N. This
is obvious and clear from the point of view of the present scientific knowledge. However, this is
a long stretch from NOx (NO and NO2) and SOx which are the criteria pollutants for which
secondary standards are being considered. I believe that legal considerations of such an
expansion of the NOx/SOx term should be discussed upfront. If not, it may be expected that
problems with various stakeholders during public discussion could take place.
Consideration for alternative secondary standards to ambient concentrations of NOx and SOx
should also be addressed. For N, all forms of reactive N could be considered as part of the
secondary standard. NOx (NO and NO2) is only a small portion of the problem, however,
theoretically it is possible to link the secondary pollutants (such as HNO3 or particulate NO3) to
the primary pollutants (NO and NO2). Ammonia is the key pollutant for which concentrations
and spatial and temporal distribution should be better characterized.
Considering high probability that relationships between NO2 concentrations (or reactive N; Nr
index) to N deposition, and between N deposition and its ecological effects may be weak, the
deposition-based indices such as critical loads (CL), should be considered as a potential
secondary standard. However, in order to consider such an approach a nation-wide monitoring of
n deposition, similar to the European ICP Forests and ICP Vegetation, should be in place. Based
on the existing monitoring efforts in the U.S., secondary standard for NOx and SOx based on N
deposition could be considered only for a small portion of the country.
Options such as "cap and trade", or emission reduction goals for specific administrative areas
(states, counties, air quality management districts) should be addresses and considered. What
was accomplished for the reductions of the SO2 emissions in the Midwest and Northeast based
on the requirements of Title IV of the CAAA or what is presently being done in Europe under
the auspices of the Convention on Long-range Transboundary Air Pollution (CLRTAP) could be
used as examples of the alternative approaches and possibilities (although such measures may be
politically difficult to be accomplished). However, a note of caution - successful reduction of the
SO2 emissions was relatively easy considering that SO2 sources are mainly large point emitters.
For NOx such possibilities could be difficult to accomplish because of the complexity of the N
chemistry and multiple sources of emissions.
10
-------
Compared to the main ISA document, transitional ecosystems are not mentioned. For
consistency this document should also treat them separately (especially since these ecosystems
are very sensitive to the N inputs).
Diagrams used in the document greatly help a reader to visualize the planned risk assessment
process envisioned by the authors. Figures 2-5, 3-1, or 3-5 are most useful.
A discussion of how the climatic conditions, or climate variability, could affect the proposed risk
assessment process is lacking. Meteorological variability is mentioned, but this is not sufficient.
More emphasis should be placed on how the other co-occurring stressors (tropospheric ozone,
drought, pests, diseases or catastrophic fires) affect N cycling and biological effects of N
deposition.
Are the CMAQ estimates of N deposition updated? If I recall the CMAQ simulation for the
western US, the latest model runs were done for 2002 (Gail Tonnessen, personal
communication).
It would greatly help in review of this document if the lines were numbered!
Specific comments
Page 1, first paragraph. A sentence "As discussed in the Integrated Review Plan (U.S. EPA,
2007a), this was done in recognition of the important linkages between ambient nitrogen and
sulfur leading to deposition of ambient particles that can have significant impacts on
environment". This is only partially true considering that particles have only very small effects
on the total N or S deposition. Gases or dissolved ions are much more important. Therefore, I
suggest deleting "of ambient particles".
Pages 7 through 9, Section 2.3.1 found this section confusing and I suppose that other readers
might have a similar experience. As I have stated in the beginning of my comments, this
document is supposed to represent the Problem Formulation Phase. As such I would expect that
the end points and conceptual models would get the same level of detail as the Analysis Plan.
Although the conceptual models are described in section 2. "Background and Overview",
specifically in section 2.3.1, a discussion of endpoints is not easily found. Sections 2.3.2.
"Overview of nitrogen deposition" and 2.3.3. "Overview sulfur oxides and sulfur deposition"
are too short to present a good discussion of the problem. I suggest skipping them completely
and referring a reader to the comprehensive review of these problems in the main ISA document.
Page 19, Table 2.1. The proposed schedule of the reviewing process is tight and very ambitious.
The CAS AC public review of the REA document in October is very early considering that the
draft will be released in August. In my particular case, a very active schedule of conferences in
August, September and October and summer vacation, does not leave enough time for a
thorough review. Therefore, I recommend moving the CAS AC review to November 2008.
11
-------
Page 27, Table 3-1. For aluminum mobilization, terrestrial ecosystems (T) should be added. In
the same table, for N concentrations, also terrestrial ecosystems are affected.
Page 31, 3.2, Step 2 (Define sensitive areas) and page 35, 3.3, Step 3 (Select risk/exposure cases
study assessment area within a sensitive area) - use of natural gradients of N and S deposition
should be considered.
Page 32, first two lines. I would argue that areas with good linkages to larger ecoregions should
be also considered for new case studies. Results from such area could be extrapolated by means
of geostatistics into much larger areas.
Page 35, last paragraph. I would like to suggest that "new N" could be called "reactive N
species", such as ammonia or nitric acid. "Old N" would be N stored in ecosystems as NO3-.
Page 38, first paragraph. Again - use of natural gradients for selection of the appropriate
assessment areas should be considered.
Page 38,1 would also suggest placing special emphasis on some areas in California, such as the
San Bernardino Mountains, coastal sage and desert ecosystems that have been intensively
studied and are sensitive to N additions.
Page 40, Table 3-6. Western states are under-represented. Please see my comment above.
Page 43, bullets. More appropriate would be using the term of "nitrate and ammonium
deposition". Ammonium would result from dry deposition of NH3, particulate NH4+ and water
dissolved NH4+ deposition.
Page 57, first paragraph. After establishment of the baseline conditions, more advanced
modeling techniques (dynamic models) would have to be used. Such models are able to take into
account time effects and may be useful for ecological risk assessment. During the last few years
such models have been used by the TCP Modeling and Maps in Europe.
The same paragraph. While particular loading "X" can cause a certain negative effect" Y", it
would be very difficult to translate it to ambient concentration of a pollutant (especially if NO
and NO2 are concerned). That effect "Y" would result from deposition of multiple N species. Is
CMAQ, or other models, able to specify what concentrations of individual pollutants are
responsible for the loading "X"? Maybe checking how the RAINS model used in Europe deals
with such issues is necessary at this step of the analysis planning?
Comments on bullets from Page 57
"What adverse effects are we trying to protect against?" Best approach would be to look at
multiple sensitive indicators (like sensitive lichens, C/N ratio, N foliar levels, mycorrhizae
colonization). Use of single indicator could be misleading. However, for practical reasons and
large-scale estimates, a single, easy to determine indicator (such as NO3- in soil solution) should
be considered as a good initial step.
12
-------
Next bullet - information on contribution of N oxidized and reduced forms to total N pollution
(reactive N) effects is still poor. More intensive field campaigns and long term monitoring using
some new techniques is urgently needed. A key problem is a need for better understanding of
ammonia concentrations in the evaluated ecosystems.
Effects do occur due to different forms of N. Effects of gaseous N species (NO, NO2, NH3 or
HNO3) are very different among each other, and quite different from the effects of particulate or
water-dissolved NO3- orNH4+.
How should alternative levels be selected? Combination of various methodologies is needed
(controlled experiments, field exposures including natural gradients, others).
What are the correct temporal and spatial scales for the ambient air indicator? Ideally hourly
values should be available. However, considering a lack of proper monitoring equipment that
could be used at large scales, the weekly, bi-weekly or even monthly averages of the key
deposition drivers would greatly help our abilities to model deposition. In such cases, passive
samplers for gaseous pollutants could be used. Information on particulate component of dry
deposition is poor, and its contribution to the total deposition may greatly vary in space and time.
Spatial coverage would depend on the chemical and topographic complexity of the area of
interest. There are new geostatistical methodologies available that may help in optimizing
monitoring networks and adequate spatial coverage.
Large-scale monitoring of throughfall could be used for integrative measure of wet and dry N
deposition. However, use of the throughfall analysis methodology in arid and semi-arid
ecosystems is questionable.
Comments on using 9 steps to determine the appropriate level and a form of a standard:
Generally this is a logical and well defined plan.
Point 1 - portion of dry deposited N, both in gaseous and aerosol forms, is not well
characterized.
Point 2 - OK.
Point 3 - here, contribution of the main contributors to dry deposition of N, ammonia and nitric
acid have to be better characterized.
Point 4 - the most sensitive indicators, and sets of various indicators, have to be considered.
Although I generally agree with a need for determining ecosystem services and bundling them in
order to find comparisons between locations, I am not sure if that would not complicate the
establishment of a link between effects, deposition, ambient concentrations and emissions.
Point 5 - OK.
13
-------
Point 6-1 suggest considering collaboration with the European scientists and modelers
experienced in using RAINS model. Impacts of meteorological (climatic) variability are of a
very high importance.
Point 7 - critical loads concept would be very useful.
Point 8 - intuitively I agree. However, I have to get acquainted better with the available literature
to say more.
Point 9 - OK.
Responses to the assigned questions:
Question 8. In the seven-step approach to the current conditions risk/exposure assessment, Step 5
(Section 3.5) discusses how to scale up case study areas to more spatially extensive sensitive
areas, where appropriate. Does the Panel agree with this approach or can they suggest
alternatives?
The proposed approach seems to make sense as long as the extrapolations are done within the
same ecoregions. However, distribution of concentrations of the key drivers of N dry deposition,
specifically nitric acid vapor and ammonia, may dramatically change between the urban and
agricultural sources and the receptor areas due to their high reactivity and deposition to
vegetation and other surfaces. Therefore the on-ground determinations of patterns of their
distribution are recommended. Passive samplers offer a relatively simple and reliable method for
determining pollution concentrations. Results of such monitoring efforts aided by geostatistics
may be used for generation of concentration surfaces of these pollutants. For estimates of dry
deposition of these gases inferential methods could be used at large geographic scales. In
addition, complexity of soils characteristics have to be also taken into account in proposed
scaling-up exercises.
Question 9. In the seven-step approach to the current conditions risk/exposure assessment, Step 6
(Section 3.6) outlines a path to assess the current conditions of sensitive ecosystems. How well
does the Panel agree with the approach outlined for calculating response curves and utilizing
mapping and ecosystem services to characterize current conditions or can the Panel recommend
alternative approaches?
This is a valid approach. Response curves from controlled experiments or natural gradient
studies encompassing wide ranges of deposition levels would provide databases needed for
estimates of current conditions and models simulating future changes. Critical loads (CL)
calculations (performed with steady state, empirical or dynamic models) have been used in the
NE USA and Canada for many years. In Europe, CL have also been successfully measured on
the ICP Forests and ICP Vegetation networks. The ICP Modeling and Maps converts that
information to maps representing annual levels of CL and exceedances for N and S deposition
and acidity. The European Regional Air Pollution Information and Simulation (RAINS), uses
those results for controlling of the internationally agreed quotas of emissions for individual
countries. A similar approach could be tested and possibly used in the U.S. as well. Instead of
14
-------
individual countries (as in Europe), states or air quality management districts could be used as
the administrative units of control.
a. How well does the Panel agree with using ecosystem services to provide a common metric
for comparing ecological risks due to nitrogen and sulfur deposition effects?
This could be very useful approach for integrating effects of deposition of N and S on
ecosystems. By presenting examples of what services are affected, potential stakeholders would
be able to better comprehend implications of the adverse effects of elevated N and S deposition.
That could bring more attention and potential support from the general public, NGOs and other
stakeholders which would help in development of the secondary, welfare effects-based standard.
b. How well does the Panel agree with collecting current valuation studies to understand the
value of bundled ecosystem services? Can the Panel recommend additional or alternative
approaches?
I generally agree with this approach. A holistic approach to the variously affected services seems
to be logical. However, it should be evaluated if these new complex attributes of changes would
not complicate the entire process of secondary standards development. Maybe as the first step
towards development of a new standard the well tested European approach (CL and RAINS
combined) should be initially considered. The idea of ecosystems services (including their
bundling) could be considered as the next step leading to further improvement of the standards.
Question 10. In the seven-step approach to the current conditions risk/exposure assessment, Step
7 (Section 3.7) describes an approach to assess degrees of protection/levels of effects under
alternative forms and levels of ambient NOx and SOx standards. This approach attempts to
describe how the methods, models, and results of the current conditions risk/exposure assessment
can inform our evaluation of the appropriate form(s) and levels(s) of a national standard. How
well does the Panel agree with the approach outlined in this section, the issues presented, and the
9 steps outlined to assess potential forms and levels of the standard? Please suggest any
additional or alternative steps we should take into consideration?
The section presents an interesting plan for defining and describing ecological changes caused by
N and S deposition and an attempt to associate them with the NOx and SOx concentration-based
standards. Although not easy to implement on a national scale, the critical loads (CL)
methodologies could be used for defining levels of the acceptable deposition of N, S and acidity.
Maps of CL exceedances, as it has been done for the NE United States, eastern Canada and
Europe, can be very useful in showing the areas where the unacceptable deposition has taken
place. Such information aided by appropriate models could lead to recommendations of where
the N deposition reductions are needed in order to prevent deterioration of sensitive ecosystems.
While it may be possible to use such information for recommending emission reductions, as it is
being done in Europe under the Convention on Long-Range Transport of Air Pollution (please
see my remarks above), a direct link to ambient NOx or SO2 concentrations would be extremely
difficult, if not impossible, to establish.
15
-------
Plan presented in Section 3.7 establishes a framework of the logically aligned steps (activities)
that theoretically could lead to development of new, concentration-based standards for NOx and
SOx.
16
-------
Ms. Lauraine Chestnut
One general comment: There will undoubtedly be limitations in data for this assessment. I want
to recommend that there be an effort throughout the process to define data that will improve this
assessment the next time around. Collection of some of these data may be ongoing, but there
may not yet be enough to answer the quantitative questions: the need for these monitoring and
data collection efforts to have continued support should be articulated. There may be further
monitoring and data collection that would be helpful. These need to be defined and prioritized so
that resources are put to their best use and support is generated for more monitoring and data
collection with focus on the specific information needed to help answer the key policy questions.
Section 2.4
1. It seems really important for this assessment to address the question of whether both S and N
have to be reduced to obtain benefits, or are they causing separate effects that are essentially
additive? This speaks to whether the standards for the two need to be linked.
2. On page 18,1 don't understand the question about whether long-term effects and cumulative
loadings are relevant to the NAAQS review. If they are important for understanding the effects
on ecosystems, why would they not be relevant?
Charge question 4:
The need to define and describe the nature of the harm to the ecological resource that might be
considered an adverse effect needs to be considered right at the beginning when the indicators
are reviewed and selected. It is important that the selected indicators be able to be linked to
changes in deposition, but also to changes in ecosystem services. There may be a threshold or a
continuous relationship between levels of the indicators and harmful effects on ecosystem
services (or even a nonlinear relationship that entails positive effects over some ranges and
negative effects over other ranges). Either way, it is important to ask whether a given change in
the selected indicator can be interpreted as showing a harmful effect, the implications of which
can be articulated for the policy maker. For example, do we know at what level for each
indicator the ecosystem looses capacity to support some species that would normally be expected
to live in this system? It will be very important to be able to explain why the change in the
indicator matters and what the significance of the change is in terms of ecosystem functions and
services.
Charge question 9b:
Once the losses or impairments in ecosystem functions and services at current deposition rates
are quantified (or otherwise characterized) then it will be useful to search for available valuation
studies that provide relevant information. It does not seem useful to try to determine total values
for all ecosystem services even in the selected case study areas, because the policy relevant
valuation issue for these pollutants is probably not a total loss in services, but a marginal change
in quality or quantity of some services. To the extent that the literature supports valuation of
bundled ecosystem services it makes sense to use this approach, especially if these services are
17
-------
perceived as connected or joint products of the ecosystem functions, and if there are multiple
services affected by the same change in pollution exposure.
Charge question 11:
I think EPA has selected the right categories of effects to focus the quantitative risk assessment
efforts. This list of other effects needs to be included and described. In some cases the
assessment might provide partial quantitative information. For example, will it be possible to say
what the expected changes in ambient concentrations of N2O might be expected as a result of
alternative standards? I agree it is not appropriate to expend resources here to try to quantify the
associated climate change effects.
18
-------
Dr. Ellis B. Cowling
Individual Comments on the Draft Scope and Methods Plan for the Risk/Exposure
Assessment: Secondary NAAQS Review for Oxides of Nitrogen and Oxides of Sulfur
Before plunging directly into my specific assignments to critically review the "Key
Policy Relevant Questions" in Section2.4 and answer Charge Questions 3-7 in preparation for
the April 3, 2008 CASAC Consultation, let me emphasize how pleased I was to see in this Draft
Scope and Methods Plan how much EPA appears to be progressing in its willingness to think in
a more holistic way about:
1) Multiple pollutants and multiple effects of airborne pollutants that often occur in the
same air parcel and frequently act together to cause similar, different, and/or sometimes
more pronounced effects than single pollutants acting alone.
2) Broadening the "Indicator of Concern" for the secondary NAAQS for "Oxides of
Nitrogen" to include chemically reduced and organic forms as well as oxidized forms of
"total reactive nitrogen,"
3) Recognizing that NAAQS designed to protect ecosystems from adverse effects of air
pollutants often require different pollutant indicators, levels, averaging times, and
statistical forms than NAAQS designed to protect human health.
4) The multiple similarities and advantages rather than just the differences among
"NAAQS," "Critical Loads" and "Target Loads" for nitrogen and sulfur air pollutants.
5) How "the words we choose often reveal the quality (and sometimes also the limitations)
in our understanding" about various phenomena in nature. Useful examples in the
current Draft Plan include the significance of such words as "ecosystem services;"
"acidification," "nutrient enrichment," "nitrogen saturation," "NOx/SOx,"
"NOx/NHx/SOx," "total reactive nitrogen," "acid precipitation," "acidifying deposition,"
"wet plus dry acidifying deposition," Etc.
Please note a few additional comments below regarding some of these important matters and
especially the Resolution displayed on page 3 of these Individual Comments after it was passed
by the Integrated Nitrogen Committee of the EPA Science Advisory Board for Communication
on October 31, 2007 to our CASAC NAAQS Review Panel on the Secondary Standards for
Oxides of Nitrogen and Oxides of Sulfur.
19
-------
My Specific Assignments in this CASAC Consultation on the Draft Scope and Methods
Plan for the Risk/Exposure Assessment Secondary NAAQS Review for
Nitrogen Dioxide and Sulfur Dioxide
My specific assignments in preparation for our April 3, 2008 CASAC Consultation on the
Draft Scope and Methods Plan as outlined in Chairman Ted Russell's memo of 11 March 2008
are:
1) Key Policy Relevant Issues. These topics are summarized in Section 2.4 in the Draft
Scope and Methods Plan, and
2) Charge questions 3-7 - Ecological Effects Characterization. These Charge Questions are
listed in Lydia Wegman's memo to Kyndall Barry dated 5 March 2008.
Chairman Russell also gave the first of these two assignments to six other CASAC panel
colleagues - Douglas Crawford-Brown, Paul Hanson, Naresh Kumar, Myron Mitchell, Kathleen
Weathers and Lauraine Chestnut. He also gave the second assignment to three other CASAC
Panel members - Charles Driscoll, Myron Mitchell, and David Shaw. Thus, I am very much
looking forward to comparing notes with all of these other people during our Consultation on
April 3rd!
Assignment 1) ~ Critical Review of section 2.4:"Key Policy Relevant Questions"
Both the current Chair of CASAC, Dr. Rogene Henderson, and others among the Statutory
Members of our Clean Air Science Advisory Committee have written in recent months about the
Need for Policy Relevancy as the Dominant Concern in the New NAAQS Review Processes.
In these carefully structured review processes, the most critical and overarching policy
question that needs to be answered is:
"What scientific evidence and/or scientific insights have been developed since the last
review to indicate if the current public-health based and/or the current public-welfare
based NAAQS need to be revised or if alternative indicators, levels, statistical forms, or
averaging times of these standards are needed to protect public health with an adequate
margin of safety and to protect public welfare?"
And similarly, especially with regard to the present integrated [simultaneous] review of the
Secondary Standards for NOx and SOx, the most critical and overarching Key Policy Question
is:
"What scientific evidence and/or scientific insights have been developed since the last
review to indicate whether, and if so, what particular ecosystem components or other air-
quality-related public welfare values, are more or less sensitive than the populations of
humans for which primary standards are established and for this reason may require a
different indicator, level, statistical form, or averaging time of a secondary standard in
order to protect public welfare."
20
-------
With the latter overarching Key Policy Question in mind, we now have the considerable
advantage that a much more complete focus can be achieved on critically important scientific
findings that are as directly relevant as possible to this overarching Key Policy Question. To
achieve this more complete focus, it is desirable that each chapter of the Integrated Science
Assessment, the Risk/Exposure Assessment, and the Policy Assessment/Rule Making
documents for each criteria pollutant should contain a summary section composed almost
entirely of very carefully crafted summary statements of conclusions and/or scientific
findings. These carefully crafted summary statements should:
1) Contain the distilled essence of the most important topics covered in each chapter,
and
2) Be as directly relevant as possible to the overarching Key Policy Question written
above.
Section 2.4 of the present Draft Scope and Methods Plan contains a series of 17 policy
relevant questions to which carefully crafted scientific answers could be developed. But these
mostly policy relevant questions need to be better organized and coordinated so that the answers
to these questions can be used to build a coherent series of linkages between the major
acidification and nutrient enrichment effects of concern and the nitrogen and sulfur
secondary NAAQS standards that will be necessary to avoid these ecological effects.
My most serious reservation about the 17 policy relevant questions listed on pages 17 and 18
of Section 2.4 is that chemically reduced forms of nitrogen (NHx), organic forms of nitrogen
(NCx), and total reactive nitrogen (TNr) all are not included in either the specific wording or the
written context of any of these 17 questions. Only the first sub-part of the sixth question listed
on page 17 deals explicitly with ammonia, ammonium ion, and/or total reactive nitrogen as
alternative "indicators" of nitrogen pollution.
"Should the current standards for NC>2 (as an indicator for NOx) and 862 (as an indicator for
SOx) be retained, revised, or revoked and/or replaced with alternative standard(s) having
different indicators to provide the required protection from known or anticipated adverse
welfare effects?"
I hope that these 17 policy relevant questions can be adjusted, revised, reorganized, and/or
augmented in order to further increase their value in optimizing the quality of summary
statements of scientific findings that will be as relevant as possible to the overarching Key
Policy Question printed above in bold italic type.
In this connection, please also note once again the attached
Resolution from the Integrated Nitrogen Committee of the Science Advisory Board
for Consideration by the CASAC Secondary NAAQS NOx and SOx Review Panel
During the ongoing meeting of the EPA Science Advisory Board's Integrated Nitrogen
Committee (INC) -- meeting at SAB Headquarters in Washington DC on October 29-31, 2007 -
the several members and Chair of the INC, Dr. James Galloway of the University of Virginia,
21
-------
asked me (as the CASAC-designated liaison person to the Science Advisory Board's Integrated
Nitrogen Committee) to present the following Resolution (which was developed and approved
by the INC) for consideration during the CAS AC review of the NAAQS for NOx and SOx
during our CASAC Conference Call Consultation on October 30, 2007.
Resolution
The current air pollution indicator for oxides of nitrogen, NOx, is an
inadequate measure of reactive nitrogen in the atmospheric environment.
The SAB's Integrated Nitrogen Committee recommends that inorganic
reduced nitrogen (ammonia plus ammonium) and total oxidized nitrogen,
NOy, be monitored as indicators of total chemically reactive nitrogen.
Assignment 2 - Charge Questions from OAQPS
My comments are organized below in response to each of the several Charge Questions
posed in Lydia Wegman's memo to Kyndall Barry dated 8 March 2008. As you will see much
more detailed attention has been given to Charge Questions 3-7, to which Chairman Ted Russell
asked me to give special attention.
1. In outlining the scope of this risk/exposure assessment, we have created a flow diagram that
represents how nitrogen and sulfur compounds move from 'source to dose' in the
environment (see Figure 2-1). How adequately does this conceptual model for evaluating
risks due to deposition-related ecological effects characterize what should be covered in the
scope of this assessment?
This conceptual model seems very adequate as an initial frame of reference for
deposition-related ecological risks
2. The main ecosystem effects areas we anticipate evaluating in this risk/exposure assessment
are:
(1) risks to terrestrial ecosystems from nitrogen enrichment effects,
(2) risks to aquatic ecosystems from nitrogen enrichment effects (eutrophication),
(3) risks to terrestrial ecosystems from acidification effects (nitrogen and sulfur), and
(4) risks to aquatic ecosystems from acidification effects (nitrogen and sulfur).
We also plan to qualitatively discuss the role of sulfur enrichment on methylmercury
production and the role of nitrous oxide in climate change. What key effects areas, if any,
have been overlooked by this approach? Should the assessment plan be modified to include
other effects?
Especially considering the court ordered time constraints under which this Risk/Exposure
Assessment must be completed, it seems very reasonable to concentrate on these the four
major aspects nutrient enrichment and acidification in terrestrial and aquatic
ecosystems but to also include some assessment of the other two effects of sulfur
enrichment on methylmercury production and N20 influences on climate change.
22
-------
3. Due to the complexity of conducting a nationwide risk/exposure assessment for welfare
effects due to NOx and SOx, we have outlined a strategy designed to identify sensitive
ecosystems and a range of harmful/adverse effects (see Figure 3-1). The seven steps are to
(1) identify documented biological, chemical and ecological effects and potential ecosystem
services,
(2) define sensitive areas using GIS mapping,
(3) select risk/exposure case study assessment areas,
(4) evaluate current loads and effects in case study assessment areas,
(5) scale up the case study assessment areas to larger sensitive areas where feasible,
(6) assess current ecological conditions in those areas, and
(7) assess alternative levels of protection under different ambient scenarios.
Does the Panel agree with this general approach? Should it be improved or modified?
Yes, I agree that this general approach is sound, But it certainly can be improved and modified in
its presentation and implementation within the Risk/Exposure Assessment for which this is a
Draft Plan.
As indicated on page 2 of these Individual Comments, my major reservation about the general
approach outlined in this Draft Scope and Methods Plan is that it remains unclear if EPA is really
willing and interested to include reduced (especially ammonia and ammonium) and organic forms
of nitrogen (amino acids and volatile amines) as well as oxidized forms of nitrogen in the
crucially important "indicators" being considered in this integrated (simultaneous) review of the
present "NOx" and "SOx" NAAQS secondary standards.
It was reassuring to read in several parts of this Draft Scope and Methods Plan: (for example) on
page 22 that:
"The risk exposure assessment will focus on ecosystem welfare effects that result from
the deposition of total reactive nitrogen " and also on page 26 that:
"To assess the impacts of total reactive nitrogen and sulfur loading we plan to identify
adverse terrestrial and aquatic effects ..."
But it was also discouraging to see that the schematic diagram chosen for inclusion on page 12 of
this Draft Scope and Methods Plan deals only with oxidized forms of nitrogen and includes no
mention or pictorial illustrations of the huge (mostly agricultural) sources of reduced and organic
forms of reactive nitrogen.
If EPA is serious about dealing with both "nutrient enrichment" and "acidification" of ecosystems
induced by atmospheric deposition of total reactive nitrogen, the agency will have to learn how to
sustain its recent success in learning to think more holistically (in part by learning how to write
more holistically) about ammonia, ammonium ion, NHx, and NCx, and TNr about "ecosystem
services" "critical loads," and "multiple pollutant/multiple effects" approaches in air quality
management.
What a different impression this Draft Scope and Methods Plan would have created if:
the more inclusive abbreviation "NOx, NHx, and SOx) had been used instead of just "NOx
and SOx," or
the words total reactive nitrogen" and "total nitrogen loading" were used more frequently and
more appropriately:
23
-------
In formulating the Key Policy Relevant Questions listed on pages 17 and 18,
In the captions of the tables and figures on pages 12, 16, 19, 28, and 29, as well as
In preparing the major subheadings and topic sentences of paragraphs on pages 7, 15, 16,
19, 21, 22, 27, 49,55,56, 59, and A-3.
4. In the seven-step approach to the current conditions risk/exposure assessment, Step 1 (Section
3.1) describes an approach to identify the documented effects, biological, chemical and
ecological indicators, and potential ecosystem services related to acidification and nutrient
enrichment. Does the Panel agree with this approach or can they suggest alternative
approaches we should consider?
I think you did very well in devising the seven step approach presented in Figure 3.1 the
first step of which is described more fully in the text in Section 3.1.1 on pages 26-30. Thus
I encourage your continuing use of this first step approach so long as you keep in mind that
reduced and organic forms of nitrogen may be even more important than oxidized forms of
nitrogen in many of the sensitive areas to which this approach is intended to be applied.
5. In the seven-step approach to the current conditions risk/exposure assessment, Step 2 (Section
3.2) outlines a path to define areas sensitive to total reactive nitrogen and sulfur inputs. Do the
Panel members agree with this approach or are there better alternatives that should be
considered?
Once again, I think your step 2 approach makes sense. I was especially pleased to see that your
intention is to use GIS mapping as a tool to further analyze and interpret the strengths and limitations
of similarities and differences among these selected sensitive areas.
a. We are attempting to characterize the risks to ecosystems from sulfur and nitrogen
deposition nationwide by clustering sensitive ecosystems where possible and by using the
linkages between these areas at different scales. Please comment on the adequacy of this
approach.
I have only very limited personal experience in using these kinds of clusterting analytical tools, thus I
am not well qualified to assess the adequacy of these methods of analysis.
b. How appropriate are the datasets and GIS maps listed in Table 3-4 for identifying ecosystems
sensitive to nitrogen and sulfur and/or are there others that have been overlooked?
The array of general types of ecosystems listed in Table 3-4 seems very adequate to me except for the
lack of near-coastal oceans waters.
6. In the seven-step approach to the current conditions risk/exposure assessment, Step 3 (Section
3.3) outlines a path to identifying risk/exposure case study assessment areas, a. Table 3-5
provides an initial list of indicators, mapping layers and multimedia models that may be used to
assess ecosystem risk and exposure.
a. Please comment on the appropriateness of these and suggest alternatives that may be better
suited for this analysis.
Over the past 20 years, I have had some interactions with the principal investigators involved in about
half of these studies of ecosystem effects, possible indicators, and the mapping layers and modules
used in these kinds of terrestrial and aquatic problem areas. From these interactions, I have the
general impression that these problem areas are appropriate for use in Step 3 and I have no
suggestions of other methods that would be more suitable for this kind of analysis.
24
-------
b. Please comment on the list of potential case study assessment areas in Table 3-6 and Table
3-7 and make recommendations or suggest any alternatives.
As in my response to the issues raised in Charge Question 6a, I have had some interactions with the
principal investigators involved in about half of these studies and believe that the array of possible
case study areas listed in Tables 3-6 and 3-7 is reasonable and is likely to be fruitful for the purposes
of this Risk/Exposure Assessment and evaluation of benefits that are likely to occur from decreases in
atmospheric loading of total reactive nitrogen. My experience is much more limited with respect to
sulfur and mercury loading.
7. In the seven-step approach to the current conditions risk/exposure assessment, Step 4 (Section
3.4) outlines a path to assess current nitrogen and sulfur loads and their effects on case study
assessment areas. Does the Panel agree with how we have described our approach to identifying
datasets, gaps, and uncertainties?
The task of identifying useful data sets is challenging but the associate tasks of identifying gaps and
characterizing uncertainties is even more complicated and challenging The approach you have
planned to use seems very sensible. I am also particularly pleased that you plan to stretch even
further and attempt to characterize uncertainties in the context of current understanding of total
reactive nitrogen impacts on ecosystem services rather than the more familiar measures of ecosystem
growth, stability, and productivity.
a. We have initially identified the primary chemical indicator that is most suitable for assessing
ecosystem acidification effects as acid neutralizing capacity (ANC), with alternatives depending
on data availability (see section 3.4.1 and Appendix B). Does the Panel agree with this selection,
or can they suggest alternative/additional key indicators?
I presume that you mean that atmospheric-deposition induced losses in ANC" rather than just ANC
itself is "the most suitable chemical indicator for assessing ecosystem acidification effects. With this
small adjustment in the phrasing of your question, I agree that ANC is probably the indicator of
choice for assessing acidification effects in aquatic ecosystems. I know of no indicator that would be
more suitable although maybe loss of cation exchange capacity might be worth some consideration.
Charles Driscoll will be better able to answer this question more authoritatively than I am able to do.
b. We have described the models being considered for this analysis (see section 3.4.2 and
Appendix C). Does the Panel agree with the choice of these models, and can they help prioritize
them for modeling the responses of the indicators recommended in Step 1 (Section 3.1)?
I have no experience on which to base an informed judgment about the comparative merits of these
models.
8. In the seven-step approach to the current conditions risk/exposure assessment, Step 5 (Section
3.5) discusses how to scale up case study areas to more spatially extensive sensitive areas, where
appropriate. Does the Panel agree with this approach or can they suggest alternatives?
I have no experience on which to base an informed judgment in response to this question.
9. In the seven-step approach to the current conditions risk/exposure assessment, Step 6 (Section
3.6) outlines a path to assess the current conditions of sensitive ecosystems. How well does the
Panel agree with the approach outlined for calculating response curves and utilizing mapping
and ecosystem services to characterize current conditions or can the Panel recommend
alternative approaches?
I agree that the approach outlined in Section 3.6 is appropriate for your assessment of the current
condition of ecosystems. As indicated in my response to Charge Question 7 (above) I am very pleased
25
-------
that you are planning to base your assessment on the magnitude of atmospheric deposition impacts on
ecosystem services.
a. How well does the Panel agree with using ecosystem services to provide a common metric
for comparing ecological risks due to nitrogen and sulfur deposition effects?
Once again, I am very pleased that you are planning to use ecosystem services as the base for
your assessment of comparative risk due to total reactive nitrogen and sulfur deposition effects.
b. How well does the Panel agree with collecting current valuation studies to understand the
value of bundled ecosystem services? Can the Panel recommend additional or alternative
approaches?
I am not sure what is meant by "bundled ecosystem services" thus I have no basis on which to
offer an informed judgment in answer to this question,
10. In the seven-step approach to the current conditions risk/exposure assessment, Step 7 (Section
3.7) describes an approach to assess degrees of protection/levels of effects under alternative
forms and levels of ambient NOx and SOx standards. This approach attempts to describe how
the methods, models, and results of the current conditions risk/exposure assessment can inform
our evaluation of the appropriate form(s) and level(s) of a national standard. How well does the
Panel agree with the approach outlined in this section, the issues presented, and the 9 steps
outlined to assess potential forms and levels of the standard? Please suggest any additional or
alternative steps we should take into consideration.
I am very pleased to see that you plan to do what appears to me to be a very thorough and effective
analysis of alternative forms and levels of ambient "NOx and SOx" standards (by which I presume
you also mean alternative "NOx, NHx, NCx" or "total reactive nitrogen" standards" ).
The nine step approach you have outlined seems very sensible to me although I could not figure out
how step 1 and 3 are different from each other, I was also puzzled about why you included the
"levels and forms" of alternative standards and not also the associated "averaging time and statistical
form" of alternative standards. Maybe the latter two aspects are so much less important than the first
two that the latter two are not essential for the purpose of this analysis.
11. Additional ecological/welfare effects due to NOx and SOx emissions that we do not currently
anticipate evaluating in detail in this review include the following:
Nitrogen saturation,
Maple decline,
Ammonia air deposition and toxicity to native mussels,
Relationships between acidity/nutrient enrichment and mercury methylation,
Sensitive areas for acidity/nutrient enrichment impacts, identified from biogeochemical
characteristics, and
Climate change effects due to N2O.
Does the Panel agree that these represent lower priority effects for the current assessment? If
not, what does the Panel recommend?
I agree that all of these effects are of lesser importance than the four main topics that are planned for
inclusion - nutrient enrichment and acidification effects on both terrestrial and aquatic ecosystems.
26
-------
But would encourage some reconsideration of both "nitrogen saturation" ~ which is hard to avoid
since it is the "limiting case" when you are already dealing with nitrogen enrichment effects.
I feel similarly about keeping climate change effects caused by N2O emissions in the present assessment
plan mainly because climate change effects are so likely to become very much more important in the
future of our country's concerns about human induced changes in the chemical and physical climates of
our planet.
27
-------
Dr. Douglas Crawford-Brown
This review focuses primarily on the Risk Assessment Framework presented, both in the
Overview sections and in the step-by-step procedures described in Chapter 3 and in the
appendices. The goal is to ensure both that the risk assessment goals are appropriate for standard-
setting, and that the methods to be used will provide the kinds of information, of sufficient
quality, to meet those goals.
The authors have focused appropriately on acidification and nutrient enrichment in the general
framework for deposition-related effects from both S and N oxides, as these are the effects for
which the scientific database is sufficiently sound. The decision to include perturbations on
methylmercury also appears sound. My primary concern on all of these measures is whether
loading, rather than subsequent concentrations, will be sufficient to accurately determine effects
levels. I realize the data are primarily in the form of loadings, and so this may be unavoidable.
But is still seems to me that there can be large variations in the relationship between loading and
concentration due to differences in hydrological characteristics and chemical reaction rates (e.g.
buffering) between water systems. But so long as the resulting variability and uncertainty can be
characterized, this should not prove a large problem for establishing a NAAQS.
With respect to N2O as a greenhouse gas, I appreciate the attempt to place oxides of nitrogen
into a framework that includes consideration of climate change, but can find little in the
document to support the idea that it will prove possible to relate emissions to either HGH
forcings or resulting changes in climate or ecosystem measures. This is an interesting idea that
probably is not yet ready for NAAQS. It was wise, therefore, to leave the discussion qualitative
on this point.
I remain unclear as to the role of the case studies. The decision to focus on especially sensitive
ecosystems is a good one so long as it can be assumed that they are representative of a
reasonably large class of such sensitive sites on the national scale (I don't think it would be
appropriate to base national standards on situations at 1 or 2 outlier sites). So, the question that
will naturally arise is what these case study sites are taken to represent in a national distribution.
It would be good if some indication could be given of the number of such places in the US that
might be equally sensitive and equally exposed (or some combination of sensitivity and exposure
that makes them experience impacts at least as large as the case studies).
I realize that much of the work will depend on what is found in preliminary assessments of the
data, but I was not at all clear as to how the spatial resolution was to be treated. Where there is
high spatial variability, the exposure side of the exposure-response relationship can be greatly
affected by choice of location. The authors do mention that they will explore this issue with
examination of spatial gradients, but don't provide much detail as to how these gradients will be
quantified to determine the appropriate grid size for atmospheric concentration, loadings and
effects. Care will need to be taken in this aspect of the assessment.
28
-------
The authors mention at one point that they will search for ways to "consolidate effects" effects
across acidification and nutrient enrichment. It would be nice to have such a summary measure,
but I doubt it will be possible to accomplish this goal. Instead, one will be left with a variety of
what are essentially hazard quotients with no way to combine these into a summary hazard
index. This won't prevent setting a secondary NAAQS, however.
Figure 3-1 is a good summary of the steps to be taken. I don't, however, know what Step 5
means (I understand it conceptually, but don't know how it will be accomplished other than
through pure judgment).
I am supportive of Step 1 in general, although the authors will need to think carefully about how
they will ensure that decisions are based on adverse effects and not simply effects. As written,
the document makes it sound as if any change in any indicator is adverse, an assumption with
which I disagree. That can in part be dealt with by relating indicator changes to alterations in
ecosystem services, although (i) the data to support such relationships are lacking and (ii) it still
will be important to define what is meant by a significant decline (and not simply a change) in an
ecosystem service).
While the use of GIS to examine the issue of ecosystem sensitivity may prove useful, I see
nothing here that convinces me such a GIS-based study can provide fundamental insights into the
causes of sensitivity. And I don't fully understand why the GIS studies are needed. I would
assume the causes of sensitivity are elucidated in the scientific literature, negating the need for a
scientific study as part of the NAAQS process. The mapping layers selected appear appropriate,
but again I can't see how they will be overlain and assessed to either identify sensitive areas that
can be used as case studies or to elucidate conditions that lead to sensitivity. But perhaps the
authors will find a way to perform a useful assessment once they data are collected, visualized
and analyzed spatially.
At one point, on Page 38, the authors comment that one wants to select a region where
atmospheric effects dominate (at least, this is what I take them to mean). That is good for a
scientific study of the effects of deposition, but a region can be sensitive to deposition precisely
because there are other sources of nutrients that push the system close to some natural limits. So
I am not sure that this criterion would be appropriate for selecting case studies.
The candidate assessment areas starting on Page 39 represent a good cross-section of potential
case study areas. They will need to be narrowed to make the study feasible.
I have little expertise on modeling deposition, and so will not comment on that aspect. I note,
however, that many of the models go well beyond modeling deposition, looking at speciation,
concentration in water, etc. I was never sure which aspects of a particular model the authors were
proposing to use, given the early parts of the document suggested that loading would be the
exposure measure in exposure-response relationships.
Step 5 remains very "black box" to me. I think the idea is good, but can't see the methodology
that is being proposed.
29
-------
Step 6 appears to be a form of hazard quotient, or margin of exposure approach. I support this in
general as it is the best that can be done.
The natural services valuation will be tricky. There are significant conceptual problems with
approaches that begin to mimic Costanza-like valuations, in large part because there is no real
market in which these services are traded. So the values are hypothetical, in the sense of a kind
of revealed preference. I suspect there will be a lot of disagreement on this aspect of the
assessment.
If the framework in the first 6 steps can be worked out, then Step 7 should be relatively straight-
forward and I support the approach defined here.
30
-------
Dr. Charles T. Driscoll
Comments on Draft Scope and Methods Plan for Risk/Exposure Assessment:
Secondary NAAQS for Oxides of Nitrogen and Oxides of Sulfur
General and specific comments:
Page 10 Does figure 2-1 imply that species alterations, eutrophication and mercury
methylation are by-products of nutrient enrichment or are they relevant to both
disturbances? Is there some significance to the fact that these are sub-bullets
under nutrient enrichment? Please clarify.
Page 14 2.3.4. Isn't nutrient enrichment (terrestrial) and eutrophi cation (aquatic) really the
same thing? Why give these separate designations? This seems to confuse the
issue.
Page 15 Last paragraph. Why are sulfur-mercury interactions exacerbated in coastal
waters more than other ecosytems? Is there any scientific basis for this
statement?
Page 16 Figure 2-5. Would it be useful to separate freshwater and coastal systems for
aquatic effects? These are very different systems.
Page 16 There is limited scientific evidence that increased N(V leaching inhibits
methylmercury production.
Page 17 Bullet 4. This bullet is very focused compared to the other bullets. Could it be
modified to make it more consistent with the other bullets? Something like
"What is total sulfur and nitrogen deposition to ecosystems and how do values
vary across the landscape?"
Page 17 Bullet 6, sub bullet 2. I'm not sure I understand the context of this question:
Does it refer to whether the effects of atmospheric nitrate deposition can be
distinguished from effects associated with other inputs of nitrogen? This is a
strange question. Can it be made more explicit?
Page 27 Some of the indicators on the table don't make sense. What is aluminum or metal
mobilization? How do you quantify or characterize forest health? Why are ANC
and alkalinity both listed? Under nutrient enrichment should SAV density or
hypolimnetic DO be listed? Should methyl or total mercury be included?
Page 28 For acidification there should be a primary indicator of soil calcium status, maybe
soil % BS, in addition to ANC which is indicative of surface water.
31
-------
Page 28 Why isn't deposition an indicator for acidification like nutrient enrichment?
Page 28 Third line. Should this be carbon to nitrogen ratio?
Page 28 Table 3-3. SAV area should be included.
Page 45 Again, it is critical to have both a soil and surface water indicator for
acidification. For example in the Northeast surface water ANC is increasing
while soils are continuing to lose exchangeable Ca2+ even under current
decreasing deposition. Because soil and surface waters are not completely
coupled, two indicators should be used: maybe ANC and soil % BS.
Page 47 If CMAQ is to be used to make deposition projections, it is critical that an effort
be made to validate simulations of total and dry deposition. Models like MAGIC
or PnET-BGC require historical (dating back to 1850) and future values of total
deposition. Will CMAQ be able to do this? A data fusion exercise will be
needed. It would be helpful to evaluate the compatibility of output data from
CMAQ with the input requirements for the watershed models. This is a critical
issue that will need to be addressed in a rigorous way.
Page 48 Paragraphs. Define GUI.
Page 49 Is it possible to summarize this text by developing a table which lists the potential
models, their characteristics, potential for application and limitations? It would
be good to indicate which of these models have been used in exercises similar to
what is proposed. My sense is that many of these models have never been applied
to anything even close to what is proposed here. Also which of the models have
been tested at any of the proposed study regions in a manner similar to what is
proposed here? Have any of these models been linked with CMAQ and how
successful has this been? Also, analysis of sulfate effects on mercury methylation
was mentioned in the introductory materials. Are there any models available to
assess this effect?
Page 50 Paragraph 1. Note PnET-BGC can be applied in a shorter than monthly time-step,
if that is desired. Most past simulations have been done in a monthly time-step to
reduce computative time in long-term simulations.
Generally, the draft S&M Plan seems appropriate. That said, what is discussed is very general
and not very specific. Also, what is proposed is very ambitious. I would imagine it would take
considerable time and resources to conduct the proposed analysis. Little text was focused on
critical loads. How does the Draft S&M Plan incorporate existing critical load efforts or is
compatible with critical load efforts or plans?
32
-------
Dr. Paul J. Hanson
The plan contains appropriate general principals, but the presentation and terminology
could be improved as outlined in the following comments. Policy relevant background levels of
NOx and SOx exposures, acidification, and N & S deposition should be discussed or planned to
be discussed somewhere in this document (see page A-8 for an exception). Adding line numbers
to this document would have facilitated its review.
Specific comments:
I would add definitions for the following terms used in the documents to the "Key Terms" pages:
Adverse Effect/Change - See possible definitions in text on page E-l.
Criteria pollutant
Dose-see use on page 10.
Ecological quality - used on page 8
Indicator - This term is used two ways in this document. On the one hand it is used (I
think correctly) in terms of the definition of a possible Secondary NAAQS. In other
places it is used to describe a measure of change driven by NOx and SOx pollutants. The
authors need to choose one form of use and stick with it throughout the document.
Multimedia Modeling - Why isn't this just ecological or process modeling?
Sensitive Ecosystem
Page vii: Within the definition of Total Reactive Nitrogen the last line might include the phrase
'N-containing organic compounds'.
Page 2 (top of the page): I'm not convinced that the authors should conclude that a more robust
database will be available after another 5-year cycle. Does EPA have plans to fund significant
new research? Are other agencies or organizations planning investments in N and S deposition
effects research?
Page 2 (middle): The regulatory authority for addressing NOx and SOx as criteria pollutants is
well described. More detail on how this authority translates to the subsequent extended
discussion on N and S deposition effects (acidification, N enrichment) would be appropriate. A
brief comment to this issue is added at the top of page 3.
Page 6: The previous NOx ACQD published in 1995 was not limited to a discussion of ambient
air effects of NO2 on vegetation. It included a large amount of data summarized from the acid
rain literature on N deposition effects (or potential effects).
Page 8: The connection between NOx emissions, subsequent deposition and the direct generation
of N2O solely from the anthropogenic N inputs is not strong. Should it really be included here?
References or direction to the ISA could also be used at the bottom of page 8.
33
-------
Page 9 second paragraph: The primary effect of NOx and SOx would be a direct effect on
vegetation function/survival or a direct impact on a material surface. I would classify all of the
effects discussed in this plan as secondary. The N and S deposition effects are at least one step
removed from the form of the primary pollutant. After rewording, however, I would retain this
paragraph. For complete appreciation of the impacts of NOx and SOx on public welfare, their
contribution to the generation of atmospheric particles and tropospheric ozone is worth
mentioning.
Figure 2-1 may reflect the proposed assessment process, but it doesn't illustrate the complexity
of the problem in my opinion. I recommend something like the following alternate figure. In this
example the bold arrows reflect the dominant processes to be focused on within the REA.
NOx & SOx
Emissions
Characterization of Exoosure
Characterization of Effects
rrep, O,
Direct Vegetation
Gaseous exposures x NOx S02, azene
\
% Chenist'y
* I Atmc
Atmos. Cone
NCx and SOx
Trans pel
pnoesses
I
CF
Anthropogenic
N and S ~
Indirect
"3lant uyla\e'fertil:2Lai.iy'
C"Yi.cal ;yce rr.plcato"s
"Acidification
N eriehfent- as verse
Visbiii"? & Pc
-------
NOx and SOx that are simply not dealt with in this document (e.g., tropospheric ozone and
aerosols, visibility, particulate matter).
The bottom full paragraph on page 16 needs a reference.
Bullets on page 17:
1 Good
2 May not be enough information
3 Good
4 Good
5 Critical
6 Current standards provide little protection nor do they inform the phenomenon of
acidification or N enrichment highlighted in this document.
7 It isn't clear how anthropogenic NOx from atmospheric pollutant sources can be
distinguished from other forms of N inputs (i.e., agricultural fertilizer).
Bottom of page 17: Ecosystem characteristics will be critical. This is a key issue.
Page 18: The bullet questions and key issues outlined on this page are very good. At the bottom
of the page it may not be clear what is intended in the discussion of long-term impacts (weeks,
months, years, decades??). A statement on the need to understand background levels of N and S
deposition and natural processes of acidification in a regulatory context should probably be
included.
Page 22: In the third paragraph 2nd line, please eliminate the word adverse. Not all effects are
adverse.
Section 3.1.1: To avoid confusion with the use of the word 'indicator' as a term to describe the
nature of the regulated pollutant (e.g., NOx and SOx atmospheric concentrations), I would
change the use of indicator throughout this section to something like 'measures of
response/change'.
Page 29: As a measure of nutrient enrichment, throughfall N deposition (Table 3-3) will fail to
measure the component of rainfall N absorbed by the canopy. I favor the use of total N
deposition as an indicator of pollutant exposure.
Page 30: A measure of ecological or biological change resulting from acidification orN
enrichment will be difficult to isolate from trends in other biogeochemical cycling drivers
(temperature, soil water status, herbivory outbreaks....).
Middle of page 35: The suggestions of the EES are well thought out and should be the basis for
this evaluation. It will be critical to show that the anthropogenic source of the driver for adverse
ecological changes is the cause for any perceived adverse effects. Can this be done?
Bottom of page 35: Can new and old N really be distinguished (or need to be distinguished) in
the context of the N and S biogeochemical cycles? Distinguishing anthropogenic contributions to
35
-------
N & S deposition might be accomplished via the use of stable isotope studies or measurements,
but such observations will not be available for EPA's analysis in this REA.
Page 45: As an indicator of the pollutant I would favor combined wet and dry N deposition.
Using throughfall would underestimate total N additions (i.e., the canopy is a good filter for the
extraction of N from rainfall).
Page 47: Of the monitoring networks mentioned on this page, several seem to be unrelated to air
pollution (e.g., agricultural runoff modeling). They may be needed to calculate total N
enrichment effects, but they will not inform air quality issues themselves.
Page 47: Add a reference for the "National Atlas".
Bottom of Page 55: The point made at the bottom of this page is really important. I don't,
however, know how one will define a national standard specific to regional areas.
Page 57: Application of modeling for the purposes proposed by EPA in this document tends to
assume that the models have validated mechanisms that 'get it right'. This may not actually be
the case. EPA should be careful not to apply models validated for one area/region to other areas
without appropriate attention to their utility in those regions. The authors should provide a more
detailed discussion of the mechanisms needed by 'useful' models ahead of their description of
the models chosen for the REA analyses.
Page 58 Bullet #5: Are sufficient data actually available to support the development of these
relationships.
Page 58 Bullet #7: Where the effects from loadings are anticipated to accumulate over long time
periods (i.e., years), it may be important to begin to also evaluate the potential implications of
climate change on such relationships.
36
-------
Dr. Rudolf Husar
Overall, the draft plan, Scope and Methods section is for the Secondary NAAQS for NOx and
SOx is thorough and executable. In particular, expanding the scope by combining NOx and SOx
is commendable. The combined review plan is consistent with the intertwined and multifaceted
nature of welfare impacts. However, there are several serious concerns regarding some aspects of
the Scope and Methods sections:
Question 1: Conceptual Model of the Welfare Risk
Consideration of other Welfare Effects of NOx/SOx.
The primary NAAQS is focused on protecting human health, while NAAQ Standards are
to protect and minimize the welfare effects of pollutions. The planned scope of the NAAQS
review document review is focused almost exclusively on the ecological effects of SOx & NOx.
The known welfare effects such as perturbations to the atmosphere (weather, climate, optics),
and the economic/esthetic effects on man-made materials are not covered.
Materials Effects
Exposure Indicators
Valuation, Economics
Atmospheric Effects
Indicators
Responses, Valuation
Marginalizing these non-ecological welfare effects will omit a significant rationale for
mitigating SOX/NOx emissions. This omission is also inconsistent with the recent trend toward
integrated assessment of interlinked pollutant systems: While Sox and NOX are considered as a
combined set of interacting pollutants, their combined effects are treated incompletely in
piecemeal fashion. This limits the ability of the document to provide a full rationale for acting on
welfare effects.
It is recognized that some discussion of welfare effects of Sox/NOx have been
incorporated in the NAAQS Criteria document for PM. If deemed desirable, this document could
make references to the specific sub-sections of the PM Document. However, the key findings
relevant to the welfare effects should be present in this document, so as to provide a complete
37
-------
assessment. Also, as stated below, most other documents do not treat the question of indicators
adequately.
The discussion at the RTF meeting has partially resolved my concern. A concise
summary of the sulfate and nitrate - related visibility effects in this document still seems
appropriate.
Question 2: Seven-Step Methodology
Apply Abridged 7 Steps to Non-Ecological Welfare Effects
The seven-step methodology proposed for the ecological effects is sound and executable.
The same methodology, suitably abridged for each welfare effect should be applied to the other
welfare effects. This would give consistency and scientific robustness to the document.
Seven-Step
Deposition
Data
Apply
(abridged) to
other welfare
effects
Elaborate versions of Steps 3, and 5 may not be necessary for some effects. However, the
development of sensitive areas, indices, current conditions would be desirable for these welfare
effects.
38
-------
2.3.2 Overview of Nitrogen Deposition
This section gives appropriate general exposition of NOx chemistry, concentrations and deposition.
In this overview, (including in Fig 2.2) it would have been helpful to indicate that
NOx deposition is primarily through gaseous HNO3
Total deposition is dominated by dry deposition
The addition of the NOx flow diagram discussed at the CAS AC meeting will reduce my concern.
Such a diagram should indicate the relative magnitudes of the wet/dry deposition processes.
The inclusion of N2O in this scoping document discussion is puzzling. It is stated that N2O is a
greenhouse gas but also that. "Since that is outside the scope of this review, it will not be a
quantitative part of this assessment". So, how was the scope defined? What is the rationale for
omission?
2.3.3 Overview of Sulfur Oxides and Sulfur Deposition
This section gives appropriate general exposition of SOx chemistry, concentrations and deposition.
It is not clear what is meant by "Due to known acute effects on plants, in previous NAAQS reviews,
SO2 served as the chemical indicator for SOx species." For the reviews over the past 30+ years, SO2
and SO4 have been clearly separated when presenting their concentrations and deposition. In what
context was SO2 a surrogate?
Pertaining sections 2.3.2 and 2.3.3, there is a general comment regarding the use of the CMAQ
model. The heavy reliance on CMAQ model for (concentration and) deposition estimation is of
concern. The CMAQ model performance for Sox is documented and for ambient SO4
concentrations, it seems adequate. However, the marginal performance of CMAQ for Sox wet
deposition and for non-sulfur species suggests considerable room for improvement. Additional
effort directed toward CMAQ validation should be included. The direct use of observational
data in estimating deposition (Step 4) and for extrapolating the case study areas (Step 5) should
be considered.
The discussion at the CASAC meeting regarding CMAQ and its calibration/evaluation for use in
this document has not eliminated my concern. Evidently, the CMAQ evaluation is still someone
else's problem...
3.2.2 CIS Mapping
This activity is commendable. This section outlines a very ambitious plan to generate many
of 'GIS layers' for use in the assessment. In particular, Table 3-5 summarizes the current plan for
GIS mapping layers and models to be applied in the risk/exposure assessment of targeted sensitive
ecosystems. These layers originate from a variety of sources and they represent very diverse content.
The preparation of each layer and the integration of the layers to support the seven steps is a
technically and scientifically challenging task.
It would appear useful to prioritize the preparation of data layers, test the layer- integration
methodologies on simple data fusion analyses. Finally, the credibility of this would be enhanced
if the GIS data layers prepared for the Review would be documented, cataloged, published and
made available for re-evaluation and re-use by the regulatory and scientific community.
The EPA presentation at the CASAC meeting has addressed these issues adequately.
39
-------
Dr. Dale Johnson
General Comments
As in my last two reviews, I stress the need for looking at all points of view, especially with
respect to nitrogen deposition. As in the other documents, the focus is on the negative effects,
which leaves this document open for severe criticism once it is released. This is especially true
for the potential effects of nitrogen deposition on carbon sequestration in US forests - this could
in fact be a negative effect of reducing N emissions! Once again, I do not advocate excusing air
polluters on this basis, but I think a fair and complete assessment requires that this side be
discussed. To fail to do so risks losing credibility for the entire effort.
Question 4. How well are the major effects of NOx and SOx on ecological acidification
identified and characterized? To what extent do the discussions and integration of evidence
across scales (e.g., species, communities, ecosystems, and regions) correctly represent and
clearly communicate the state of the science?
The effects of NOx and SOx on ecological acidification are fairly well characterized - that is not
the problem. The problem is the scope of this assessment - namely, the focus on acidification.
Nitrogen is the limiting nutrient for most terrestrial ecosystems and therefore there is the distinct
possibility of beneficial effects as well. I believe that it is a both a scientific and strategic mistake
to ignore the potential benefits of N deposition to the national C budget while at the same time
considering only one part of the greenhouse situation by including the negative effects of N2O. I
daresay that climate change is potentially a much larger environmental issue than acidic
deposition or N saturation is - at least it certainly looms much larger in the media and the mind
of the public these days. Nadelhoffer et al. (1999) dismiss the potential benefits of N deposition
on forest growth largely on the assumption that most deposited N is immobilized by microbial
uptake in the forest floor and soil and only a small proportion (20% in their studies) is recovered
by trees. This is in fact typical of previous forest fertilization studies which show that microbes
typically out compete trees for N. However, the argument is disputed by Jenkinson et al (1999)
find their argument unconvincing because it ignores foliar uptake: most atmospherically-
deposited N first impinges on the foliage, not the forest floor, as in fertilizer studies and in the
studies cited by Nadelhoffer et al (1999), and microbial competition in the forest canopy is
considerably less than in the soil. At a minimum, the arguments by Nadelhoffer et al (1999)
imply that greater rates of N deposition would lead to greater rates of forest C sequestration, an
argument also implied by the analysis of Hungate et al (2003). Finally, a recent study by
Pregitzer et al (2008) has conclusively demonstrated that experimental inputs of N at relatively
low levels (30 kg ha-1 yr-1) caused significant increases in C sequestration in live trees and
forest floor in forests in Michigan.
Again, I emphasize that the point of delving into this issue in some detail is not to justify
increases in N deposition but to fully ventilate both sides of that issue. The potential contribution
of N deposition, especially at higher levels, to N deficient forests has not unresolved and needs
some experimentation and careful analysis. The current proposed methods include only those
40
-------
sites where N deposition has proven to have negative effects or is likely to have negative effects,
and these sites are simply not typical of forests in the United States which are mostly N-limited
rather than N saturated. Furthermore, the results from these arguably unusual N-saturated sites,
which do not reflect the normal N limited condition of forests in the US, will then be
extrapolated, apparently to represent N effects over the entire country. This is bias in the extreme
- case studies need also to include chronically N deficient sites where growth and indeed even
ecosystem health might improve with increased N deposition! I believe that such an analysis
should be a part of the plan for assessing the effects of atmospheric deposition in the document
being reviewed. It makes no sense to partition out the acidification effects of air pollution from
the potential effects of C sequestration.
5. How well has the ISA characterized the relationship between acidifying deposition levels of
NOx and SOx and environmental effects?
All of the comments made above apply to this question as well. The ISO has some technical
problems in the way acidification is handled, as noted in my review, but the major problem is the
complete focus on acidification and the lack of perspective on the larger C budget and global
climate issues looming ahead.
6. How well characterized is the relative importance of the oxidized and the reduced forms of
nitrogen on ecosystem acidification?
There is some lack of clarity on this in the ISA document, as noted in my detailed review on p. 4-
12, lines 2-9. Ammonium is acidifying whether it nitrifies or is taken up by plants or microbes
(in the latter case, causing the release of H+ during uptake). The oxidized form of N, nitrate, is
not in itself acidifying - in fact, uptake of nitrate should cause the release of OH" to maintain
charge balance in the plant or microbe.
Specific Comments:
p. 2. paragraph 2: Effects on climate are mentioned here - such effects must include the potential
increases in C sequestration associated with N deposition from pollution in order to be a
complete picture.
p. 14, last paragraph: "can be characterized as a positive or negative effects..." Only negative
effects are sought in the case studies proposed in this document. No mention is made of potential
positive effects on the bullet list on p. 15.
p. 26, top: "we plan to identify adverse terrestrial and aquatic effects.." Why only adverse
effects? Positive effects were mentioned earlier - was this only lip service? Where is the
balance?
p. 26, bottom: "focused on what soil features should be tracked, including how organic matter
affects microbial processes." I do not understand what is meant here by how organic matter
affects microbial processes nor how it relates to the problem at hand.
41
-------
p. 27, Table 3-1: Carbon budget, growth, etc are highlighted here - but this will apparently be
assessed only in sites specifically selected to show the most adverse effects in the case study
assessment. Furthermore, the results from these arguably unusual N-saturated sites, which do not
reflect the normal N limited condition of forests in the US, will then be extrapolated, apparently
to represent N effects over the entire country. This is bias in the extreme - case studies need also
to include chronically N deficient sites where growth and indeed even ecosystem health might
improve with increased N deposition!
p. 30, paragraph 1: among ecosystem services, the example or primary production is given. See
comments directly above. Assessment of primary production appears to be limited to sites where
N effects are known to be negative.
p. 31, paragraph 1: Most ecosystems in the US are sensitive to N deposition in the sense that
growth increases might occur. In some cases this will be "bad", as in exotic species invasions,
and in some cases this will be "good" as in commercial forests.
p. 32, GIS mapping: I firmly believe that the GIS mapping exercise should also include a
thorough analysis of potential effects of increased N deposition on C sequestration.
p. 36, Table 3-5: Indicators should include foliar N concentration and should be stratified by
species. We have reasonably good data on foliar N levels for deficiency, and even toxicity levels
for many species. Foliar N should be included in the indicator list.
p. 38: The list of prospective sites here should include fast growing forests of the southeast
(loblolly pine, for example) where N inputs may well cause greater growth and C sequestration.
Also should include the San Bernardino sites in southern California where N saturation is
extreme. San Bernardino's are buried in Table 3-7, but should be included here. This list is
biased toward sites with negative effects and incomplete.
p. 44: Lake Tahoe should include forest effects, were forests are thought to be typically N
limited. Lake Tahoe is also somewhat unique in that much of the pollution enters the lake
directly since ratio of land to water in the watershed is very low.
References:
Hungate, B.A., J.S. Dukes, M.R. Shaw, Y. Luo, and C.B. Field. 2003. Nitrogen and climate
change. Science 302: 1512-1513.
Jenkinson, D.W., K. Goulding, and D.S. Powlson. 1999. Nitrogen deposition and carbon
sequestration. Nature
Nadelhoffer, KJ.,. B.A. Emmett, P. Gunderson, OJ. Kj0naas, CJ. Koopmans, P. Schleppl, A.
Tietema and R.F. Wright. 1999. Nitrogen deposition makes a minor contribution to carbon
sequestration in temperate forests. Nature 398: 145-148.
42
-------
Pregitzer, K.S., AJ. Burton, D.R. Zak, and A.F. Talhelm. 2009. Simulated chronic nitrogen
deposition increases carbon storage in Northern Temperate forests. Global Change Biology
14: 142-153.
43
-------
Dr. Donna Kenski
I had 2 assignments for this document so I'll address those first:
I was asked to comment specifically on Sec. 2.3.1. While it minimally fulfilled its stated purpose
(i.e., to provide an overview of the risk assessment framework), it was too abbreviated (4
paragraphs!) to be as effective as it could be. Section 3.7 actually did a much more
comprehensive and coherent job explaining the risk assessment and its application to the
standard-setting process. In fact, Section 3.7 was the first time I felt like this document actually
pulled all the pieces together and made a case for how the process should work. Consequently I
recommend beefing up Sec. 2.3.1 with much of the text from Sec. 3.7 (the bulleted lists
especially, as well as the revised version of Fig. 2.1).
My second assignment was Charge question 11: "Additional ecological/welfare effects due to
NOx and Sox emissions that we do not currently anticipate evaluating in detail in this review
include the following:
Nitrogen saturation,
Maple decline,
Ammonia air deposition and toxicity to native mussels,
Relationships between acidity/nutrient enrichment and mercury methylation,
Sensitive areas for acidity/nutrient enrichment impacts, identified from biogeochemical
characteristics, and
Climate change effects due to N2O.
Does the Panel agree that these represent lower priority effects for the current assessment? If
not, what does the Panel recommend?"
While I agree that these MAY be lower priority effects, I don't think they necessarily should be
neglected altogether. Although the plan says that N2O, for example, will be included in the
scope of the review in a non-quantitative way (p. 13, Sec. 2.3.2), it never mentions how these
lower priority effects will be included, even in a qualitative way. The rest of the plan seems to
focus solely on the modeling of the 4 defined risks (p. 15, aquatic and terrestrial acidification and
nitrogen enrichment) and never circles back to pick up these other effects. At what point do they
get incorporated and how? It this is part of step 5, it was not obvious. It should be explicitly
stated.
Of these lower priority effects, the only one that seemed really tangential is the ammonia
deposition and toxicity to mussels. It is well established that mussels are sensitive to ammonia in
water, but I was unable to find any papers on ammonia deposition and subsequent impact on
surface water concentrations. These should be cited, at least in Appendix A where the subject is
briefly reviewed.
44
-------
Other general comments:
I found this document very difficult to follow, but I'm not sure exactly why. Organizationally it
seemed sound. There were some terminology issues that complicated the presentation. For
example, total reactive nitrogen is defined in the glossary as oxidized and reduced species as well
as organic compounds, but this is not the standard terminology in atmospheric chemistry, I
believe (it may be in the deposition/ecological world). Seinfeld and Pandis (in Atmospheric
Chemistry and Physics) and many others define reactive nitrogen as NOy (oxidized and organic
species), not including reduced nitrogen species. It seemed like the document was inconsistent,
using reactive nitrogen to mean NOy sometimes and other times to mean NOy plus NHx. For
example, p. 11, Sec. 2.3.2, 4th line, defines total reactive nitrogen as NOy (or at least I think
that's what it's trying to do, with an unfortunate typo using reduced for reactive). Then again on
page 23, end of 1st paragraph, it says EPA plans to evaluate the contribution of NOx to total
reactive nitrogen relative to reduced forms of nitrogen. So I'm left quite confused about how
and when ammonia/um will or won't be incorporated into this assessment. This leads to a larger
concern when we get to the very end (p. 59), where some useful bullets itemize questions about
the levels and forms of the standard that might be considered. Somewhat ominously, it sounds as
if these questions might not be discussed as part of the risk assessment but only as part of the
policy assessment (i.e., the ANPR). I hope that is not the case and that the coming risk/exposure
assessment will discuss these options thoroughly. More important, it is not clear from this short
list whether the indicator itself will be evaluated, and again whether it will include total nitrogen
effects, versus total reactive nitrogen (NOy + NHx or just NOy??).
The general approach, especially the reliance on GIS mapping, seems sound, but the real
question is how much data exist to support this effort in a meaningful way. Not possible to
assess this from the Scope and Methods Plan.
Appendix B is too short to be useful. Rather than just describing the PnET-N-DNDC model and
its input parameters, it should have incorporated model results.
45
-------
Dr. Naresh Kumar
Charge Question 1. In outlining the scope of this risk/exposure assessment, we have created a
flow diagram that represents how nitrogen and sulfur compounds move from 'source to dose' in
the environment (see Figure 2-1). How adequately does this conceptual model for evaluating
risks due to deposition-related ecological effects characterize what should be covered in the
scope for this assessment?
The conceptual model shown in Figure 2-1 is a simple, but efficient way to represent risk
assessment framework for deposition-related ecological effects. Various endpoints, such as
"acidification", "nutrient enrichment", "climate change", etc. are listed outside the dotted line
suggesting that these will be described in detail in the assessment document. However, later it is
mentioned that the N2O effect (climate change) is outside the scope of this assessment and will
be included only quantitatively. It is fine to do so, but Figure 2-1 should be modified to make it
clear which endpoints will be assessed only quantitatively (one suggestion is to italicize such
endpoints).
In Section 2.3, Page 8, where the N2O effect is discussed the first time in the text, a discussion on
the scope of assessment for N2O effect should be included. For example, it should be made clear
whether the N2O effect will be part of final deliberations for review/revision of the NAAQS for
NOx and SOx.
Also, carbon sequestration should be added as an endpoint under "Nutrient Enrichment" in
Figure 2-1 and a discussion should be included in the text describing this effect. Similar to the
N2O effect, it should be made clear whether the carbon sequestration effect will be part of final
deliberations for review/revision of the NAAQS for NOx and SOx.
46
-------
Dr. Myron J. Mitchell
Final Comments on: Draft Scope and Methods Plan for Risk/Exposure Assessment: Secondary
NAAQS Review for Oxides of Nitrogen and Oxides of Sulfur
General Comments
More consistency is needed throughout the document on the different definitions and forms of
chemical species including the grouping of these chemical species into classes. The linkage of
acidification with cation (e.g., Ca and Mg) nutrient depletion and the mobilization of toxic
cations (e.g., aluminum) needs to occur earlier in the document so that the total ramification of
acidification is clearer. The document needs considerable work with respect to consistency and
the use of precise descriptions. Some statements lack clarity. The criteria for selecting case
studies are discussed, but an unambiguous delineation of the importance of specific criteria in
making these selections is needed. Throughout the document there are statements about the
needs for data. It would be better to state that "information" is what is required. Data
availability without previous interpretation and analyses are less useful than those data sets that
have been evaluated. Those data sets that have been evaluated by multiple approaches would be
especially useful. Some reference to the European experiences in developing methods for
Risk/Exposure Assessment might be insightful including the relatively wide application of the
MAGIC model at various European sites.
Within these proposed scope and methods in some instances the plan seems very ambitious and I
am not certain that completion is possible within the time frame of the assessment and with the
resources available to do these tasks. This issue may be especially important in evaluating
impacts on ecosystem services. The document needs to more clearly articulate priorities with
respect to various approaches. I would suggest that it may be most facile to focus especially on
the case studies to show most clearly the linkages between sulfur and nitrogen deposition and
adverse damage to ecosystems.
Specific Comments
Page Comment
2 It is not clear whether N2O is included with the definitions of "oxides of nitrogen"
and or "nitrogen oxides." On page 11 it is stated that "the family of nitrogen
oxides includes any gaseous combination of nitrogen and oxygen, e.g., NO2, NO,
N2O, N2O3, N2O4, and N2O5."
6 Change to: in previous NAAQS reviews on nitrogen or sulfur
47
-------
11 Change to: In an analytical approach unconstrained by data and other resource
limitations, one could envision a comprehensive risk assessment
covering all potentially affected ecosystems and all scientifically
supported effects on those ecosystems in the United States.
11 The identification of the N chemical species provided below needs to be better
aligned with the definitions of N chemical species given on page 2: "The sum of
mono-nitrogen oxides, NCh and NO, typically are referred to as nitrogen oxides
(NOx) in the atmospheric science community. More formally, the family of nitrogen
oxides includes any gaseous combination of nitrogen and oxygen, e.g., NCh, NO,
N2O, N2O3, N2O4, and N2Os. Total reduced nitrogen (NOy) includes all nitrogen
oxides as well as gaseous and particulate nitrate species such as HNOs, PAN, and
aerosol phase ammonium nitrates. Reduced atmospheric nitrogen species include
ammonia gas (NFb) and ammonium ion (NH4+), the sum of which is referred to as
NHx."
13 The following statement is confusing with respect to how N2O is treated in this
assessment: "Since the definition of "welfare effects" includes effects on climate
[CAA Section 302(h)], we will include N2O within the scope of this review.
However, it is most appropriate to analyze the role of N2O in anthropogenic climate
change in the context of all of the greenhouse gases. Since that is outside the scope of
this review, it will not be a quantitative part of this assessment."
13 I don't believe that generally the SO2 or SOX includes particulate sulfate particles as
stated: "SO2 is one of a group of substances known as SOx, which include multiple
gaseous (e.g., SO2, SO, SOs, 8203, 820?) and particulate (e.g., ammonium sulfate)
species (Figure 2-4)."
21 Is this statement really necessary: "In addition to these four effects, we plan to
address, as appropriate and within our time constraints, impacts associated with
nitrous oxide (N2O) and the influence of sulfur enrichment on methylmercury
production."
22 This statement needs to be reworded to clearly suggest the challenge associated with
making an assessment of impacts on a diverse set of ecosystem types over a range of
atmospheric inputs: "The anticipated spatial extent and diversity of ecological
effects due to deposition of nitrogen and sulfur do not facilitate a nationwide
analysis."
24 Change to: case study areas, data may not be sufficient to perform a quantitative
assessment for each
28 In Table 3-2, why are K and Na not included with Ca and Mg as base cations?
28 Change to: "exchange capacity (CEC) are more widely available at the present time
than information on biological indicators"
48
-------
28 The following statement implies that CASTNET underestimates dry deposition of
nitrogen. If this is the case, further explanation is warranted: "This creates
complications in developing estimates for total nitrogen deposition levels because
dry-deposition data sources will likely be underestimated."
29 The following statement is not true since there is a linkage with Ca availability and
tree species (e.g., sugar maple) that produce litter this has higher N mineralization
and nitrification rates: "For terrestrial ecosystems, low calcium to nitrogen ratios in
soils are commonly related to increased nitrification and potential increases in soil
acidity and releases in NCb to receiving waters; however, these measurements are not
always widely available." Literature examples include:
Christopher, S.F., B.D. Page, J.L. Campbell and MJ. Mitchell. 2006. Contrasting
stream water NCV and Ca2+ in two nearly adjacent catchments: the role of soil Ca
and forest vegetation. Global Change Biology 12:364-381.
Page, B.D. and MJ. Mitchell. 2008. Influences of a calcium gradient on soil
inorganic nitrogen in the Adirondack Mountains, New York. Ecological
Applications. (In Press)
29 Change to: For aquatic ecosystems, the indicators for "nutrient enrichment" effects
reflect a combination of inputs from various sources
29 Not sure what "are encouraged" implies in the following: "are encouraged for
inclusion in numeric criteria as part of EPA-approved state water quality standards
(U.S. EPA, 2000)".
29 Clarification is needed on the "data-fusion approach."
29 In table 3-3 change to: "Reflects a combination of inputs from various sources
(air, discharges to water, diffuse runoff, and groundwater inputs)".
29 In table 3-3, the following statement (Repeated twice) is confusing: "Relative role
of air deposition should ideally be compared with air deposition data and also
with available (preferably multi-media) models."
How can the models be separated from air deposition data since the models are
needed to calculate deposition?
30 Is it correct to include within ecosystem services "cultural services including
spiritual or religious values, aesthetic values?"
30 What is meant by "type of environmental system?"
30 It is not clear how the following approach will produce the needed outcome:
"Identify Databases of Indicator Conditions: The indicators selected will relate to
49
-------
available compendiums of literature abstracts or actual database systems (as stand
alone files or accessed through Web portals) to provide readily available and
transparent ways to document the nature of the indicators and the indicator conditions
used to define the environmental impairments."
"Compendiums" should be "compendia."
30 Change to: Identify and Address Temporal Issues: Different ecoregions,
landscapes, biological provinces will have differing degrees of susceptibility to
impairments or differing recovery potential, depending on edaphic characteristics,
past land use and/or pollution histories.
30 For the following statement, be more explicit on the actual times needed for recovery
(fairly rapid and much longer recovery times need some approximate values):
"Some ecological systems may be capable of fairly rapid recovery responses
once pollutant loadings are significantly abated; other systems, such as larger
estuarine aquatic systems, may require much longer recovery times."
31 Provide better description versus the following j argon: or at least define these terms
"potential near-field and far-field linkages."
31 The terms in Figure 3-2 should be reflected in the text.
31 Change to: Inland acid-sensitive waters in the eastern United States and nitrogen-
sensitive ecosystems in the Rocky Mountains and other parts of the western United
States may require large-scale, special-area assessments.
31 Clarify why this area is special (Is this different from the sensitivity of other areas?):
"Ecosystem effects in special areas"
31 How could linkages not be geographically significant? Note the statement: "If the
linkages are geographically significant"
32 Why this specific example: (e.g., local research of MeHg formation in Devil's Lake,
WI).
If this example is used, at least a citation is needed.
32 Change to: Of special interest will be the characterization of linkages that can be used
for synthesizing results for the entire U.S.
32 Why "may" in the following statement: "Information that deals with special case
study areas (e.g., the Adirondack Mountains or special alpine and sub-alpine"
Previous statements have suggested the importance of focusing on sensitive
systems.
32 For heading "3.2.2 GIS Mapping" delete mapping; this is redundant.
50
-------
32 Reword for clarity and precision: "To describe the national picture, we plan to map
the locations of those sensitive ecosystems identified in Section 3.2.1 and identify the
characteristics of the biological and biogeochemical properties that create the
sensitivity."
32 By "unmapped areas" does this imply areas in which an appropriate GIS is not
available?
32 There are different issues associated with scale and defining the boundaries of a GIS.
Clarify the following statement: "Sensitive areas can be identified at different spatial
scales by using different approaches for defining the boundaries of the mapped
units."
32 The description under 3.2.2 needs considerable modification to clarify how GIS will
be applied to this assessment.
33 The linkages need to be made more explicit in some components of: Figure 3-3.
Documented biological, biogeochemical, and physiographic linkages.
34 Provide further details on the source of the "ESRI 8.3 data disks."
35 It is stated that "Where case study or ecosystem-specific data are available, a subset
of maps for the case study assessment area may be created."
Are only "data" needed rather than information associated with the interpretation
of these data?
35 It is suggested that "Complementary to these efforts, we may use a statistical cluster
analysis to group ecosystem units into similar sets. Clustering ecosystems might
reduce the number of locations that need to be modeled to adequately characterize the
variability in ecosystem response to changes in nitrogen and sulfur deposition."
It would also be helpful to identify key ecosystem attributes that are most critical
in the classification and grouping of these ecosystems with respect to effects of N
and S pollutants.
35 It is stated that "In selecting areas to assess ecological effects from air deposition, the
SAB Ecological Effects Subcommittee (EES) suggests consideration of (1) clear
quantifiable ecological effects due to air pollution, (2) the degree to which a
significant component of ecological effects are attributable to air pollution, (3) the
responsiveness of ecosystem services to changes in air pollution, (4) the cumulative
impacts of multiple air pollutants, (5) the abundance of ecological effects and
economic benefit cost analysis, and (6) the visibility to the public and value of
resources at risk (U.S. EPA, 2005). While these recommendations were made in the
context of a prospective cost-benefit analysis, many of these recommendations are
sound in the context of our NAAQS risk analysis."
51
-------
Having some hierarchy in these selection criteria would be useful since there could be
distinct differences among ecosystems on the relative value of each of these criteria.
35-37 In addition specific recommendations are provided from the ESS regarding
selection criteria. These factors need to be blended into a more cohesive set of
criteria for selection.
36-37 The information provided in Table 3-5 (Summary of Indicators, Mapping Layers,
and Models for Targeted Ecosystems) is very diverse with respect to types of
information, spatial scales, level of previous analyses, etc. It will be very useful
to clearly indicate priorities in the selection of attributes from this table. This
selection should be based upon well-identified criteria.
45 The following statement is weak: "Depending on the adequacy and abundance of
data for areas, the evaluation may entail computer modeling, statistical analysis, or
qualitative analysis."
Clearer criteria of the importance of the type and amount of information available
need to be provided in selecting case study sites.
45 I am not sure that the following indicator is useful due the role of canopy
exchange processes in markedly affecting the concentration and fluxes of N
chemical species in throughfall: "Nitrate, ammonia, organic nitrogen throughfall
deposition for terrestrial ecosystems."
46 It is stated that: "We plan to evaluate the spatial adequacy of available monitoring
data including GIS mapping of documented data to identify any meaningful
spatial gaps."
What will be done if these spatial gaps are found?
46 It is stated that: "For each ecosystem effect, we plan to determine if there is a
temporal dimension to exposure."
Isn't there always a temporal dimension to exposure? Aren't more critical issue
aspects associated with non-linearities, thresholds, lag-times, etc.? Some of this is
discussed in the following sentences.
46-47 Each of these sources of loading information has limitations. Including some
details on the errors of these estimates would be helpful
47 In discussing the "CMAQ Deposition Modeling" the issues related to
uncertainties associated with the calculation of deposition velocities as a function
of surface types needs some mention with respect to the overall confidence in the
predictions.
52
-------
48 In discussing acidification the importance of the contribution of mobile anions
(sulfate and nitrate) to the loss nutrient and toxic cations needs more emphasis.
The pH effect is only a small part of the impacts.
50 An important criterion for selecting regions to model should include availability
of previous modeling efforts.
50-51 Further clarity is needed in the section on "Assess Uncertainty in Loading and
Exposure Computations." For example does resolution include both temporal and
spatial considerations? What is meant by "differing complexities?" Does this
mean landscape complexities, complex processes, or something else? If more
inventories of reactive N species are needed for the modeling, how will these
inventories be done in the context of this effort?
51 Is it feasible to determine "the contributions from air deposition sources can be
analyzed relative to all other major anthropogenic or natural sources for sensitive
ecosystems across the country?"
51 How will the assessment balance the needs of doing rigorous analyses in one or
possibly few regions with the greatest level of accuracy and precision versus a
broader range of regions with lower accuracy and precision?
52 Be more specific in which parameters will be evaluated with respect to "response
curves."
52 More clearly differentiate how the Calculate Desired Exposure Endpoint
compares with the determination of critical load including the criteria used in
these determinations.
53 The following approach is quite vague and ambiguous: "The specific methods
used to evaluate adversity will depend on the availability of data and methods for
the indicators of interest related to acidification and nutrient enrichment and on an
assessment of the appropriateness of each type of quantification in comparing
different levels and forms of the standards. In our initial assessment of the
available data, given the timeframe for this NAAQS review, we have determined
that the most useful approaches will be those focusing on quantifying the link
between changes in ecosystem indicators and ecosystem services."
Can some priorities be suggested on which indicators would have highest
priority?
53 Doing the following is certainly a daunting challenge: "Thus, when ecosystem
services are quantified and their ecological response functions to NOX and SOX are
modeled, it is imperative that the entire bundle of services be evaluated, and that
the linkages and tradeoffs among ecosystem services are included in the
quantification (i.e., ecological tradeoff functions [ETFs])."
53
-------
54 Is it feasible to do the following within the time frame of this assessment?
"Therefore, at this time, data mining will be central to developing at least a
preliminary assessment of potential impacts of NOX/SOX deposition and acidity on
ecosystem services. In the current plan, process-based models are being
considered to be used to (1) synthesize/link the suite of ERFs and ETFs and (2)
generate maps and summaries of ecosystem services and tradeoffs in response to
current and future ambient air indicators for NOX and SOX. The collection of
response and tradeoff functions will aid in the valuation of the services at risk to
these criteria pollutants where possible."
55 To accomplish the tasks that are stated might require a very selective choice of a
region or a few regions where the available information is sufficient to do this
work within a relatively short time:
1. Identify areas/regions of the country receiving high levels of NOx/SOx
deposition and acidity impacts.
2. In those regions, identify ecosystems sensitive to elevated levels of nitrogen
and sulfur, using some common selection criteria.
3. In those sensitive ecosystems, ask what ecosystem services are expected to be
prevalent and "valued" (i.e., some subset of all potential services).
4. In those areas, identify what data are available to develop ERFs and ETFs, at
least to a qualitative degree that would enable production of spatial and temporal
maps to identify different degrees of protection that would exist under alternative
secondary NAAQS. Chan et al., (2006) produced such maps for several
ecosystem services in the Central Coast ecoregion of California. The linkage and
comparison of multiple ecosystem services in the region would provide
information for consideration of tradeoff value of one service versus another.
58 The following evaluation approach is extremely ambitious:
"1. Identify the relative contribution of loadings associated with atmospheric
deposition of nitrogen and sulfur.
2. Identify the most critical impacts from nitrogen and/or sulfur loadings (i.e.,
acidification, nutrient enrichment, or eutrophication).
3. Identify the contribution to atmospheric loadings from total reactive nitrogen,
NOx, and SOx.
4. Identify the biogeochemical indicators/resources of concern in the assessment
area and the ecosystem services associated with those indicators.
a. Determine the ecosystem service effects associated with the most-critical
impacts.
b. Bundle ecosystem services to find common metrics for comparison across
locations.
5. Define the exposure-response (loading-response) functions (ERFs) for the
ecological indicators of concern.
54
-------
6. Estimate the loadings/exposures associated with current and alternative levels
of the NOx and SOx standards (using CMAQ modeling).
a. Analyze the relationships between NOx, SOx, and other reactive forms
of nitrogen.
b. Assess the impacts of meteorological variability on these relationships.
7. Estimate the ecosystem impacts associated with estimated loadings.
8. Convert estimates of individual ecosystem risks to common units using
a. economic valuation based on benefits transfer from existing literature
estimates
b. biogeochemical equivalents using ecological tradeoff functions (ETFs).
9. Combine individual risk estimates to produce overall impact estimates"
Would it be more feasible to determine a hierarchal approach with respect to the
selection of case study areas where the entire suite of approaches is utilized while
for others a more limited assessment will be made?
59 The questions posed are valid and interesting, but addressing them will be a major
task:
"How do alternative levels and forms of the standards relate to a given exposure
metric?
What are the appropriate averaging times for alternative levels and forms of the
standards?
What alternative levels of the standards should be considered?
Should there be alternative levels of the standards (i.e., individual NOx and SOx
standards or a combined NOx/SOx standard)?
Do the ambient air indicator forms allow for site-specific protection while
maintaining national consistency?
Does the ambient air indicator adequately account for the effects of total reactive
nitrogen?
Does the form of the standard have an impact on the risk?
As I have said previously with respect to the overall approach, should some of
these questions have greater priority?
A-l In determining "Sensitive Ecosystems" having some criteria related to information
availability would facilitate the assessment of "Targeted Ecosystem Effects."
A-l The susceptibility of the ecosystems to the depletion of nutrient cations should
also be an important factor with respect to evaluating sensitivity.
A-l In grouping ecosystems "statistical cluster analysis" should be supplemented with
consideration of the major factors that are most critical in determining the impacts
of N and S deposition. Can the cluster analysis include weighting of the most
critical factors?
55
-------
A-2 In addressing the issues of N enrichment the importance of the generation of the
mobile nitrate anion with respect to mobilization of toxic Al and H ions and loss
of nutrient cations (especially Ca and Mg) needs to be included.
A-2 The quantification and analyses of ecosystem services in the context of this
assessment will be a major challenge.
A-3
to
A-5 A major task will be the separation of atmospheric derived fixed N versus other N
sources especially fertilizer and waste treatment in making evaluations for the
larger coastal watersheds.
A-5 The heading "A. 1.3 Aquatic Sulfur Enrichment" is likely not appropriate since
much of the discussion focuses on the generation of MeHg which most likely is
predominantly generated in wetlands. Maybe a separate section on wetlands
would be helpful.
A-6 The following statement is not correct in the context of sulfate/MeHg
relationships: "The "cause-and-effect relationship" between sulfur and mercury
deposition from the atmosphere has been demonstrated in the lab and in
small-scale field experiments".
A-6
to
A-7 The section on "Terrestrial Acidification Due to Nitrogen and Sulfur" with respect
to the discussion on nutrient cation depletion should briefly indicate the
importance of specific tree species such as sugar maple in this assessment.
A-7 In discussing episodic acidification, some mention of climate linkages would be
appropriate. This is relevant not only to snowmelt episodes but also the summer
storm events, especially those following droughts.
A-8 It would be better to indicate that this is "Sugar Maple Decline". Some maple
species are very resilient to atmospheric pollutants and nutrient depletion.
A-9 Data suggests that sugar maple has higher Ca requirements than other dominant
species in the northeastern forests of the U.S. making it more sensitive to Ca
depletion. See for example: Page, B.D., T.D. Bullen and MJ. Mitchell. 2008.
Influences of calcium availability and tree species on the cycling of Ca isotopes in
soil, vegetation, and stream water. Biogeochemistry. DOI 10.1007/sl0533-008-
9188-
56
-------
B-l
to
B-2 The importance of N2O emissions is due to its role as a green house gas. It may
not be feasible to address fully the importance of this factor in this assessment
without spending considerable effort integrating green house gas discussion into
the document.
C-l
to
C-4 Some consideration is needed on the accuracy and precision of the estimates
associated with the use of the CMAQ MODELING. How will the uncertainties
associated with the modeling of the atmospheric deposition of sulfur and nitrogen
affect the ability of the assessment to predict the effects of spatial and temporal
patterns?
57
-------
Mr. Richard L. Poirot
Supplemental Post Meeting (4/2-3/08) Individual CASAC Comments on Secondary SOx &
NOx Integrated Science Assessment and Ecological Risk/ Exposure Assessment Draft
Scope and Methods Document
Shortly after the 4/2-3/08 CASAC Review of the Secondary SOx & NOx Integrated Science
Assessment and Ecological Risk/ Exposure Assessment Plan, it has come to my attention that
EPA funding for several critical long-term atmospheric deposition-related monitoring and
research programs, which have been invaluable to understanding the continuing effects of sulfur
and nitrogen deposition, are scheduled to be substantially reduced or eliminated in the
Administration's proposed 2009 budget. Specifically, the Clean Air Status and Trends Network
(CASTNET) air quality and dry deposition network is scheduled for a 25% funding reduction in
2009, and the Temporally-Integrated Monitoring of Ecosystems and the Long-Term
Monitoring (TIME/LTM) programs (which had recently been saved from proposed cuts in 2008,
but) will be considered "completed" (i.e. terminated) in 2009. See attached letter from Assistant
Administrator George Gray to Gary Lovett (chair of the Northeast Ecosystem Research
Cooperative steering committee).
The timing for reduction and/or termination of these valuable long term programs is especially
poor, since the Agency is currently considering secondary standards for SOx and NOx, which are
being specifically focused on the environmental effects of S and N deposition. At the same time,
large additional S and N emissions reductions are scheduled in the Eastern US under the Clean
Air Interstate Rule (CAIR), while new sources of S and N emissions are coming on line in many
western states, northern Mexico and western Canadian provinces.
The ISA and Ecological Risk/Exposure documents should be modified to reflect these planed
changes in the Agency's sulfur and nitrogen deposition-related monitoring and research
programs, clearly indicating:
CASTNET, TIME & LTM are scheduled for large cuts or elimination in 2009,
- why the Agency plans to reduce or eliminate these long-term programs,
how the Agency plans to evaluate environmental effects of future emission changes from
CAIR, new SOx or NOx NAAQS, and/or other emissions changes,
- how the Agency plans to consider secondary SOx, NOx or PM standards in this or future
NAAQS review cycles without the information on deposition and ecological effects that
these programs provide.
58
-------
%
CTION AGENCY
D.C.
\
S -«<*«
OF
»JO
Mr, Gary M. Lovett
Chair, Steering Committee
Northeastern Ecosystem Research Cooperative
C'ary Institute of Ecosystem Studies
Box AB. Millbnxik, NY
Dear Mr,
Thank you for your letter of January 29. 2008, requesting information on Environmental
Protection Agency's (EPA) to the Temporally- Integrated Monitoring of Ecosystems
and the Long-Term Monitoring (TiMli'LTM) associated with RPA's Acid Rain
Research Prop-am, We are pleased in report that we are able to both of these programs in
FY at the FY 2007 resource levels of seven ($720.000.00). As
you suggest, the of these monitoring is ant! the result of many years of
by the EPA's Office of Research and Development (ORD), These two programs have
been extremely valuable in demonstrating the effects of the reduction of sulfur on aquatic
resources, reflecting changes from ihe most recent Clean Air Act amendments
It should be noted that the focus of the research in theTlMBLTM programs was on the
of the monitoring program, of indicators to and reporting
»n those as a means of verifying the intended results. With the FY 2008 ORD
will complete its defined goal for both of these programs, EPA will the appropriate
disposition of these programs in FY 2009.
Again, thank you for your letter. If you have additional questions, please contaet me or
your staff may call in EPA's Office of Congressional and Intergovernmental
Relations, at 202-564-1695.
Dr. George Gray
Administrator
Office of Research and Development
Lek Kadeli
Kevin Tcichinan
Car! Ma/y.a
Becky Iliggins
59
-------
Comments on SOx/NOx Eco-Risk Assessment Draft Scope & Methods Document
It was a pleasure to read through this planning document. It was clearly written and well-
organized, founded on a sound conceptual model and guided by an excellent series of key policy-
relevant questions (on pages 17 & 18). Because of the new and unfamiliar NAAQS review
process, its difficult to judge how far the information should be developed in each stage of the
review (what belongs in the domains of the ISA, the risk assessment or the policy assessment)
but it appears like the planned ecological risk/exposure assessment should provide an adequate
basis for consideration of policy options for secondary NOx and SOx NAAQS.
At the same time, I think the potential for effective secondary standards could be substantially
enhanced or limited by how broadly or narrowly the pollutant effects of interest are defined (for
example including or excluding aerosol-phase effects), and by how much flexibility is allowed in
considering the measured indicators by which compliance with the secondary standards will be
determined (for example: SO2, total atmospheric sulfur, sulfur deposition, exceedance of critical
loads for S or S+N deposition, etc.). Statements on page 56 and 57 of the plan imply that
deposition-based indicators of ecological risk are not at all the same as and will need to be
related to comparable ambient air concentration indicators - perhaps limited to the nominal
gaseous criteria pollutants of sulfur dioxide and nitrogen dioxide. If that is the case, I doubt very
much that any useful secondary standards can be developed for these pollutants, and I think this
critically important issue and should be discussed in some detail at this review meeting.
Assuming this issue can be satisfactorily resolved, I also much admire the carefully reasoned
conceptual model as well as the very ambitious approach for conducting the assessment. I don't
personally have enough familiarity with the many types of atmospheric, geological, chemical,
biological and socio-economic data and model parameters needed to conduct the proposed
assessment, but it strikes me as a very challenging undertaking to first collect all these necessary
layers of information and then and then make them all fit together to produce the proposed
quantitative assessments, especially since each step in the process depends upon successful
previous steps. I note the phrases "we plan to" or "we intend to" appear 43 times in the
document, but there are no "we have done"s. In many cases, I assume you have already gathered
some of these data and tested their use in some of the intended applications - i.e. have some
sense that this approach will actually work. It would be helpful for reviewers to see some
example intermediate demonstration products to give a better sense that this plan will actually
work.
Charge Questions 8-10:
8. In the seven-step approach to the current conditions risk/exposure assessment, Step 5 (Section
3.5) discusses how to scale up case study areas to more spatially extensive sensitive areas, where
appropriate. Does the Panel agree with this approach or can they suggest alternatives?
Its not entirely clear how geographically limited your case study areas will be, but generally the
process of scaling results to larger spatial domains should be reasonable if your case studies
identify the key biogeochemical parameters associated with the effects, and if you have data with
sufficient quality and resolution to identify similar conditions over larger regions. I imagine this
60
-------
will often involve moving from intensive locations where detailed on-site measurements are
available, to broader spatial areas where surrogate data estimates will need to be employed.
Presumably it will be possible to test some of your projected scaled effects at other sites where
detailed data are available. Possibly you could also test scalability by projecting site-specific
results to higher or lower deposition rates from the recent past. It should also be possible and
desirable to include some bounding estimates as you extrapolate results over space (& time).
9. In the seven-step approach to the current conditions risk/exposure assessment, Step 6 (Section
3.6) outlines a path to assess the current conditions of sensitive ecosystems. How well does the
Panel agree with the approach outlined for calculating response curves and utilizing mapping
and ecosystem services to characterize current conditions or can the Panel recommend
alternative approaches?
How well does the Panel agree with using ecosystem services to provide a common metric for
comparing ecological risks due to nitrogen and sulfur deposition effects?
How well does the Panel agree with collecting current valuation studies to understand the value
of bundled ecosystem services? Can the Panel recommend additional or alternative approaches?
I have only general comments here, as I really don't know enough about the "ecosystem
services" approach for assessing valuation. I think it should be possible to develop reasonable
ecosystem response curves for a variety of endpoints as well as to provide a detailed descriptive
summary of the various short and long-term consequences that are expected to result if critical
loads or other such effects limits are exceeded. I would think this descriptive information of
expected consequences could be carried through to future steps in the process, regardless of
whether the "value of bundled ecosystem services" can be effectively communicated to your
readers. I don't think it's absolutely necessary to directly compare/contrast the risks of S vs. N
pollution (its not like we can have a standard for only one or the other), but if you have a way to
present the very different effects of acidification & nitrification in common units than yes, this
could be very helpful.
10. In the seven-step approach to the current conditions risk/exposure assessment, Step 7
(Section 3.7) describes an approach to assess degrees of protection/levels of effects under
alternative forms and levels of ambient NOx and SOx standards. This approach attempts to
describe how the methods, models, and results of the current conditions risk/exposure
assessment can inform our evaluation of the appropriate form(s) andlevel(s) of a national
standard. How well does the Panel agree with the approach outlined in this section, the issues
presented, and the 9 steps outlined to assess potential forms and levels of the standard? Please
suggest any additional or alternative steps we should take into consideration.
As indicated above, I have a major problem with the implied constraint on NOx/SOx indicators.
If you are talking only SO2 and NO2 as NAAQS indicators, I think chances of useful secondary
NAAQS are slim. A comparison of spatial & temporal fields of S & N emissions, ambient
concentrations of gaseous SO2, NO, NO2, HNO3, NH3, aerosol SO4, NO3, NH4 and wet & dry
S and N (oxidized & reduced) deposition (recent past, current, and projected [post CAIR] future)
should be a relatively straight-forward modeling exercise - which we would want to see
confirmed to the extent possible with measurements. This kind of information can and should be
conducted in parallel with and independently from the eco-risk assessment, and results
61
-------
distributed as soon as feasible to this panel for review (if you can provide all of the above in
gridded GIS data layers so much the better). At the same time, I think it would also be useful to
consider the possible benefits of a wider range of indicators than SO2 and NO2.
In making this broader assessment, I would encourage you to continually consider what the best
ecological indicators would be and then consider whether we have in place adequate
measurement programs to track them in the locations of concern. I would also encourage full
consideration of the sources, effects and benefits of controlling reduced N, regardless of whether
anticipated policy-relevant "NOx" NAAQS indicator metrics are likely to include it or not.
62
-------
Mr. David J. Shaw
General Comments
Outcome ofREA
Is this REA the same one that is described as due August 2008 (Table 2-1)? The REA Plan is
difficult to assess in terms of what outcome to expect because it has so many uncertainties built-
in, for instance "depending on available data", or "if feasible" is used a lot. Also, who will do
the actual assessment work, over what time period? What funds are available for the work?
Who is the audience?
Adopting a Longer View
From what I have looked at so far, I think it is safe to say that most of the environmentally
sensitive areas do not have adequate monitoring in place to track the changes that we will need to
quantify when the next review of the these pollutants become due in 5 -10 years. It seems
appropriate to make it a requirement that this REA make specific recommendations on what
needs to be in place to make the next assessment more adequately quantitative or populated with
data. We don't want to see more sophisticated models with the same poor data. We do want to
see a stronger data record of results on the ground.
This longer term view (5 to 10 year) is also an opportunity to examine in this REA what metrics
would be more useful that simply going with those that are currently available, e.g. the base
cation surplus is a better indicator than ANC or pH where DOC is changing like in the
Adirondacks.
Reliance on Modeling
A model like CMAQ certainly can produce estimates of dry deposition, but one of the drawbacks
is that there is no way to assess if these estimates are reasonable. For this reason, it probably
makes more sense to emphasize those measurements and model predictions that we have the
most confidence in, namely wet deposition. (Section 3.4.2, pages 46-47)
Using five years of model results certainly is better than relying on a single modeling year. Were
these five years modeled in a consistent fashion - same version of MM5, consistent emission
inventories, consistent CMAQ features/parameters, etc.? Have they been evaluated against all
available ambient concentration and deposition measurements, both for individual years and for
the entire five-year period? (Appendix C)
Data Certainty
Few, if any, of the references to existing data provide the years of data availability. The length
and current-ness of records are critical to any assessment especially because the last decade has
63
-------
been a time of significant emissions/deposition changes and ecosystem response signals have
been detected.
Missing from the document is a discussion (with maps) of the documented areas of the US where
sensitive ecosystems and high deposition rates of N and S have been identified (e.g. NAPAP
Biennial Report to Congress: An Integrated Assessment, May 1998, Figure 19, page 55 ).
PM& Ozone
I feel it is important to address particulate matter and ozone in this ISA as well as the PM and
ozone ISAs.
The omission of particulate phase NOX and SOX appears to limit the potential for setting
standards in the future using both PM and oxides of Nitrogen and Sulfur standards. Furthermore,
PM plays a significant role in nitrogen and sulfur deposition.
Also, separating out the effects of gas- versus aerosol-phase S/N will be difficult, since wet and
dry deposition can include both phases, and atmospheric chemistry and transport affect both
phases. The ISA clearly states that "particulate NOX and SOX will be addressed with the
secondary PM NAAQS review," and it therefore becomes crucial that these two review process
tracks are highly consistent with each other. One cannot proceed independently of the other
track.
Climate Change
Climate change discussion appears to be limited to N2O. I feel that the climate change issue may
be addressed more appropriately by including foliar injury data (USDA).
Section 2.3.1, Figure 2-1, page 10: Consider amending this figure. Climate change may affect
the "Characterization of Exposure" and "Characterization of Ecological Effects," so consider
putting arrows from "Climate Change" to these boxes. Suggest changing
"Climate Change (N2O)" slightly to "Climate Change (N2O, etc.)."
Charge Questions
The 7 step approach seems thorough, but I do not feel that the gaps and uncertainties are properly
described.
Identifying documented effects seems appropriate, but I feel that not all indicators are identified.
Specifically, foliar injury (USDA) and base cation surplus.
Specific Comments
Section 2.3, page 8: Why is the document disregarding damage to materials, including decay of
buildings, statues, and sculptures that are part of our national heritage this time around?
64
-------
Section 2.3.2, pages 11-13: Although N2O is included in this review process, it is not part of
NOy. Nor does it play a role in N deposition, so while it could be included elsewhere in this
document it does not really belong in this section.
Section 2.4, last bullet on page 17: Is it necessary to be able to distinguish NOx from total
reactive N? Isn't the purpose of this to examine all oxides of N?
Section 3.1.1, page 28: add 'base cation surplus' to Table 3-2. Key indicators of nutrient
enrichment. This first paragraph is about deposition and applies generically to both acidification
and nutrient enrichment issues.
Section 3.3, page 36-37: Table 3-5 More detail needed on the indicator data. Add dates of the
mapping layers, whether point or polygon data, the number of data points or polygons in the
study area (maybe the whole US), and the resolution of the base data.
Section 3.3, page 38: I feel that including the Catskills of New York will provide important
information and data.
Section 3.4.1, page 45: Selection of ANC as an indicator. Have all sensitive regions and case
study areas been evaluated independently for selection of best indicators? Just because ANC is
most widely available, it may not necessarily be the best, e.g. see discussion on using base cation
surplus for Adirondack streams in the ISA. Also the three first bullets of indicators are not really
indicators (see page A-4 paragraph for indicators).
Section 3.4.2, page 47: The maps are too small.
There are many acronyms throughout this document, and not all of them are commonly used.
Please include a listing of these.
65
-------
Dr. Kathleen Weathers
This is a useful document for catalyzing discussion and getting feedback from the committee.
The approach relies almost exclusively on process-based and spatial modeling analysis. A short
discussion of whatif anyother types of analytical and assessment tools are available would
be useful.
Specific and general comments, questions and suggestions are listed below.
Section 2.3.1
I applaud the use of a conceptual model and framework.
Explicitly missing from the conceptual framework, however, is the key link (flow arrow)
between concentrations and deposition. Also, the significance of the arrows and boxes should be
identified (e.g., what's the distinction between the dotted and solid lines surrounding boxes. And
the arrows?
What's the difference between ecosystem effects and ecosystem responses, or services (in
boxes), for that matter?
I think that it would be helpful to define general ecosystem functions (e.g., productivity, nutrient
cycling, etc.) and then point to places where the deposition of S and N has been demonstrated to
alter them.
Page 1: Consider altering the sentence beginning with "Our plan..." to: "...and the levels of
deposition," and the environmental effects on ecosystems of this deposited material"
Page 1: "Along with these case studies, we plan to conduct statistical and spatial
characterizations..." Are these analyses intended to be new work/research?
Section 2.1: The legislative mandate reads to me as the request for determination of critical
loads, albeit within a narrow range.
Section 2.2, page 6 awkward phrase: ....importance of acid neutralizing capacity in surface water
acidification.
Page 8, 2nd para: tropospheric ozone
Page 8, last para: "Against this broad background..." awkward and, in places incorrect.
Page 9, "In this current review...appropriateness of NO2 as an indicator..." I'm not sure what this
means.
66
-------
Words such as harmful (page 11) should be modified by ecological or environmental.
The consideration of critical load analysis as an ecosystem-type specific risk may be more useful
than considering it as a site-specific risk.
Page 11, the paragraph beginning with "Each component of the framework..." isn't clear.
Fig 2-2 might note explicitly the formation of ammonium sulfate, since both N and S are of
interest here.
2.3.3. The first couple of sentences are awkward. Also, acidification of the environment should
include lowering the natural pH of rain, water bodies and/or terrestrial ecosystems.
2.3.4: are the deposition of NOx and SOx meant to infer the deposition of nitric and sulfuric
acids? If not the second sentence is not right.
2.3.4: Be specific, for example: "Nitrogen and sulfur enrichment represents a continuum of
effects...and it can be characterized as positive (increase in environmental parameter y with
enrichment, or additional N or S, or with incremental additions of N or S) or negative (decrease
in environmental parameter y with increase in N or S, etc)
Say clearly why methylmercury is being considered here vs other linked biogeochemical
interactions (the effect of sulfur on phosphorous release/availability in aquatic systems, for
example).
Figure 2.5: consider depicting feedbacks and linked processes
Section 2.4:
Clarify..."rather than on the effects of aerosol NOx and SOx that remain in the atmosphere."
Page 17, second bullet: responses = effects? and variability is meant to be spatial and temporal,
correct?
Page 18, second set of bullets:
Identifying important chemical species in the atmosphere? Important means relative to other
species, because they can cause environmental effects? Over space and time?
Page 18, last bullet: Changes in land use might be added to the meteorologic and climate
considerations
Section 3: Seven-step approach
The 7-step approach is appealing. It uses both spatial and mechanistic models to explore inputs
of N and S across the US as well as the effects of N and S deposition. I remain concerned,
however, that (1) the data that are necessary for many of the analyses do not exist or (2) available
data are not at an appropriate spatial or temporal scale and/or cannot readily or reasonably be
67
-------
linked together or extrapolated using geostatistical or other tools. (3) Finally, I cannot tell
whether or how experimental data, for example N or S or combined N and S additions would be
used to inform this analysis.
Since both the use of biogeochemical process models in new geographic locations as well as the
creation of GIS-based models and analysis are active areas of scientific research, to what extent
is it necessary that these model analyses and extrapolations appear in peer reviewed publication
before they can be effectively used in this assessment? In addition, the steps that will be used to
validate or verify model results should be identified.
Under Key Indicators of Nutrient Enrichment: What's the basis of the suggestion that wet-
deposition monitoring stations can provide more... an extensive range of nitrogen species than is
possible for dry-deposition monitoring stations?
As mentioned in my review of the science assessment, throughfall N may not reflect N
deposition levels. Also, C:N in soils have been related to nitrate in surface waters draining
watersheds.
Table 3-3: Technically, there aren't "dry deposition monitoring" networks, rather there exist air
concentration monitoring networks and those data are used to estimate dry deposition.
See comment above about indicators and throughfall. Sulfur has been used successfully as an
index of wet+dry deposition in the eastern US.
Recovery should be defined clearly and carefully for the purposes of this assessment.
What's the operative definition of a multi-media model?
Step 5 (Section 3.5):
The spatial extrapolation of case study areas is an interesting approach; my answer to whether I
agree with the approach depends upon answers to several questions. It is not clear to me the basis
upon which case study areas could be scaled, or whether the interpolation techniques alluded to
are appropriate spatial extrapolation tools. Presumably deposition output from the CMAQ model
will be mapped on to the sensitive areas, for instance. Are the spatial scales comparable?
Step 6 (Section 3.6):
The goal of creating response curves needs more explanation. Is the goal to identify thresholds,
or identify inflection points? What's a desired exposure endpoint? How will, or how could,
published loading experiments be used to help identify response curves?
An example of "quantifying the link between changes in ecosystem indicators and ecosystem
services" would make easier a discussion on the utility of this approach.
68
-------
I concur with the idea of using the Millennium Ecosystem Assessment's definition of ecosystem
services.
I think that parts of the mapping steps identified on page 55 are a good start:
It makes sense to me that the first data layer should be a total (wet + dry) S and N deposition
map, apparently generated from CMAQ.
The acidity impact datalayer is presumably derived, but from which data or datalayers?
And, from here on out with the mapping exercises, the devil is in the details. For example, what
are the common selection criteria? How will the prevalent ecosystem services be identified? A
highlighted example from the Chan et al. 2006 paper would aid this discussion.
The valuation of ecosystem services is out of the realm of my direct expertise, but it seems to me
that there is always the danger of undervaluing ecosystem services, especially when considering
biogeochemical cycling of multiple, interacting elements.
Step 7 (Section 3.7):
Based on the heterogeneity in S and N loading, ecosystem sensitivity and responses across the
United States, I think it very important to include as part of the secondary NAAQS a "form that
allows for consideration of regional heterogeneity."
Resilience (and recovery) will need to be defined.
The need to relate ambient concentrations with actual deposition loads appears again in this
section.
In the list of associated issues, "If total nitrogen..." concentration, content in what?
I agree that it will be most fruitful to do intensive analyses on smaller regions.
Will the proposed analyses be done on watersheds? Or on ecosystems with some other boundary
and if so, what and why? It would be useful to define boundaries for these assessment areas.
Many, if not all, of the 9 steps entail significant analysis (which is evident to those who will be
working on this document, of course). Are all steps equally important to come to policy relevant
conclusions? Is it possible to identify what the most "sensitive" steps are, meaning which ones
have the greatest influence on the policy-relevant conclusions. What is the relative importance of
assessing the impacts of meteorological variability on loadings and exposures across the US
(presumably) vs analyzing the relationships between NOx, SOx and other reactive forms of N (in
the atmosphere?)? My assessment of the approach depends, in part, on answers to these
questions.
69
------- |