1
 2
   &EPA
United! S*K
Enviraimerilfll PiutmBmi
Agamy
7  Policy Assessment for the Review of the
s  Secondary National Ambient Air Quality
9  Standards for NOX and SOX
10
11
12  First External Review Draft
13

-------
 1                                                          EPA-452/P-10-006
 2                                                               March 2010
 O
 4
 5
 6
 7
 8   Policy Assessment for the Review of the

 9   Secondary National Ambient Air Quality Standards

10   for NOX and SOX:
11

12

13   First External Review Draft
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37                        U.S. Environmental Protection Agency
3 8                            Office of Air and Radiation
39                      Office of Air Quality Planning and Standards
40                      Health and Environmental Impacts Division
41                     Research Triangle Park, North Carolina 27711
42
43
44

-------
 1                                       DISCLAIMER
 2
 O
 4          This document has been reviewed by the Office of Air Quality Planning and Standards,
 5   U.S. Environmental Protection Agency (EPA), and approved for publication. This draft
 6   document has been prepared by staff from the Office of Air Quality Planning and Standards,
 7   U.S. Environmental Protection Agency. Any opinions, findings, conclusions, or
 8   recommendations are those of the authors and do not necessarily reflect the views of the EPA.
 9   Mention of trade names or commercial products is not intended to constitute endorsement or
10   recommendation for use. This document is being provided to the Clean Air Scientific Advisory
11   Committee for their review, and made available to the public for comment. Any questions or
12   comments concerning this document should be addressed to Dr. Bryan Hubbell, U.S.
13   Environmental Protection Agency, Office of Air Quality Planning and Standards, C504-02,
14   Research Triangle Park, North Carolina 27711 (email: hubbell.bryan@epa.gov ).
15

-------
 1                                  TABLE OF CONTENTS

 2    List of Figures	iv
 3    List of Tables	vii
 4    List of Acronyms and Abbreviations	ix
 5    List of Key Terms	xii
 6    1.  Introduction	1
 7       1.1  Definitions of NOX and SOX for this Assessment	3
 8       1.2  Policy Objectives	4
 9       1.3  Critical Policy Elements	6
10       1.4  Historical Context	8
11            1.4.1  History of NOX and SOXNAAQS Review	8
12            1.4.2  History of Related Assessments and Agency Actions	10
13       1.5  Proposed Conceptual Framework for Combined NOX SOX Standards	13
14       1.6  Policy Relevant Questions	16
15    2.  Known or anticipated ecological effects	22
16       2.1  Acidification: Evidence of effects on structure and function of terrestrial and
17            freshwater ecosystems	23
18            2.1.1  What is the nature of acidification related ecosystem responses to
19                   reactive nitrogen and/ sulfur deposition?	24
20            2.1.2  What types of ecosystems are sensitive to such effects? In which ways
21                   are these responses affected by atmospheric, ecological, and landscape
22                   factors?	26
23            2.1.3  What is the magnitude of ecosystem responses to acidifying deposition?	26
24            2.1.4  What are the key uncertainties associated with acidification?	35
25       2.2  Nitrogen enrichment: Evidence of effects on structure and function of terrestrial
26            and freshwater ecosystems	37
27            2.2.1  What is the nature of terrestrial and freshwater ecosystem responses to
28                   reactive nitrogen and/ sulfur deposition?	37
29            2.2.2  What types of ecosystems are sensitive to such effects? How are these
30                   responses affected by atmospheric, ecological, and landscape factors	39
31            2.2.3  What is the magnitude of ecosystem responses to nitrogen deposition?	40
32            2.2.4  What are the key uncertainties associated with nutrient enrichment?	48
33       2.3  What Ecological effects are associated with gas-phase NOX  and SOX?	49
34            2.3.1  What is the nature of ecosystem responses to gas-phase nitrogen and
35                   sulfur?	50
36            2.3.2  What types of ecosystems are sensitive to such effects? How are these
37                   responses affected by atmospheric, ecological, and landscape factors?	50
38            2.3.3  What is the magnitude of ecosystem responses to gas phase effects of
39                  NOxandSOx?	51
40       2.4  Summary	51
41       2.5  References	52
      March 2010                                 i                 Draft - Do Not Quote or Cite

-------
 1    3.  Considerations of Adversity to Public Welfare	62
 2       3.1  How do we characterize adversity to public welfare? What are the relevant
 3            factors and how are they addressed in this document?	62
 4            3.1.1   What are the benchmarks for adversity from other sources?	62
 5            3.1.2   Other EPA Programs and Federal Agencies	65
 6       3.2  What are ecosystem services and how does this concept relate to public
 7            welfare?	69
 8       3.3  What is the role of economics?	75
 9       3.4  What is the evidence for effects  on ecosystem services? How do we link
10            ecological indicators to services?	78
11       3.5  References	89
12    4  Addressing the Adequacy of the Current Standards	101
13       4.1  Are the structures of the current NOX and SOX secondary standards based on
14            relevant ecological indicators such that they are adequate to determine and
15            protect public welfare against adverse effects on ecosystems?	101
16       4.2  To what extent are the structures of the current NOX and SOX secondary
17            standards meaningfully related to relevant ecological indicators of public
18            welfare effects?	103
19       4.3  To what extent do current monitoring networks provide a sufficient basis for
20            determining the adequacy of current secondary NOX and SOX standards?	106
21            4.3.1   What does the NADP monitoring network provide and what are the
22                   major limitations?	Ill
23            4.3.2   How do we characterize  deposition through Monitoring and Models?	112
24       4.4  What is our best characterization of atmospheric concentrations of NOy and
25            SOX, and deposition of N and S?	114
26            4.4.1   What are the current atmospheric concentrations of reactive nitrogen,
27                   NOy, reduced nitrogen, NHX, sulfur dioxide, SO2, and sulfate, SO4?	115
28       4.5  Are adverse effects on the public welfare occurring under current air quality
29            conditions for NO2 and 802 and  would they occur if the nation met the current
30            secondary standards?	130
31            4.5.1   To what extent do the current NOX and SOX secondary  standards provide
32                   protection from adverse effects associated with deposition of
33                   atmospheric NOX, and SOX which results in acidification in sensitive
34                   aquatic and terrestrial ecosystems?	133
35            4.5.2   To what extent does the current NOX secondary standard provide
36                   protection from adverse effects associated with deposition of
37                   atmospheric NOX, which results in nutrient enrichment effects in
38                   sensitive aquatic and  terrestrial ecosystems?	138
39            4.5.3   Aquatic Nutrient Enrichment	139
40            4.5.4   Terrestrial Nutrient Enrichment	141
41       4.6  To what extent do the current NOX and/or SOX secondary standards provide
42            protection from other ecological effects (e.g., mercury methylation) associated
43            with the deposition of atmospheric NOX, and/or SOX?	142
44       4.7  References	143
45    5.  Conceptual Design of an Ecologically Relevant Multi-pollutant Standard	145
46       5.1  Components of the design	145

      March 2010                                ii                  Draft - Do No Quote or Cite

-------
 1            5.1.1   For which effects is there sufficient information to support setting
 2                   standards?	146
 3       5.2  Ecological Components of the Standard: Aquatic Acidification	147
 4            5.2.1   Conceptual design  considerations from the ISA and REA	149
 5            5.2.2   Design options for  aquatic acidification	157
 6       5.3  Ecological Components of the Standard: Terrestrial Acidification, Terrestrial
 7            Nutrient Enrichment and Surface water Nutrient Enrichment	167
 8            5.3.1   Terrestrial Acidification	167
 9            5.3.2   Terrestrial and surface water nutrient enrichment	168
10            5.3.3   Summary	169
11       5.4  Linking Deposition to Atmospheric Concentration	169
12            5.4.1   Background	169
13            5.4.2   Aggregation Issues	170
14            5.4.3   AirQuality Simulation Models	171
15            5.4.4   Oxidized Sulfur and Nitrogen Pollutant Species	172
16            5.4.5   Example Calculations	173
17       5.5  Example calculation  for the conceptual  design and derivation of AAPI	177
18            5.5.1   Example calculation for the conceptual design	177
19            5.5.2   Derivation of the Atmospheric Acidification Potential Index (AAPI):	185
20       5.6  References	188
21    6.  Options for Elements of the Standard	190
22       6.1  What atmospheric indicators of oxidized nitrogen and sulfur are appropriate for
23            use in a secondary NAAQS that provides protection for public welfare from
24            exposure related to deposition of N and S? What averaging times and statistics
25            for such indicators are appropriate to consider?	191
26       6.2  What is the appropriate averaging time for the air quality indicators NOy and
27            SOX to provide protection of public welfare from adverse effects from
28            acidification?	193
29       6.3  What form(s) of the standard are most appropriate to provide protection of
30            sensitive ecosystems from the effects of acidifying deposition related to ambient
31            NOX and SOX concentrations?	194
32       6.4  What are the appropriate spatial extents of the boundaries for evaluating AAPI?
33            Within those boundaries, what are the appropriate statistics to use in calculating
34            the parameters of the AAPI,  e.g. G, VNoy, Vs, and NHX? Within those
35            boundaries, what s the appropriate spatial averaging for the air quality indicators
36            NOy and SOX to provide protection of public welfare from adverse effects from
37            acidification?	203
38       6.5  What are the options for specifying the targets for the ecological indicator for
39            aquatic acidification?	203
40            6.5.1   What levels of impairment are related to  alternative levels of ANC?	204
41       6.6  What are the appropriate ambient air monitoring methods to consider in
42            developing the standards?	208
43            6.6.1   What measurements would be used to characterize NOy and SOX
44                   ambient air concentrations for the purposes of the AAPI based standard?	208
45            6.6.2   What sampling frequency would be required?	208
      March 2010                                iii                 Draft - Do No Quote or Cite

-------
 1            6.6.3   What are the spatial scale issues associated with monitoring for
 2                   compliance, and how should these be addressed?	209
 3       6.7  Taking into consideration information about ecosystem services and other
 4            factors related to characterizing adversity for the ecological effects being
 5            assessed in this review, what is an appropriate range of alternative standards for
 6            the Agency to consider?	210
 7    7.  Co-protection for Other Effects Using Standards to Protect Against Acidification	213
 8       7.1  To what extent would a standard specifically defined to protect against aquatic
 9            acidification likely provide protection from terrestrial acidification?	213
10       7.2  To what extent would a standard specifically defined to protect against aquatic
11            acidification likely provide protection from terrestrial nutrient enrichment?	214
12       7.3  To what extent would a standard specifically defined to protect against aquatic
13            acidification likely provide protection from aquatic nutrient enrichment?	215
14    8.  Consideration of Issues Regarding Reduced and Oxidized Forms of Nitrogen	216
15       9.1  Conclusions	219
16       9.2  Summary of key uncertainties and research recommendations related to setting
17            a secondary standard forNOx and SOX	223
18            9.2.1   Research Needs to Reduce Uncertainty in the Next Review (focused on
19                   aquatic acidification)	223
20            9.2.2   Data Needs to Reduce Uncertainty in the Next Review (focused on
21                   aquatic acidification)	223
22

23                                    LIST OF FIGURES

24    Figure 1-1. Framework of an alternative secondary standard	16
25    Figure 2-1. Ecological Effects Associated with Alternative Levels of Acid Neutralizing
26                   Capacity (ANC)	28
27    Figure 2-2. Average NOs" concentrations (orange), SO42" concentrations (red), and ANC
28                   (blue) across the 44 lakes in the Adirondack Case Study Area modeled
29                   using MAGIC for the period 1850 to 2050	29
30    Figure 2-3. ANC concentrations of preacidification (1860) and current (2006) conditions
31                   based on hindcasts of 44 lakes in the Adirondack Case Study Area
32                   modeled using MAGIC.  [Note: in this map, the symbol for red is
33                   reversed and should be < 0. The figure will be revised in the next draft.]	30
34    Figure 2-4. Critical loads of acidifying deposition that each surface water location can
35                   receive in the Adirondack Case Study Area while maintaining or
36                   exceeding an ANC concentration of 50 ueq/L based on 2002 data.
37                   Watersheds with critical load values <100 meq/m2/yr (red and orange
38                   circles) are most sensitive to surface water acidification, whereas
39                   watersheds with values >100 meq/m2/yr (yellow and green circles) are
40                   the least sensitive sites	31
41    Figure 2-5. Average NOs" concentrations orange), SO42"concentrations (red), and ANC
42                   (blue) levels for the 60 streams in the Shenandoah Case Study Area
43                   modeled using MAGIC for the period 1850 to 2050	32


      March 2010                                iv                 Draft - Do No Quote or Cite

-------
 1    Figure 2-6. ANC levels of 1860 (preacidification) and 2006 (current) conditions based on
 2                   hindcasts of 60 streams in the Shenandoah Case Study Area modeled
 3                   using MAGIC	33
 4    Figure 2-7. Critical loads of surface water acidity for an ANC of 50 [j,eq/L for
 5                   Shenandoah Case Study Area streams. Each dot represents an estimated
 6                   amount of acidifying deposition (i.e., critical load) that each stream's
 7                   watershed can receive and still maintain a surface water ANC >50 ueq/L.
 8                   Watersheds with critical load values <100 meq/m2/yr (red and orange
 9                   circles) are most sensitive to surface water acidification, whereas
10                   watersheds with values >100 meq/m2/yr (yellow and green circles) are
11                   the least sensitive sites	34
12    Figure 2-8. Benchmarks of atmospheric nitrogen deposition for several ecosystem
13                   indicators with the inclusion of the diatom changes in the Rocky
14                   Mountain lakes (REA 5.3.1.2)	42
15    Figure 2-9 (from REA figure 5.3-9). Observed effects from ambient and experimental
16                   atmospheric nitrogen deposition loads in relation to using CMAQ 2002
17                   modeling results and NADP monitoring data. Citations for effect results
18                   are from the ISA, Table 4.4 (U.S. EPA, 2008)	43
19    Figure 3-1. Common anthropogenic  stressors and the essential ecological attributes they
20                   affect. Modified from Young and Sanzone (2002)	64
21    Figure 3-2. Representation of the benefits assessment process indicating where some
22                   ecological benefits may remain unrecognized, unquantified, or
23                   unmonetized. (Source: EBASP USEPA 2006)	71
24    Figure 3-3. Conceptual model showing the relationships among ambient air quality
25                   indicators and exposure pathways and the resulting impacts on
26                   ecosystems, ecological responses, effects and benefits to characterize
27                   known or anticipated adverse effects to public welfare. [This figure to be
28                   revised for Second Draft Policy Assessment Document]	73
29    Figure 3-4. Locations of Eastern U.S. National Parks (Class I areas) relative to deposition
30                   of Nitrogen and Sulfur in sensitive aquatic areas	74
31    Figure 3-5. Location of Western U.S. National Parks (Class I areas) relative to deposition
32                   of Nitrogen and Sulfur	75
33    Figure 3-6. Conceptual model linking ecological indicator (ANC) to affected ecosystem
34                   services	79
35    Figure 4-1. Routinely operating surface monitoring stations measuring forms of
36                   atmospheric nitrogen	107
37    Figure 4-2. Routinely operating surface monitoring stations measuring forms of
38                   atmospheric sulfur	108
39    Figure 4-3. Anticipated network of surface based NOy stations based on 2009 network
40                   design plans. The NCore stations are scheduled to be operating by
41                   January, 2011	110
42    Figure 4-4. Location of approximately 250 National Atmospheric Deposition Monitoring
43                   (NADP) National Trends Network (NTN) sites illustrating annual
44                   ammonium deposition for 2005. Weekly values of precipitation based
45                   nitrate, sulfate and ammonium are provided by NADP	112

      March  2010                                v                 Draft - Do No Quote or Cite

-------
 1   Figure 4-5. 2005 CMAQ modeled annual average NOy (ppb). These maps will be
 2                  replaced with full CONUS maps in the next draft	117
 3   Figure 4-6. 2005 CMAQ modeled annual average total reduced nitrogen (NHX) (as ng/m3
 4                  nitrogen)	118
 5   Figure 4-7. 2005 CMAQ modeled annual average ammonia, NHs, (as ng/m3 N)	119
 6   Figure 4-8. 2005 CMAQ modeled annual average ammonia, NH4, (as ng/m3 N)	120
 7   Figure 4-9. 2005 CMAQ modeled annual average SOX, (as ng/m3 S from SO2 and SO4)	121
 8   Figure 4-10. 2005 CMAQ modeled annual average SO2 (as ng/m3 S)	122
 9   Figure 4-11. 2005 CMAQ modeled annual average SO4 (as ng/m3 S)	123
10   Figure 4-12. 2005 annual average sulfur dioxide concentrations based on CASTNET
11                  generated by the Visibility Information Exchange Web Sysytem
12                  (VIEWS)	124
13   Figure 4-13. 2005 annual average sulfate concentrations based on CASTNET generated
14                  by the Visibility Information Exchange Web Sysytem (VIEWS)	124
15   Figure 4-14. Annual average 2005 NOy concentrations from reporting stations in AQS	125
16   Figure 4-15. 2005 CMAQ modeled Oxidized Nitrogen Deposition (kgN/Ha/Yr)	126
17   Figure 4-16. 2005 CMAQ modeled Oxidized Sulfur Deposition (kgS/Ha/Yr)	127
18   Figure 4-17. Three hour average maximum 2005 SO2 concentrations based on the
19                  SLAMS reporting to EPA's Air Quality System (AQS) data base. The
20                  current SO2 secondary standard based on a the maximum 3 hour average
21                  value is 500 ppb, a value not exceeded. While there are obvious spatial
22                  gaps, the majority of these stations are located to capture maximum
23                  values generally in proximity to major sources and high populations.
24                  Lower relative values are expected in more remote acid sensitive areas	128
25   Figure 4-18. Annual average 2005 NO2 concentrations based on the SLAMS reporting to
26                  EPA's Air Quality System (AQS) data base. The current NO2 secondary
27                  standard is 53 ppb, a value well above those observed. While there are
28                  obvious spatial gaps, the stations are located in areas of relatively high
29                  concentrations in highly populated areas. Lower relative values are
30                  expected in more remote acid sensitive areas	129
31   Figure 4-19. 2005 CMAQ derived annual average ratio of (NOy - NO2)/NOy. The
32                  fraction of NO2 contributing to total NOy generally is less than 50% in
33                  the Adirondack and Shenandoah case study areas. The ratio reflects the
34                  relative air mass aging associated with transformation of oxidized
35                  nitrogen beyond NO and NO2 as one moves from urban to rural
36                  locations	130
37   Figure 4-20. National map highlighting the 9 case study areas evaluated in the REA	133
38   Fig 5-1. Schematic diagram of the conceptual design of the standard	146
39   Fig 5-2. Schematic diagram of the conceptual design of the standard based on aquatic
40                  acidification. From left to right, if a desired level of ANC is known, then
41                  the concentration of the atmospheric indicators that will cause that level
42                  may be calculated. From right to left,  if the if the concentration of the air
43                  quality indicators are known than the  ANC that will be caused may be
44                  calculated	148

     March 2010                               vi                 Draft - Do No Quote or Cite

-------
 1    Figure 5-3. The depositional load function	158
 2    Fig 5-4. A map of acid sensitive areas of the Eastern U.S. developed from a lithology-
 3                  based five-unit geologic classification system after methods in Sullivan
 4                  etal. (2007)	163
 5    Figure 5-5. VS/N values for each grid cell in the eastern (right) and western (left) U.S.
 6                  domains. The top maps are for sulfur and the bottom are for nitrogen	174
 7    Figure 5-6. Schematic Diagram illustrating the procedure for converting deposition
 8                  tradeoff curves of sulfur and nitrogen to atmospheric concentrations of
 9                  SOxandNOx	175
10    Figure 5-7. Inter-annual coefficients of variation (CV) of a) nitrogen and b) sulfur VS/N
11                  values, based on a series of 2002-2005 CMAQ v4.7 simulation	176
12    Figure 5-8. Tradeoff curve for S and N deposition to protect from aquatic acidification in
13                  the Adirondacks using Neco equation 2	181
14    Figure 5-9. Tradeoff curve for S and N deposition to protect from aquatic acidification in
15                  the Adirondacks using Neco equation 3	181
16    Figure 5-10. Tradeoff curve for S and NOy deposition to protect from aquatic
17                  acidification in the Adirondacks using Neco equation 2	183
18    Figure 5-11. Tradeoff curve for S and NOy deposition to protect from aquatic
19                  acidification in the Adirondacks using Neco equations	183
20    Figure 5-12. Tradeoff curve for atmospheric concentration of SOX and NOy to protect
21                  from aquatic acidification in the Adirondacks using Neco equation 2	184
22    Figure 5-13. Tradeoff curve for atmospheric concentration of SOX and NOy to protect
23                  from aquatic acidification in the Adirondacks using Neco equation 3	185
24    Figure 6-1. Ecosystems sensitive to acidifying deposition in the Eastern U.S. (Note that
25                  Florida represents a special case where high levels of natural
26                  acidification exist unrelated to deposition) This map does not include all
27                  sensitive areas in the U.S. Certain mountainous areas of the Western
28                  U.S. are also sensitive to acidifying deposition	202
29    Figure 6-2. Number offish species per lake or stream versus ANC level and aquatic
30                  status category (colored regions) for lakes in the Adirondack Case Study
31                  Area (Sullivan et al., 2006)	206
32

33                                     LIST OF TABLES

34    Table 3-1. Crosswalk between Ecosystem Services and Public Welfare Effects	70
35    Table 5-1. Illustration of how selected models and water chemistry data were used to
36                  calculate critical loads in the REA	151
37    Table 5-2. Summary of the ecological components of design option 1	166
38    Table 5-3. Oxidized sulfur and nitrogen species currently available in CMAQ
39                  simulations. Note that PNA concentrations are not available in current
40                  CMAQ extractions	174
41    Table 5-4. Example Calculations for Determining the Percent of Water Bodies Achieving
42                  Target ANC Levels	180


      March 2010                               vii                 Draft - Do No Quote or Cite

-------
 1   Table 5-5. Values for N and S deposition tradeoff curves for ANC = 50, protecting 32
 2                  and 50% of the population, in Adirondacks case study area as illustrated
 3                  on Fig 5-8 and Fig 5-9. Units are in meq/m2/yr unless noted otherwise	180
 4   Table 5-6. Values for NOy and S deposition tradeoff curves for ANC = 50, protecting 32
 5                  and 50% of the population in Adirondacks case study area as illustrated
 6                  on Fig 5.10 and Fig 5.11. Units are in meq/m2/yr unless noted otherwise	182
 7   Table 7-1. Results of comparing aquatic ANC50 critical loads to average terrestrial
 8                  watershed area Bc:Al ratios. Left numbers in each column are the
 9                  number of lakes or streams that had a lower critical load than the
10                  terrestrial calculated critical load. Right numbers in each column are the
11                  number of lakes that had a higher critical load than the watershed
12                  calculated terrestrial critical loads	214
13
14
15
     March 2010                               viii                 Draft - Do No Quote or Cite

-------
            LIST OF ACRONYMS AND ABBREVIATIONS
2
O
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
AAPI
ADR
A13+
ANC
AQCD
AQRV
ASSETS El
Bc/Al
C
Ca/Al
n 2+
Ca
CAA
CASAC
CASTNet
CCS
Chi a
CLE
CMAQ
CSS
CWA
DIN
DO
DOT
EMAP
EPA
FHWAR
FIA
FWS
GIS
GPP
H+
H2O
H2SO4
ha
HAB
HFC
Hg+2
Hg°
HNO3
HONO
HUC
IMPROVE
ISA
K+
Atmospheric Acidification Potential Index
Adirondack Mountains of New York
aluminum
acid neutralizing capacity
Air Quality Criteria Document
air quality related values
Assessment of Estuarine Trophic Status eutrophi cation index
Base cation to aluminum ratio, also Be: Al
carbon
calcium to aluminum ratio
calcium
Clean Air Act
Clean Air Scientific Advisory Committee
Clean Air Status and Trends Network
coastal sage scrub
chlorophyll a
critical load exceedance
Community Multiscale Air Quality model
coastal sage scrub
Clean Water Act
dissolved inorganic nitrogen
dissolved oxygen
U.S. Department of Interior
Environmental Monitoring and Assessment Program
U.S. Environmental Protection Agency
fishing, hunting and wildlife associated recreation survey
Forest Inventory and Analysis National Program
Fish and Wildlife Service
geographic information systems
gross primary productivity
hydrogen ion
water vapor
sulfuric acid
hectare
harmful algal bloom
hydrofluorocarbon
reactive mercury
elemental mercury
nitric acid
nitrous acid
hydrologic unit code
Interagency Monitoring of Protected Visual Environments
Integrated Science Assessment
potassium
March 2010
                               IX
Draft - Do No Quote or Cite

-------
1
2
O
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
kg/ha/yr
km
LRMP
LTER
LTM
MAGIC
MCF
MEA
Mg2+
N
N2
N2O
N203
N204
N205
Na+
NAAQS
NADP
NAPAP
NAWQA
NEEA
NEP
NH3
NH4+
(NH4)2SO4
NHX
NO
NO2
NO2-
MV
NOAA
NOX
NOy
NPP
NFS
NRC
NSWS
NTN
NTR
03
OAQPS
OW
PAN
PFC
pH
ppb
ppm
kilograms per hectare per year
kilometer
Land and Resource Management Plan
Long Term Ecological Monitoring and Research
Long-Term Monitoring
Model of Acidification of Groundwater in Catchments
Mixed Conifer Forest
Millennium Ecosystem Assessment
magnesium
nitrogen
gaseous nitrogen
nitrous oxide
nitrogen trioxide
nitrogen tetr oxide
dinitrogen pentoxide
sodium
National Ambient Air Quality Standards
National Atmospheric Deposition Program
National Acid Precipitation Assessment Program
National Water Quality Assessment
National Estuarine Eutrophi cation Assessment
net ecosystem productivity
ammonia gas
ammonium ion
ammonium sulfate
category label for NH3 plus NH4+
nitric oxide
nitrogen dioxide
reduced nitrite
reduced nitrate
National Oceanic and Atmospheric Administration
nitrogen oxides
total oxidized nitrogen
net primary productivity
National Park Service
National Research Council
National Surface Water Survey
National Trends Network
organic nitrate
ozone
Office of Air Quality Planning and Standards
Office of Water
peroxyacyl nitrates
perfluorocarbons
relative acidity
parts per billion
parts per million
March 2010
Draft - Do No Quote or Cite

-------
1
2
O
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
ppt
PSD
REA
REMAP
S
S203
S207
SAV
SF6
SMP
SO
S02
SO3
so32-
SO4
SO42"
SOM
sox
SPARROW
SRB
STORE!
TIME
TMDL
TP
USFS
USGS
ueq/L
Hg/m3

                      parts per trillion
                      prevention of significant deterioration
                      Risk and Exposure Assessment
                      Regional Environmental Monitoring and Assessment Program
                      sulfur
                      thiosulfate
                      heptoxide
                      submerged aquatic vegetation
                      sulfur hexafluoride
                      Simple Mass Balance
                      sulfur monoxide
                      sulfur dioxide
                      sulfur trioxide
                      sulfite
                      wet sulfate
                      sulfate ion
                      soil organic matter
                      sulfur oxides
                      SPAtially Referenced Regressions on Watershed Attributes
                      sulfate-reducing bacteria
                      STORage and RETrieval
                      Temporally Integrated Monitoring of Ecosystems
                      total maximum daily load
                      total phosphorus
                      U.S. Forest Service
                      U.S. Geological Survey
                      microequivalents per liter
                      micrograms per cubic meter
30
      March  2010
                                                 XI
Draft - Do No Quote or Cite

-------
 1                                  LIST OF KEY TERMS

 2    Acidification: The process of increasing the acidity of a system (e.g., lake, stream, forest soil).
 3          Atmospheric deposition of acidic or acidifying compounds can acidify lakes, streams,
 4          and forest soils.
 5    Air Quality Indicator: The substance or set of substances (e.g., PM2.5, NC>2, 862) occurring in
 6          the ambient air for which the National Ambient Air Quality Standards set a standard level
 7          and monitoring occurs.
 8    Alpine: The biogeographic zone made up of slopes above the tree line, characterized by the
 9          presence of rosette-forming herbaceous plants and low, shrubby, slow-growing woody
10          plants.
11    Acid Neutralizing Capacity: A key indicator of the ability of water to neutralize the acid or
12          acidifying inputs it receives. This ability depends largely on associated biogeophysical
13          characteristics, such as underlying geology, base cation concentrations, and weathering
14          rates.
15    Arid Region: A land region of low rainfall, where "low" is widely accepted to be less than 250
16          mm precipitation per year.
17    Base Cation Saturation:  The degree to which soil cation exchange sites are occupied with base
18          cations (e.g., Ca2+, Mg2+, K+) as opposed to A13+ and H+. Base cation saturation is a
19          measure of soil acidification, with lower values being more acidic.  There is a threshold
20          whereby soils with base saturations less than 20% (especially between 10%-20%) are
21          extremely sensitive to change.
22    Ecologically Relevant Indicator: A physical, chemical, or biological entity/feature that
23          demonstrates a consistent degree of response to a given level of stressor exposure and
24          that is easily measured/quantified to make  it a useful predictor of ecological risk.
25    Critical Load: A quantitative estimate of an exposure to one or more pollutants, below which
26          significant (as defined by the analyst or decision maker) harmful effects on specified
27          sensitive elements of the environment do not occur, according to present knowledge.
28    Denitrification: The anaerobic reduction of oxidized nitrogen (e.g., nitrate or nitrite) to gaseous
29          nitrogen (e.g., N2O or N2) by denitrifying bacteria.
30    Dry Deposition: The removal of gases and particles from the atmosphere to surfaces in the
31          absence of precipitation (e.g., rain, snow) or occult deposition (e.g., fog).
32    Ecological Risk: The likelihood that adverse ecological effects may occur or are occurring as a
33          result of exposure to one or more stressors (U.S. EPA, 1992).
34    Ecological Risk Assessment: A process that evaluates the likelihood that adverse ecological
35          effects may occur or are occurring as a result of exposure to one or more stressors (U.S.
36          EPA, 1992).
37    Ecosystem: The interactive system formed from all living organisms and their abiotic (i.e.,
38          physical and chemical) environment within a given area. Ecosystems cover a hierarchy of
39          spatial scales and can comprise the entire globe, biomes at the continental scale, or small,
40          well-circumscribed systems such as a  small pond.
41    Ecosystem Benefit: The value, expressed qualitatively, quantitatively,  and/or in economic terms,
42          where possible, associated with changes in ecosystem services that result either directly
43          or indirectly in improved human health and/or welfare. Examples of ecosystem benefits
44          that derive from improved air quality include improvements in habitats for sport fish
45          species, the quality of drinking water and recreational areas, and visibility.

      March 2010                                xii                 Draft - Do No Quote or Cite

-------
 1    Ecosystem Function: The processes and interactions that operate within an ecosystem.
 2    Ecosystem Services: The ecological processes or functions having monetary or non-monetary
 3          value to individuals or society at large. These are (1) supporting services, such as
 4          productivity or biodiversity maintenance; (2) provisioning services, such as food, fiber, or
 5          fish; (3) regulating  services, such as climate regulation or carbon sequestration; and (4)
 6          cultural services, such as tourism or spiritual and aesthetic appreciation.
 7    Eutrophication: The process by which nitrogen additions stimulate the growth of autotrophic
 8          biota, usually resulting in the depletion of dissolved oxygen.
 9    Nitrogen Enrichment: The process by which a terrestrial system becomes enhanced by nutrient
10          additions to a degree that stimulates the growth of plant or other terrestrial biota, usually
11          resulting in an increase in productivity.
12    Nitrogen Saturation: The point at which nitrogen inputs from atmospheric deposition and other
13          sources exceed the biological requirements of the ecosystem; a level beyond nitrogen
14          enrichment.
15    Occult Deposition: The removal of gases and particles from the atmosphere to surfaces by fog
16          or mi st.
17    Semi-arid Regions: Regions of moderately low rainfall, which are not highly productive and are
18          usually classified as rangelands. "Moderately low" is widely accepted as between  100-
19          and 250-mm precipitation per year.
20    Sensitivity: The degree to which a system is affected,  either adversely or beneficially, by NOX
21          and/or SOX pollution (e.g., acidification, nutrient enrichment). The effect may be direct
22          (e.g., a change in growth in response to a change in the mean,  range, or variability of
23          nitrogen deposition) or indirect (e.g., changes in growth due to the direct effect of
24          nitrogen consequently altering competitive dynamics between species and decreased
25          biodiversity).
26    Total Reactive Nitrogen:  This includes all biologically, chemically,  and radiatively active
27          nitrogen compounds in the atmosphere and biosphere, such as NFL?, NH4+, NO, NO2,
28          HNOs, N2O, NO3-, and organic compounds (e.g., urea, amines, nucleic acids).
29    Valuation: The economic or non-economic process of determining either the value of
30          maintaining a given ecosystem type, state, or condition, or the value of a change in an
31          ecosystem, its components, or the services it provides.
32    Variable Factors: Influences which by themselves or in combination with other factors may
33          alter the effects on public welfare of an air pollutant (section 108 (a)(2))
34          (a) Atmospheric Factors: Atmospheric conditions that may influence transformation,
35          conversion, transport, and deposition,  and thereby, the effects  of an air pollutant on
36          public welfare, such as precipitation, relative humidity,  oxidation state, and co-pollutants
37          present in the atmosphere.
38          (b) Ecological Factors: Ecological conditions that may influence the effects of an air
39          pollutant on public welfare once it is introduced into an ecosystem, such as soil base
40          saturation, soil thickness, runoff rate, land use conditions, bedrock geology, and
41          weathering rates.
42    Vulnerability: The degree to which a system is susceptible to,  and unable to cope with, the
43          adverse effects of NOX and/or SOX air  pollution.
44    Welfare Effects: The effects on soils, water,  crops, vegetation, man-made materials, animals,
45          wildlife, weather, visibility, and climate; as well as damage to and deterioration of
46          property, hazards to transportation, and the effects on economic values and on personal
      March 2010                                xiii                 Draft - Do No Quote or Cite

-------
1          comfort and well-being, whether caused by transformation, conversion, or combination
2          with other air pollutants (Clean Air Act Section 302[h]).
3   Wet Deposition: The removal of gases and particles from the atmosphere to surfaces by rain or
4          other precipitation.
5
6
7
    March 2010                               xiv                 Draft - Do No Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX


 1                                  1.     INTRODUCTION
 2          The U.S. Environmental Protection Agency (EPA) is presently conducting a review of
 3    the secondary National Ambient Air Quality Standards (NAAQS) for oxides of nitrogen (NOX)
 4    and oxides of sulfur (SOX). The EPA's overall plan and schedule for this review were presented
 5    in the Integrated Review Plan for the Secondary National Ambient Air Quality Standards  for
 6    Nitrogen Dioxide and Sulfur Dioxide (US EPA, 2007). The Integrated Review Plan (IRP)
 7    outlined the Clean Air Act (CAA or the Act) requirements related to the establishment and
 8    reviews of the NAAQS, the process and schedule for conducting the current review, and the key
 9    components in the NAAQS review process: an Integrated Science Assessment (ISA), Risk and
10    Exposure Assessment (REA), and policy assessment/rulemaking. It presented key policy -
11    relevant issues to be addressed in this review as a series of questions that frames our
12    consideration of whether the current secondary (welfare-based) NAAQS for NOX and SOX should
13    be retained or revised.
14          As part of this review, staff in the U.S. Environmental Protection Agency's (EPA) Office
15    of Air Quality Planning and Standards (OAQPS) prepared this first draft Policy Assessment.1
16    The objective of this assessment is to evaluate the policy implications of the key scientific
17    information contained in the document Integrated Science Assessment for Oxides of Nitrogen
18    and Sulfur-Ecological Criteria (USEPA, 2008; henceforth referred to as the ISA), prepared by
19    EPA's National Center for Environmental Assessment (NCEA) and the results from the analyses
20    contained in the Risk and Exposure Assessment for Review of the Secondary National Ambient
21    Air Quality Standards for  Oxides of Nitrogen and Oxides of Sulfur (U.S. EPA, 2009; henceforth
22    referred to as the REA). This first draft also presents preliminary  staff conclusions on a range of
23    policy options that we believe are appropriate for the Administrator to consider concerning
24    whether, and if so how, to revise the secondary (welfare-based) NOX and SOX NAAQS.
      1 Preparation of a PA by OAQPS staff reflects Administrator Jackson's decision to modify the NAAQS review
      process that was presented in the IRP. See http://www.epa.gov/ttn/naaqs/review.html for more information on the
      current NAAQS review process.

      February, 2010                              1                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1           This policy assessment is intended to help "bridge the gap" between the scientific
 2    assessment contained in the ISA and the judgments required of the EPA Administrator in
 3    determining whether it is appropriate to retain or revise the secondary NAAQS for NOX and SOX.
 4    This policy assessment considers the available scientific evidence and quantitative risk-based
 5    analyses, together with related limitations and uncertainties,  and focuses on the basic elements of
 6    air quality standards: indicators2, averaging times, forms3,  and levels. These elements, which
 7    serve to define each standard, must be considered collectively in evaluating the welfare
 8    protection afforded by the secondary NOX and SOX NAAQS. Our development of this policy
 9    assessment is based on the assessment and integrative synthesis of information presented in the
10    ISA and on staff analyses and evaluations presented in this document, and is further informed by
11    comments and advice received from an independent scientific review committee, the Clean Air
12    Scientific Advisory Committee (CASAC), in their review of the previous integrated science and
13    risk and exposure assessments. The Policy Assessment is further informed by  comments
14    submitted by the public4. To view related documents developed as part of the planning, science,
15    and risk assessment phases of this review see
16    http://www.epa.gov/ttn/naaqs/standards/no2so2sec/index.html.
17           This document is organized around a conceptual framework for a combined NOX and SOX
18    secondary NAAQS and is focused on answering key policy questions related to the
19    implementation of that conceptual framework. Chapter 2 provides a summary of ecological
20    effects from the deposition of ambient NOX and SOX to sensitive ecosystems, drawing from the
21    ISA and REA. Chapter 3 places those ecological effects within the context of "public welfare"
22    by linking effects to ecosystem services or other benchmarks of public welfare. Chapter 4
23    addresses the adequacy of the current NOX and SOX secondary NAAQS in addressing the impacts
24    on public welfare from ecological effects. Chapter 5 develops the conceptual design for
25    ecologically relevant multi-pollutant standards.  Chapter 6 presents options for developing critical
26    elements of a secondary NAAQS necessary  to implement the conceptual design. Chapter 7
27    describes how secondary NAAQS designed  to protect a specific ecological endpoint may also
28    provide protection for other ecological endpoints. Chapter 8  provides a consideration of issues
       The "indicator" of a standard defines the chemical species or mixture that is to be measured in determining
      whether an area attains the standard.
      3 The "form" of a standard defines the air quality statistic that is to be compared to the level of the standard in
      determining whether an area attains the standard.
      4 Summary information on public comments will be provided in a later draft of the policy assessment

      March 2010                                 2                 Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    regarding reduced and oxidized forms of nitrogen. Chapter 9 concludes with preliminary staff
 2    conclusions regarding ranges of options for pollutant indicators, averaging times, forms, and
 3    levels for the secondary NOX and SOX NAAQS, including a discussion of staff initial conclusions
 4    on what levels of the secondary NAAQS might be requisite to protect public welfare.
 5          In this document we consider how the available scientific evidence and quantitative risk-
 6    based analyses, together with related limitations and uncertainties, inform the review of each
 7    element of the NAAQS: indicator, averaging times, forms, and levels. These elements must be
 8    considered collectively in evaluating the welfare protection afforded by the secondary NAAQS
 9    standards. This draft document does not contain final staff conclusions as to all the necessary
10    components of an alternative secondary standard for NOX  and/or SOX but rather describes the
11    current state of thinking with regard to potential policy options and provides an appropriate
12    context of information for the Administrator to consider in making decisions regarding the
13    standards.
14          While this policy assessment should be of use to all parties interested in the secondary
15    NOX and SOX NAAQS review,  it is written with an expectation that the reader has some
16    familiarity with the technical discussions contained in the  ISA and REA.
17          EPA will be preparing a second draft Policy Assessment subsequent to receiving advice
18    from the CASAC. The second draft will incorporate responses to comments received from
19    CASAC, as well as comments submitted by the public. The second draft will also provide a more
20    complete development of the conceptual model, and will provide a more complete set of staff
21    conclusions on critical elements of the standards. EPA's final Policy Assessment will address
22    additional CASAC comments on the second draft, and will include sufficient information to
23    inform the Administrator on critical  elements  of the standards, and staff conclusions regarding
24    alternative levels  of the standards.

25    1.1    DEFINITIONS OF NOX AND  SOX FOR THIS ASSESSMENT
26          As discussed in detail in the REA (REA 1.3.1), in  the atmospheric science community
27    NOX is typically referred to as the sum of nitrogen dioxide (NC^), and nitric oxide (NO). From a
28    Clean Air Act perspective, the family of NOX  includes  any gaseous combination of nitrogen and
29    oxygen (e.g., NC>2, NO, nitrous oxide pSPzO], nitrogen trioxide [N^Os], nitrogen tetroxide PS^O^,
30    and dinitrogen pentoxide pS^Os]). The term used by the scientific community to represent the

      March  2010                                3                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    complete set of oxidized nitrogen compounds, including those listed in CAA Section 108(c), is
 2    total oxidized nitrogen (NOy). NOy includes all nitrogen oxides, including e.g. total reactive
 3    oxidized atmospheric nitrogen, defined as NOX (NO and NO2) and all oxidized NOX products:
 4    NOy = NO2 + NO + HNO3 + PAN +2N2O5 + HONO+ NO3 + organic nitrates + paniculate NO3
 5    (Finlayson-Pitts and Pitts, 2000). In this document, unless otherwise indicated, we use the term
 6    NOX interchangeably with NOy to refer to the complete set of oxidized nitrogen compounds.
 7          For this assessment, SOX is defined to include all oxides of sulfur, including multiple
 8    gaseous substances (e.g., SO2, sulfur monoxide [SO], sulfur trioxide [SO3], thiosulfate [S2O3],
 9    and heptoxide [S2O?], as well as particulate species, such as ammonium sulfate [(NFL^SOJ).
10    Throughout this text we refer to sulfate as SO4 and nitrate as NO3, recognizing that they have
11    charges of -2 for sulfate and -1 for nitrate.

12    1.2   POLICY OBJECTIVES
13          In conducting this periodic review of the NOX and  SOX secondary NAAQS, EPA has
14    decided to jointly assess the scientific information, associated risks, and standards relevant to
15    protecting the public welfare from adverse effects  associated with oxides of nitrogen and sulfur.
16    Although EPA has historically adopted separate secondary standards for oxides of nitrogen
17    (NOX) and oxides of sulfur (SOX), EPA is conducting a joint secondary review of these standards
18    because NOX, SOX, and their associated transformation products are linked from an atmospheric
19    chemistry perspective, as well as from an environmental effects perspective. The National
20    Research Council (NRC) has recommended that EPA consider multiple pollutants, as
21    appropriate, in forming the scientific basis for the NAAQS (NRC, 2004). There is a strong basis
22    for considering these pollutants together, building  upon EPA's and CAS AC's past recognition of
23    the interactions of these pollutants and on the growing body of scientific information that is now
24    available related to these interactions and associated ecological effects.
25          EPA sets secondary standards for two criteria pollutants related to NOX and SOX: ozone
26    and particulate matter (PM). NOX is a precursor to the formation of ozone in the atmosphere, and
27    under certain conditions, can combine with atmospheric ammonia to form ammonium nitrate, a
28    component of fine PM. SOX is a precursor to the formation of particulate sulfate, which is a
29    significant component of fine PM in many parts of the U.S. While there are a number of welfare
30    effects associated with ozone and fine PM, including ozone damage to vegetation, and visibility

      March 2010                                4                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    degradation related to PM, protection against those effects is provided by the ozone and fine PM
 2    standards. This review focuses on evaluation of the protection provided by NOX and SOX
 3    secondary standards for effects associated with direct atmospheric concentrations of NOX and
 4    SOX, and effects associated with deposition of NOX and SOX to ecosystems, including deposition
 5    in the form of particulate nitrate and sulfate in their component forms.
 6           The ISA highlights the ecological effects associated with deposition of ambient NOX and
 7    SOX to ecosystems other than commercially managed forests and agricultural lands. This
 8    assessment evaluates information on gas-phase effects of NOX and SOX via stomatal exposure on
 9    vegetation, but primarily focuses on the effects of gas-phase NOX and SOX exposure via
10    deposition on multiple ecological receptors. Highlighted effects include those associated with
11    acidification and nitrogen nutrient enrichment. Based on these highlighted effects, EPA's policy
12    objective is to develop a framework for NOX and SOX standards that incorporate factors that will
13    lead to standards that are ecologically relevant, and that recognizes the interactions between the
14    two pollutants as they deposit to sensitive ecosystems, with an ultimate goal of setting standards
15    that, based on the ecological criteria described in the ISA, and consistent with the requirements
16    of the Clean Air Act, "are requisite to protect the public welfare from any known or anticipated
17    adverse effects associated with the presence of such air pollutant in the ambient air."
18           In presenting policy options for the Administrator's consideration, we note that the final
19    decision on retaining or revising the current secondary standards for NOX and SOX is largely a
20    public welfare policy judgment based on the Administrator's informed assessment of what
21    constitutes requisite protection against adverse effects to public welfare. A final decision should
22    draw upon scientific information and analyses about welfare effects, exposure and risks, as well
23    as judgments about the appropriate response to the range of uncertainties that are inherent in the
24    scientific evidence and analyses. The ultimate  determination as to what level of damage to
25    ecosystems and the services provided by those ecosystems is adverse to public welfare is not
26    wholly a scientific question, although it is informed by  scientific studies linking ecosystem
27    damage to losses in ecosystem services, and economic information on the value of those losses in
28    ecosystem services. Our approach to informing these judgments, as discussed below, is
29    consistent with the requirements of the NAAQS provisions of the Clean Air Act and with how
30    EPA and the courts have historically interpreted the Act. These provisions require the
31    Administrator to establish secondary NAAQS that, in the Administrator's judgment, are requisite

      March 2010                                 5                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    to protect public welfare from any known or anticipated adverse effects associated with the
 2    presence of NOX and SOX in the ambient air. In so doing, the Administrator seeks to establish
 3    standards that are neither more nor less stringent than necessary for this purpose.
 4          For this first draft policy assessment, we have chosen to focus much of our discussion on
 5    the effects of ambient NOX and SOX on ecological impacts associated with acidifying deposition
 6    of nitrogen and sulfur, which is a transformation product of ambient NOX and SOX. We have the
 7    greatest confidence in the causal linkages between NOX and SOX and aquatic acidification effects,
 8    and we have the most complete information available with which to develop an ecologically
 9    meaningful structure for the standards. In future drafts, we expect to be able to explore whether
10    and how the standards can be expanded to directly address effects of acidification on terrestrial
11    ecosystems, and to address the effects of nutrient enrichment in terrestrial and aquatic
12    ecosystems.

13    1.3   CRITICAL POLICY ELEMENTS
14          Our policy objective is guided by the information in the ISA and REA, framed within the
15    legislative requirements of the CAA. This framing leads us to focus on critical policy elements
16    (CPE) consistent with elements of Clean  Air Act language.
17          Sections 108 and 109 of the CAA govern the establishment and periodic review of the
18    NAAQS and of the air quality criteria upon which the  standards are based. The NAAQS are
19    established for pollutants that are listed under section 108, based on three criteria, including
20    whether emissions of the air pollutant cause or contribute to air pollution which may reasonably
21    be anticipated to endanger public health or welfare and whose presence in the ambient air results
22    from numerous or diverse mobile or stationary sources. The NAAQS are based on air quality
23    criteria that reflect the latest scientific knowledge, useful in indicating the types and extent of
24    identifiable effects on public health or welfare that may be expected from the presence of the
25    pollutant in ambient air. The criteria refer to criteria issued pursuant to §108 of the Clean Air
26    Act, which include "(A) those variable factors (including atmospheric conditions) which of
27    themselves or in combination with other factors may alter the effects on public health or welfare
28    of such air pollutant; (B) the types of air pollutants which, when present in the atmosphere, may
29    interact with such pollutant to produce an adverse effect on public health of welfare; and (C) any
30    known or anticipated adverse effects on welfare."

      March 2010                                 6                  Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1           The following critical policy elements for the design of ecologically relevant secondary
 2    standards for NOX and SOX are identified:
 3           (CPE 1)   An evaluation of the effects of ambient NOX and SOX on ecosystems, and the
 4                     relationship between those effects and the measure of dose in the ecosystem,
 5                     indicated by the deposit!onal loadings of N and S.
 6           (CPE 1.1) Evaluation of the relationship between response of ecological receptors, e.g.
 7                     changes in diversity offish species, and the response related to public welfare,
 8                     e.g. loss in recreational fishing services.
 9           (CPE 1.2) Evaluation of the extent to which identified effects are occurring under recent
10                     conditions, and the extent to which meeting the current standards would
11                     provide protection against these effects.
12           (CPE 2)   An assessment of how best to characterize, in defining the standards, the
13                     variable ecosystem factors that affect the relationship between ecological
14                     effects and deposit! onal loadings of N and S.
15           (CPE 2.1) Specification  of potential indicators of ecological effects, e.g. acid
16                     neutralizing capacity (ANC) that incorporates variability in ecosystem factors.
17           (CPE 3)   Characterization of the complex atmospheric transformations between
18                     ambient concentrations of NOX and SOX and deposition of N and S in the
19                     specification of a standard.
20           (CPE 4)   Specification  of those factors, such as precipitation, which interact with
21                     ambient NOX  and SOX to produce adverse effects on welfare, by affecting
22                     deposition of N and S.
23           (CPE 5)   Specification  of the form for the standard(s), including ambient atmospheric
24                     indicators for NOX and SOX, with consideration of averaging times, and
25                     options for levels of the standard(s).
26           The development of the conceptual framework for the NOX and SOX standards described
27    in Section 1.4 will be motivated  by these critical policy elements. However, in order to provide a
28    historical context for this new framework, the next section provides a brief history of previous
29    reviews of the NOX and SOX secondary NAAQS, as well as other relevant historical reviews of
30    welfare effects associated with these pollutants.
      March 2010                                 7                 Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    1.4   HISTORICAL CONTEXT

 2          1.4.1   History of NOX and SOX NAAQS Review

 3          1.4.1.1  NOx NAAQS
 4          EPA began the most recent previous review of the NOX secondary standards in 1987 and
 5    in November 1991, EPA released an updated draft AQCD for CAS AC and public review and
 6    comment (56 FR 59285). This draft document provided a comprehensive assessment of the
 7    available scientific and technical information on health and welfare effects associated with NO2
 8    and other NOX. CAS AC reviewed the draft document at a meeting held on July 1, 1993, and
 9    concluded in a closure letter to the Administrator that the document "provides a scientifically
10    balanced and defensible summary of current knowledge of the effects of this pollutant and
11    provides an adequate basis for EPA to make a decision as to the appropriate NAAQS for NO2"
12    (Wolff, 1993). The AQCD Air Quality Criteria for Oxides of Nitrogen was then finalized (U.S.
13    EPA, 1993). EPA also prepared a Staff Paper that summarized and integrated the key  studies and
14    scientific evidence contained in the revised NOX AQCD and identified the critical elements to be
15    considered in the review of the NO2 NAAQS. CASAC reviewed two drafts of the Staff Paper and
16    concluded in a closure letter to the Administrator that the document provided a "scientifically
17    adequate basis for regulatory decisions on nitrogen dioxide" (Wolff, 1995). In October 1995, the
18    Administrator announced her proposed decision not to revise either the primary or secondary
19    NAAQS for NO2 (60 FR 52874; October 11, 1995). A year later, the Administrator made a final
20    determination not to revise the NAAQS for NO2 after careful evaluation of the comments
21    received on the proposal (61 FR 52852; October 8, 1996). The level for both the existing primary
22    and secondary NAAQS for NO2 is 0.053 ppm (100 micrograms per cubic meter [jjg/ms] of air),
23    annual arithmetic average, calculated as the arithmetic mean of the 1-hour NO2 concentrations.

24          1.4.1.2  SOX NAAQS
25          Based on the 1970 SOX criteria document (DHEW, 1970), EPA promulgated primary and
26    secondary NAAQS for SO2 on April 30, 1971 (36 FR 8186). The secondary standards included a
27    standard at 0.02 ppm in an annual arithmetic mean and a 3-hour average of 0.5 ppm, not to be
28    exceeded more than once per year. These secondary standards were established solely on the
29    basis of evidence of adverse effects on vegetation. In  1973, revisions made to Chapter 5

      March 2010                              8                 Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    ("Effects of Sulfur Oxide in the Atmosphere on Vegetation") of Air Quality Criteria for Sulfur
 2    Oxides (U.S. EPA, 1973) indicated that it could not properly be concluded that the vegetation
 3    injury reported resulted from the average SO2 exposure over the growing season, rather than
 4    from short-term peak concentrations. Therefore, EPA proposed (38 FR 11355) and then finalized
 5    (38 FR 25678) a revocation of the annual mean secondary standard. At that time, EPA was aware
 6    that SOX have other public welfare effects, including effects on materials, visibility, soils, and
 7    water. However, the available data were considered insufficient to establish a quantitative
 8    relationship between specific ambient SOX concentrations and effects (38 FR 25679).
 9          In 1979, EPA announced that it was revising the Air Quality Criteria Document (AQCD)
10    for sulfur oxides concurrently with that  for particulate matter and would produce a combined
11    particulate matter and sulfur oxides criteria document. Following its review of a draft revised
12    criteria document in August 1980, CAS AC concluded that acid deposition was a topic of
13    extreme  scientific complexity because of the difficulty in establishing firm quantitative
14    relationships among (1) emissions of relevant pollutants (e.g., SC>2 and  oxides of nitrogen), (2)
15    formation of acidic wet and dry deposition products, and (3) effects on  terrestrial and  aquatic
16    ecosystems. CAS AC also noted that acid deposition involves, at a minimum, several different
17    criteria pollutants: oxides of sulfur, oxides of nitrogen, and the fine particulate fraction of
18    suspended particles. CAS AC felt that any document on this subject should address both wet  and
19    dry deposition, since dry deposition was believed to account for at least one half of the total acid
20    deposition problem.
21          For these reasons, CAS AC recommended that a separate, comprehensive document on
22    acid deposition be prepared prior to any consideration of using the NAAQS as a regulatory
23    mechanism for the control of acid deposition. CASAC also suggested that a discussion of acid
24    deposition be included in the AQCDs for nitrogen oxides and PM and SOX. Following CASAC
25    closure on the AQCD for SC>2 in December 1981, EPA's Office of Air  Quality Planning and
26    Standards published a Staff Paper in November 1982, but the paper did not directly assess the
27    issue of acid deposition. Instead, EPA subsequently prepared the following documents: The
28    Acidic Deposition Phenomenon and Its Effects: Critical Assessment Review Papers, Volumes I
29    and II (U.S. EPA,  1984a, b), and The Acidic Deposition Phenomenon and Its Effects:  Critical
30    Assessment Document (U.S. EPA, 1985) (53 FR 14935  -14936). These  documents, though they
      March  2010                                9                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    were not considered criteria documents and did not undergo CASAC review, represented the
 2    most comprehensive summary of relevant scientific information completed by EPA at that point.
 3          On April 26, 1988 (53 FR 14926), EPA proposed not to revise the existing primary and
 4    secondary standards for SC>2. This proposal regarding the secondary SC>2 NAAQS was due to the
 5    Administrator's conclusions that (1) based upon the then-current scientific understanding of the
 6    acid deposition problem, it would be premature and unwise to prescribe any regulatory control
 7    program at that time, and (2) when the fundamental scientific uncertainties had been reduced
 8    through ongoing research efforts, EPA would draft and support an appropriate set of control
 9    measures.

10          1.4.2   History of Related Assessments and Agency Actions
11          In 1980, the Congress created the National Acid Precipitation Assessment Program
12    (NAPAP) in response to growing concern about acidic deposition. The NAPAP was given a
13    broad 10-year mandate to examine the causes and effects of acidic deposition and to explore
14    alternative control options to alleviate acidic deposition and its effects. During the course of the
15    program, the NAPAP issued a series of publicly available interim reports prior to the completion
16    of a final report in 1990 (NAPAP, 1990).
17          In spite of the complexities and significant remaining uncertainties associated with the
18    acid deposition problem, it soon became clear that a program to address acid deposition was
19    needed. The Clean Air Act Amendments of 1990 included numerous separate provisions related
20    to the acid deposition problem. The primary and most important of the provisions, the
21    amendments to Title IV of the Act, established the Acid Rain Program to reduce emissions of
22    SC>2 by 10 million tons and NOX emissions by 2 million tons from  1980 emission levels in order
23    to achieve reductions over broad geographic regions. In this provision, Congress included a
24    statement of findings that led them to take action, concluding that (1) the presence of acid
25    compounds and their precursors in the atmosphere and in deposition from the atmosphere
26    represents a threat to natural resources, ecosystems, materials, visibility, and public health; (2)
27    the problem of acid deposition is of national and international significance; and (3) current and
28    future generations of Americans will be adversely affected by delaying measures to remedy the
29    problem.
      March  2010                               10                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1           Second, Congress authorized the continuation of the NAPAP in order to assure that the
 2    research and monitoring efforts already undertaken would continue to be coordinated and would
 3    provide the basis for an impartial assessment of the effectiveness of the Title IV program.
 4           Third, Congress considered that further action might be necessary in the long term to
 5    address any problems remaining after implementation of the Title IV program and, reserving
 6    judgment on the form that action could take, included Section 404 of the 1990 Amendments
 7    (Clean Air Act Amendments of 1990, Pub. L. 101-549,  § 404) requiring EPA to conduct a study
 8    on the feasibility and effectiveness of an acid deposition standard or standards to protect
 9    "sensitive and critically sensitive aquatic and terrestrial  resources." At the conclusion of the
10    study, EPA was to submit a report to Congress. Five years later, EPA submitted its report,
11    entitled Acid Deposition Standard Feasibility Study: Report to Congress (U.S. EPA, 1995) in
12    fulfillment of this requirement. The Report concluded that establishing acid deposition standards
13    for sulfur and nitrogen deposition may at some point in  the future be technically feasible,
14    although appropriate deposition loads for these acidifying chemicals could not be defined with
15    reasonable certainty at that time.
16           Fourth, the 1990 Amendments also added new language to sections of the CAA
17    pertaining to the scope and application of the secondary NAAQS designed to protect the public
18    welfare. Specifically, the definition of "effects on welfare"  in Section 302(h) was expanded to
19    state that the welfare effects include effects ".. .whether caused by transformation, conversion, or
20    combination with other air pollutants."
21           In 1999,  seven Northeastern states cited this amended language in Section 302(h) in a
22    petition asking EPA to use its authority under the NAAQS program to promulgate secondary
23    NAAQS for  the criteria pollutants associated with the formation of acid  rain. The petition stated
24    that this language "clearly references the transformation of pollutants resulting in the inevitable
25    formation of sulfate and nitrate aerosols and/or their ultimate environmental impacts as wet and
26    dry deposition, clearly signaling Congressional intent that the welfare damage occasioned by
27    sulfur and nitrogen oxides be addressed through the secondary standard provisions of Section
28    109 of the Act." The petition further stated that "recent federal studies, including the NAPAP
29    Biennial Report to Congress: An Integrated Assessment, document the continued-and increasing-
30    damage being inflicted by acid deposition to the lakes and forests of New York, New England
31    and other parts of our nation, demonstrating that the Title IV program had proven insufficient."

      March 2010                                11                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    The petition also listed other adverse welfare effects associated with the transformation of these
 2    criteria pollutants, including impaired visibility, eutrophication of coastal estuaries, global
 3    warming, and tropospheric ozone and stratospheric ozone depletion.
 4           In a related matter, the Office of the Secretary of the U.S. Department of Interior
 5    requested in 2000 that EPA initiate a rulemaking proceeding to enhance the air quality in
 6    national parks and wilderness areas in order to protect resources and values that are being
 7    adversely affected by air pollution. Included among the effects of concern identified in the
 8    request were the acidification of streams, surface waters, and/or soils; eutrophication of coastal
 9    waters; visibility impairment; and foliar injury from ozone.
10           In a Federal Register notice in 2001, EPA announced receipt of these requests and asked
11    for comment on the issues raised in them. EPA stated that it would consider any relevant
12    comments and information submitted, along with the information provided by the petitioners and
13    DOI, before making any decision concerning a response to these requests for rulemaking (65 FR
14    48699).
15           The most recent 2005 NAPAP report states that"...  scientific studies indicate that the
16    emission reductions achieved by Title IV are not sufficient to allow recovery of acid-sensitive
17    ecosystems. Estimates from the literature of the scope of additional emission reductions that are
18    necessary in order to protect acid-sensitive ecosystems range from approximately 40-80%
19    beyond full implementation of Title IV.... The results of the modeling presented in this Report to
20    Congress indicate that broader recovery is not predicted without additional emission reductions"
21    (NAPAP, 2005).5
22           Given the state of the science as described in the ISA and in other recent reports, such as
23    the NAPAP's above, EPA believes it is appropriate, in the context of evaluating the adequacy of
24    the current NC>2 and SC>2 secondary standards in this review, to revisit the question of the
25    appropriateness and the feasibility of setting a secondary NAAQS to address remaining known
26    or anticipated adverse public welfare effects resulting from the acidic and nutrient deposition of
27    these criteria pollutants
      5 Note that a new NAPAP report is expected to be released later in 2010.  The findings of that report will be
      considered in the final policy assessment.

      March 2010                                 12                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    1.5   PROPOSED CONCEPTUAL FRAMEWORK FOR COMBINED NOX
 2          SOX STANDARDS
 3          There is a strong basis for considering NOX and SOX together at this time, building upon
 4    EPA's and CASAC's past recognition of the interactions of these pollutants and on the growing
 5    body of scientific information that is now available related to these interactions and associated
 6    ecological effects. The REA introduced a conceptual framework for ecologically meaningful
 7    secondary standards that recognized the complex processes by which ecosystems are exposed to
 8    ambient NOX and SOX. That framework provided a flow from ambient concentrations exposures
 9    via deposition to ecological indicators and effects (see Figure ES-2 in the REA Executive
10    Summary). This sequence represents the process by which we can determine the risks associated
11    with ambient concentrations of NOX and SOX. However, for the purposes of discussing a
12    conceptual framework for design of standards to protect against those risks, a modified version
13    of the risk frame work is needed.
14          Figure 1-1 depicts the framework by which we are considering the structure  of an
15    ecologically meaningful secondary standard. It is a conceptual diagram that illustrates how a
16    level of protection related to an indicator of ecological effect(s) equates to atmospheric
17    concentrations of NOX and SOX indicators. This conceptual diagram illustrates the linkages
18    between ambient air concentrations and resulting deposition metrics, and between the deposition
19    metric and the ecological indicator of concern. The Atmospheric Deposition Transformation
20    Function translates ambient atmospheric concentrations of NOX and SOX to nitrogen  and sulfur
21    deposition metrics, while the Ecological Effect Function transforms the deposition metric into
22    the ecological indicator.
23          Development of a form for the standard that reflects this structure is a critical step in the
24    overall standard setting process. The atmospheric levels of NOX and SOX that satisfy  a particular
25    level of ecosystem protection are those levels that result in an amount of deposition that is less
26    than the amount of deposition that a given ecosystem can accept without excessive degradation
27    of the ecological indicator for a targeted effect.
28          The details of this conceptual framework are discussed in Chapter 5, including
29    discussions of modifying factors that alter the relationship between ambient atmospheric
30    concentrations of NOX and SOX and depositional loads of nitrogen and sulfur,  and those that
31    modify the relationship between deposition loads and the ecological indicator.

      March 2010                               13                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1           In setting NAAQS to protect public health and welfare, EPA has historically established
 2    standards which require the comparison of monitored concentrations of an air pollutant against a
 3    numerical metric of atmospheric concentration that does not vary geographically. This approach
 4    has appropriately protected public health as at-risk populations are widely distributed throughout
 5    the nation. As more is learned about the effects of pollutants such as NOX and SOX and the
 6    environment, however, such an approach may not be appropriate to provide the requisite level of
 7    protection to public welfare from effects on sensitive ecosystems. EPA is considering in this
 8    review of the secondary standard for NOX and SOX whether a standard that takes into account
 9    variable factors, such as atmospheric variables and ecosystem sensitivities, is the appropriate
10    approach to protect the public welfare from the effects associated with the presence of these
11    pollutants in the ambient air.
12           EPA must undertake  a thorough review of the air quality criteria for the pollutant at issue
13    in reviewing a secondary NAAQS, and determine whether a current standard is requisite to
14    protect the public welfare. Under section 108 of the CAA, air quality criteria are to "reflect the
15    latest scientific knowledge useful in indicating the kind and extent of all identifiable effects"
16    associated with the presence of the pollutant in the ambient air. It is clear from the language of
17    the CAA that where the state of the science provides a basis for considering  such effects, the
18    review of the air quality criteria should encompass a broad analysis of "any" known or
19    anticipated  adverse effects, as well as the ways in which variable conditions  such as atmospheric
20    conditions may impact the effect of a pollutant and the ways in which other air pollutants may
21    interact with the criteria pollutant to produce adverse effects. Specifically, section  108(a)(2) of
22    the CAA provides that:
23           Air  quality criteria for an air pollutant shall accurately reflect the latest scientific
24    knowledge  useful in indicating the kind and extent of all identifiable effects  on public health or
25    welfare which may be expected from the presence of such pollutant in the ambient air, in varying
26    quantities. The criteria for an air pollutant to the extent practicable, shall include information on:
27      •   (A)     those variable factors (including atmospheric conditions) which of themselves or
28          in combination with other factors may alter the effects on public health or welfare of such
29          air pollutants;
30      •   (B)     the types of air pollutants which, when present in the atmosphere, may interact
31          with  such pollutants to produce an adverse effect on public health or welfare; and

      March 2010                                 14                  Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1      •   (C)     any known or anticipated adverse effects on welfare.
 2          Based on this extensive review of the air quality criteria for an air pollutant, the
 3    Administrator is required to review and to revise, as appropriate, the secondary standard to
 4    ensure that the standard "is requisite to protect public welfare from any known or anticipated
 5    adverse effects associated with the presence of such air pollutant in the ambient air." CAA §
 6    109(b) & (d). "Effects on welfare," in turn, is defined to include a broad array of effects,
 7    including effects on soil, water, crops, vegetation, and manmade materials, "whether caused by
 8    transformation,  conversion, or combination with other air pollutants." CAA § 302(h). Thus, as
 9    with the sections of the CAA describing the issuance of air quality criteria, the CAA uses
10    expansive language in describing the scope of EPA's responsibility and the range of effects that
11    EPA should take into account in setting a standard that is requisite to protect public welfare. The
12    term "requisite," however, indicates that section 109 is not open-ended. In considering the
13    meaning of the term "requisite" in the context of the primary standards, the Supreme Court has
14    agreed with EPA that such a standard is one that is "sufficient, but not more than necessary" to
15    protect public health. Whitman v. American Trucking, 531 U.S. 457, 473 (2001).
16          While EPA has most often considered the results of direct exposure to an air pollutant in
17    the ambient air in  assessing effects on public health  and welfare, such as the health effects on
18    humans when breathing in an air pollutant or the effects on vegetation through the uptake of air
19    pollutants from  the ambient air through leaves, EPA has also considered, where appropriate, the
20    effects of exposure to air pollutants through more indirect mechanisms. For example, both in
21    1978 and in 2008, EPA established a NAAQS for lead that addressed the health effects of
22    ambient lead whether the lead particles were inhaled or were ingested after deposition on the
23    ground or other surfaces. 73 FR 66964 (November 12, 2008), Lead Industries v. EPA, 647 F.2d
24    1130 (DC Cir. 1980) (1978 NAAQS). The deposition of ambient NOX and SOX to terrestrial and
25    aquatic environments can impact ecosystems through both direct and indirect mechanisms, as
26    discussed in the REA and this document. Given Congress' instruction to set a standard that "is
27    requisite to protect the public welfare from "any known or anticipated adverse effects associated
28    with the presence  of such air pollutant in the ambient air," 42 U.S.C. § 109 (b)(2) (emphasis
29    added), this review appropriately attempts to take into consideration widely acknowledged
30    effects, such as acidification and nutrient enrichment, which are associated with the presence of
31    ambient SOX and NOX.

      March 2010                                15                 Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
       In this review, EPA is also attempting to develop a standard that takes into account the
variability in effects from ambient levels of SOX and NOX. The CAA requires EPA to establish
"national" standards, based on the air quality criteria, that provide the requisite degree of
protection, but does not clearly address how to do so under the circumstances present here. One
approach is to develop a secondary standard such as the one discussed in this Policy Assessment
Document. Such a standard is designed to provide a generally uniform degree of protection
throughout the country by allowing for varying concentrations of allowable ambient NOX and
SOX, depending on  atmospheric conditions and other variabilities, to achieve that degree of
protection. Such a standard protects sensitive ecosystems wherever such ecosystems are found.
This approach recognizes that setting a standard that is sufficient to protect the public welfare but
not more than is necessary calls for consideration of a standard such as the one discussed in this
document.
              Structure of an Ecologically-based  Standard
                          Variable/Fixed
                            Factors:
                            Atmospheric
                            Landscape
                           Atmospheric
                            Deposition
                           Transformation
                            Function
                                       Form of the Standard
                                       Level of the Standard
       Figure 1-1. Framework of an alternative secondary standard.
1.6    POLICY RELEVANT QUESTIONS
       In this policy assessment, a series of general questions frames our approach to identifying
a range of policy options for consideration by the Administrator regarding secondary NAAQS
for NOX and SOX. These questions are drawn from our Integrated Review Plan with
     March 2010
                                          16
Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    modifications based on further consideration by staff and comments from CAS AC and the
 2    public. Our policy assessment begins by characterizing "known or anticipated adverse effects"
 3    on public welfare within our conceptual model (CPE 1). As noted earlier, this review is focusing
 4    on effects in unmanaged ecosystems (not commercial forests or agricultural lands6) resulting
 5    from ambient concentrations of NOX and SOX through deposition of N and S. In Chapter 2, we
 6    draw from the information and conclusions presented in the ISA and REA to address the
 7    following questions:
 8           1.  What are the nature and magnitude of ecosystem responses to reactive nitrogen and
 9              sulfur deposition?
10              a. How are these responses affected by landscape factors?
11              b. What types of ecosystems are sensitive to such responses?
12           2.  To what extent can ecosystem responses to nitrogen deposition be separated into
13              responses related to oxidized and reduced forms of reactive nitrogen compounds?
14           In Chapter 3, we address the following questions related to linking effects to measures of
15    adversity (CPE 1.1):
16           1.  How do we characterize  adversity to public welfare? What are the sources of
17              potentially relevant characterization for this policy assessment?
18           2.  What is the evidence of effects on ecosystem services, and how can those ecosystem
19              services be linked to ecological indicators?
20           3.  To what extent are identified ecosystem effects important from a public welfare
21              perspective, and what are the important uncertainties associated with estimating such
22              effects?
23           Once we have described ecological effects, we then provide an assessment of the
24    adequacy of the existing NOX and SOX standards (CPE 1.2). We begin this assessment by
25    drawing from the information and conclusions presented in the ISA and REA to address in
26    Chapter 4 the following questions, which allow us to identify whether the structure of the current
27    standards is appropriate relative to the key ecological effects assessed in the ISA and REA,
      6 The decision to focus on unmanaged ecosystems is based on the weight of evidence of effects in those ecosystems.
      The majority of the scientific evidence regarding acidification and nutrient enrichment is based on studies in
      unmanaged ecosystems. Non-managed terrestrial ecosystems tend to have a higher fraction of N deposition
      resulting from atmospheric N (ISA 3.3.2.5). In addition, the ISA notes that agricultural and commercial forest lands
      are routinely fertilized with amounts of N (100 to 300 kg N/ha) that exceed air pollutant inputs even in the most
      polluted areas (ISA 3.3.9)

      March 2010                                 17                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    including acidification and excess nutrient enrichment and whether there is adequate information
 2    and analyses available at this time to assess the extent to which potentially adverse effects on
 3    aquatic and terrestrial ecosystems can be associated with current levels of atmospheric reactive
 4    nitrogen, accounting for the contributions of oxidized and reduced forms, and SOX and with
 5    levels that are at or below the current secondary standards:
 6           1.  To what extent are effects that could reasonably be judged to be adverse to public
 7              welfare occurring under current conditions and would such effects occur if the nation
 8              met the current standards? To what extent do the current NOX and SOX secondary
 9              standards provide protection from effects  associated with deposition of:
10              a.  Sulfur and oxidized nitrogen from atmospheric NOX, and SOX which results in
11                 acidification in sensitive aquatic and terrestrial ecosystems?
12              b.  Oxidized nitrogen from atmospheric NOX, which results in nutrient enrichment
13                 effects in sensitive aquatic and terrestrial ecosystems?
14              c.  Sulfur and oxidized nitrogen from atmospheric NOX and  SOX which results in
15                 other ecological effects (e.g. mercury methylation)?
16           2.  In what way are the structures of the current NOX and SOX secondary standards
17              inadequate to protect against public welfare effects?
18           In Chapter 5, we follow our adequacy assessment by developing in greater detail the
19    conceptual framework for the design of ecologically relevant multi-pollutant standards
20    introduced in Section 1.4  above. To the extent that the available information calls into question
21    the adequacy of protection afforded by the current  standards and/or the appropriateness of the
22    structure of the  standards, we explore the extent to which available information supports
23    consideration of alternative standards, in terms of atmospheric and ecological indicators and
24    related averaging times, forms, and levels. This conceptual framework is designed to focus on
25    resolving the following questions:
26           1.  (CPE 2.1) Does the available information provide support for the use of ecological
27              indicators to characterize the responses  of aquatic and terrestrial ecosystems to
28              oxidized nitrogen and sulfur deposition?
29           2.  (CPE 1) Does the available information provide support for the development of
30              appropriate ecological response to deposition relationship(s) that meaningfully relates
31              oxidized nitrogen and sulfur deposition to relevant ecological indicators? Does a

      March 2010                                18                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1              quantified relationship exist between the level of a relevant ecological indicator and
 2              an amount of nitrogen and sulfur deposition?
 3           3.  (CPE 2) What are the important variables in the ecological response to deposition
 4              relationship(s)? Are these relationships applicable nationally? What are the
 5              appropriate temporal scales for these relationships?
 6              a.  How does ecological  response to deposition relationship(s) depend upon spatially
 7                 heterogeneous geologic factors (e.g. bedrock type, weathering rates) that govern
 8                 sensitivity?
 9              b.  How do we consider areas with high natural background acidification or nutrient
10                 loadings?
11           4.  (CPE 3) Does the available information provide support for the development of
12              appropriate functions that characterize the relationships between atmospheric NOX
13              and SOX and the wet and  dry deposition of total reactive nitrogen and sulfur? (CPE 4)
14              How do these relationships depend upon relevant atmospheric factors  (e.g., reduced
15              forms of nitrogen, meteorological factors) and landscape factors?
16              a.  What deposition function is appropriate to use for the purpose of relating an
17                 amount of nitrogen and/or sulfur deposition in sensitive ecosystems to ambient
18                 concentrations of atmospheric reactive nitrogen, including oxides and reduced
19                 forms, and/or sulfur?  What are the important variables in such a function? What
20                 are appropriate spatial and temporal scales to use in specifying such variables?
21           Based on the conceptual framework for the structure of the ecologically relevant multi-
22    pollutant standards, we then address  in Chapter 6 the elements of the standard needed to develop
23    options for consideration by the Administrator. Development of these options will focus on
24    addressing the following questions:
25           1.  (CPE 2.1) What ecological indicators are appropriate to use for the purpose of
26              developing an alternative standard for the various ecological effects assessed in this
27              review?
28           2.  (CPE 5) What indicators  of oxides of nitrogen and sulfur are appropriate to use for
29              the purpose of determining whether the resultant deposition is within the target values
30              needed to achieve the desired degree of protection? What averaging times and forms
31              are appropriate to consider?

      March 2010                                 19                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1           3.  (CPE 4) What approaches are available to specify non-atmospheric elements of the
 2              standard, e.g. weathering rates? Are there approaches that can simplify the structure
 3              of the standard by using discrete representations (bins) of continuous variables?
 4           4.  What are the available approaches for accounting for reduced N in the structure of the
 5              standard?
 6           5.  What is the most appropriate form for the standards to reflect the relationships
 7              between ambient NOX and SOX, acidifying deposition, and the ecological indicator for
 8              acidification?
 9           Several follow-up questions derive from our assessment of options for specifying the
10    elements of a multipollutant standard. In Chapter 7, we address the questions:
11           1.  To what extent would a standard specifically defined to protect against one ecological
12              effect (i.e., aquatic acidification) likely provide protection from other relevant
13              ecological effects?
14           2.  What are the available approaches for combining multiple indicators into a single
15              standard, e.g. using nitrogen effects to bound the tradeoff curve for NOX/SOX for
16              aquatic acidification effects
17           3.  What are the available approaches to integrate potential standards for aquatic and
18              terrestrial acidification and/or aquatic and terrestrial N enrichment?
19           In Chapter 8, we plan to address in the second draft policy assessment issues regarding
20    the adequacy of the current definitions of oxides of nitrogen and sulfur in specifying standards
21    for protection against effects associated with deposition of nitrogen and sulfur. This discussion
22    will be focused on the following questions:
23           1.  To what extent are effects associated with atmospheric nitrogen deposition reduced
24              when NOX related deposition is reduced?
25           2.  To what extent can appropriate protection from relevant ecological effects be
26              achieved by specifying indicators of atmospheric reactive nitrogen and sulfur
27              compounds in terms of gas- and particle-phase nitrogen oxides and/or sulfur oxides?
28           3.  To what extent does the available information on welfare effects provide a basis for
29              considering expanding the list of criteria pollutants to include all reactive nitrogen or
30              gas-phase ammonia? What are the relative merits of listing total reactive nitrogen
31              versus gas phase ammonia for protection of public welfare effects?

      March 2010                                20                 Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1          We conclude with a discussion of a range of options to consider in selecting pollutant
 2    indicators, averaging times, forms, and levels for the secondary NOX and SOX standards,
 3    including a discussion of staff initial conclusions on what levels of the standard for NOX and SOX
 4    would be requisite to protect public welfare against adverse ecological effects. This discussion is
 5    informed by a consideration of the role of ecosystem services in helping to characterize what
 6    adversity to public welfare, focused on the following questions:
 7          1.  (CPE 5) What are the risks of ecosystem service impairment under alternative levels
 8              of potential standards for NOX and SOX?
 9          2.  (CPE 5) To what extent can information about ecosystem services be used to help
10              characterize the extent to which differing levels of relevant ecological indicators
11              reflect impacts that can reasonably be judged to be adverse from a public welfare
12              perspective?
13          3.  (CPE 5) Are there relevant benchmarks for adversity to public welfare that can be
14              derived from other sources?
15          4.  (CPE 5) Taking into consideration information about ecosystem services and other
16              factors related to characterizing adversity to public welfare for the ecological effects
17              being assessed in this review, what is an appropriate  range of levels of protection to
18              be achieved by alternative standards for the Agency to consider?
19
     March  2010                                21                 Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX


 1           2.     KNOWN OR ANTICIPATED ECOLOGICAL EFFECTS
 2          This chapter addresses Critical Policy Element 1, evaluation of the effects of ambient
 3    NOX and SOX on ecosystems,  and the relationship between those effects and the measure of dose
 4    in the ecosystem, indicated by the deposit!onal loadings of N and S. In section 302(h) of the
 5    Clean Air Act, welfare effects addressed by a secondary NAAQS include, but are not limited to,
 6    "effects on soils, water, crops, vegetation, man-made materials, animals, wildlife, weather,
 7    visibility and climate, damage to and deterioration of property, and hazards to transportation, as
 8    well as effects on economic values and on personal comfort and well-being". Of these welfare
 9    effects categories, the effects  of NOX and SOX on aquatic and terrestrial ecosystems, which
10    encompass soils, water, vegetation, wildlife, and contribute to economic value and well-being,
11    are of most concern at concentrations typically occurring in the U.S. Direct effects of NOX and
12    SOX on vegetation are also discussed in this chapter, and have been the focus of previous
13    reviews. However, for this review, the focus of this chapter is on the known and anticipated
14    effects to ecosystems caused by exposure to NOX and SOX through deposition.
15          The information presented here is a concise summary of conclusions from the ISA and
16    the REA. This chapter focuses on  effects on specific ecosystems with a brief discussion on
17    critical uncertainties associated with acidification and nutrient enrichment; Chapter 3 evaluates
18    those effects within the context of alternative definitions of, including assessments of potential
19    impacts on ecosystem services. Effects are broadly categorized into acidification and nutrient-
20    enrichment in the proceeding sections. This is background information intended to support new
21    approaches for the design of ecologically relevant secondary NOX and SOX standards which are
22    protective of U.S. ecosystems. More detailed information on the conceptual  design and specific
23    options for the proposed standards are presented in Chapters 5 and 6 of this policy assessment
24    document. While we provide  a summary of effects for all four of the primary effects categories,
25    we reiterate that the focus of this first draft policy assessment is on effects related to aquatic
26    acidification, without downplaying the potential significance of effects in other categories.
      March  2010                               22                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1   2.1    ACIDIFICATION: EVIDENCE OF EFFECTS ON STRUCTURE AND
 2          FUNCTION OF TERRESTRIAL AND FRESHWATER
 3          ECOSYSTEMS
 4          Sulfur oxides (SOX) and nitrogen oxides (NOX) compounds in the atmosphere undergo a
 5   complex mix of reactions and thermodynamic processes in gaseous, liquid, and solid phases to
 6   form various acidic compounds. These acidic compounds are removed from the atmosphere
 7   through deposition: either wet (e.g., rain, snow), fog or cloud, or dry (e.g., gases, particles).
 8   Deposition of these acidic compounds leads to ecosystem exposure and effects on ecosystem
 9   structure and function. Following deposition, these compounds can, in some instances, leach out
10   of the soils in the form of sulfate (SC>42") and nitrate (N(V), leading to the acidification of surface
11   waters. The effects on ecosystems depend on the magnitude of deposition, as well as a host of
12   biogeochemical processes occurring in the soils and waterbodies (REA 2.1).  The chemical forms
13   of nitrogen that may contribute to acidifying deposition include both oxidized and reduced
14   species.
15          When sulfur or nitrogen leaches from soils to surface waters in the form of SC>42" or N(V,
16   an equivalent amount of positive cations, or countercharge, is also transported. This maintains
17   electroneutrality. If the countercharge is provided by base cations, such as calcium (Ca2+),
18   magnesium (Mg2+), sodium (Na+), or potassium (K+), rather than  hydrogen (H+) and dissolved
19   inorganic aluminum, the acidity of the soil water is neutralized, but the base saturation of the soil
20   is reduced. Continued SC>42 or N(V leaching can deplete the base cation supply of the soil. As
21   the base cations are removed, continued deposition and leaching of SO42"  and/or NO3" (with
22   H+and A13+) leads to acidification of soil water, and by connection, surface water. A watershed's
23   ability to neutralize acidic deposition is determined by a host of biogeophysical factors, including
24   base cation concentrations, weathering rates, uptake by vegetation, rate of surface water flow,
25   soil depth, and bedrock. (REA 2.1)  Some of these factors such as vegetation  and soil depth are
26   highly variable over small spatial scales, but others vary over larger spatial scales like geology.
27   For the purpose of a national  secondary standard, the most relevant characteristics are those that
28   are less variable over small scales.
29          Acidifying deposition of NOX and SOX and the chemical and biological responses
30   associated with these inputs vary temporally. Chronic or long-term deposition processes result in
     March 2010                              23                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    increases of N and S and the associated effects of acidifying deposition in the time scale of years
 2    to decades. Episodic or short term (i.e., hours or days) deposition refers to events in which the
 3    level of the acid neutralizing capacity (ANC) of a lake or stream is temporarily lowered. In
 4    aquatic ecosystems, short-term (i.e., hours or days) episodic changes in water chemistry can have
 5    significant biological effects. Episodic chemistry refers to conditions during precipitation or
 6    snowmelt events when proportionately more drainage water is routed through upper soil horizons
 7    that tend to provide less acid neutralizing than was passing through deeper soil horizons (REA
 8    4.2). Some streams and lakes may have chronic or base flow chemistry that is suitable for aquatic
 9    biota, but may be subject to occasional acidic episodes with lethal consequences.
10           The following summary is a concise overview of the known or anticipated effects caused
11    by acidification to ecosystems within the United States. Acidification affects both terrestrial and
12    freshwater aquatic ecosystems. Terrestrial and aquatic processes are often linked; therefore
13    responses to the following questions address both types of ecosystems  unless otherwise noted.

14           2.1.1  What is the nature of acidification related ecosystem responses to reactive
15                 nitrogen and/ sulfur deposition?
16           The ISA concluded that deposition of SOX, NOX, and NHX leads to the acidification of
17    ecosystems (EPA 2008). In the process of acidification, geochemical components of terrestrial
18    and freshwater aquatic ecosystems are altered in a way that leads to effects on  biological
19    organisms. Deposition to terrestrial ecosystems  often moves through the soil and eventually
20    leaches into adjacent water bodies, moreover deposition to the land effects the water as well.
21           The scientific evidence is sufficient to infer a causal relationship between acidifying
22    deposition and effects on biogeochemistry and biota in aquatic ecosystems (ISA 4.2.2). The
23    strongest evidence comes from studies of surface water chemistry in which acidic deposition is
24    observed to alter sulfate and nitrate concentrations in surface waters, sum  and surplus of base
25    cations, acid, ANC, inorganic aluminum, calcium, and surface water pH (ISA 3.2.3.2).
26    Consistent and coherent documentation from multiple studies on various species from all major
27    trophic levels of aquatic systems shows that geochemical alteration caused by acidification can
28    result in the loss of acid-sensitive biological species (ISA 3.2.3.3). For example, in the
29    Adirondacks, of the 53 fish species recorded in Adirondack lakes about half (26  species) were
30    absent from lakes with pH below 6.0 (Baker et al.,  1990b). Biological effects are linked to
      March 2010                                24                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    changes in water chemistry including ANC, inorganic Al, and pH. Decreases in ANC and pH
 2    and increases in inorganic Al concentration contribute to declines in taxonomic richness of
 3    zooplankton, macroinvertebrates, and fish, which often are sources of food for birds and other
 4    animal species in the ecosystem, as well as serving as a source of food and recreation for
 5    humans. Acidification of ecosystems has been shown to disrupt food web dynamics causing
 6    alteration to the diet, breeding distribution and reproduction of certain species of birds (ISA
 7    4.2.2.2. and Table 3-9). For example, breeding distribution of the common goldeneye
 8    (Bucephala clangula) an insectivorous duck, may be affected by changes in acidifying deposition
 9    (Longcore and Gill,  1993). Similarly, reduced prey diversity and quantity have been observed to
10    create feeding problems for nesting pairs of loons on low-pH lakes in the Adirondacks (Parker
11    1988).
12           In terrestrial ecosystems, the evidence is sufficient to infer a causal relationship between
13    acidifying deposition and changes in biogeochemistry (ISA 4.2.1.1). The strongest evidence
14    comes from studies of forested ecosystems, with supportive information on other plant
15    communities, including shrubs and lichens (ISA 3.2.2.1.). Three useful indicators of chemical
16    changes and acidification effects on terrestrial ecosystems, showing consistency and coherence
17    among multiple studies: soil base saturation, Al concentrations in soil water and soil C:N ratio
18    (ISA 3.2.2.2).
19           In soils with base saturation less than about 15 to 20% exchange ion chemistry is
20    dominated by Al (Reuss, 1983). Under this condition, responses to inputs of sulfuric acid and
21    nitric acid largely involve the release and mobilization of inorganic Al through cation exchange.
22    The effect can be neutralized by weathering from geologic parent material or base cation
23    exchange. The Ca2+ and Al in soils are  strongly influenced by soil acidification and both have
24    been shown to have quantitative links to tree health, including Al interference with Ca2+uptake
25    and Al toxicity to roots (Parker et al., 1989; U.S. EPA, 2009). Effects of nitrification and
26    associated acidification and cation leaching have been consistently shown to occur only in soils
27    with a C:N  ratio below about 20 to 25 (Aber et al., 2003; Ross et al., 2004).
28           Acidification has been shown to cause decreased growth and increased susceptibility to
29    disease and injury in sensitive tree species. Red spruce (Picea rubens) dieback or decline has
30    been observed across high elevation areas in the Adirondack, Green and White mountains
31    (DeHayes et al., 1999). The frequency of freezing injury to red spruce needles has increased over

      March 2010                                25                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    the past 40 years, a period that coincided with increased emissions of S and N oxides and
 2    increased acidifying deposition (DeHayes et al., 1999). Acidifying deposition may be
 3    contributing to episodic dieback in Sugar maple {Acer saccharum) through depletion of nutrient
 4    cations from marginal soils (Horsley et al., 2000; Bailey et al., 2004). Grasslands are likely less
 5    sensitive to acidification than forests (Blake et al., 1999; Kocky and Wilson 2001).

 6           2.1.2  What types of ecosystems are sensitive to such effects? In which ways are
 7                 these responses affected by atmospheric, ecological, and landscape factors?
 8           The intersection between current deposition loading, historic loading, and sensitivity
 9    defines the ecological vulnerability to the effects of acidification. Freshwater aquatic and
10    terrestrial ecosystems are the ecosystem types which are most sensitive to acidification. The ISA
11    reports that the principal factor governing the sensitivity of terrestrial and aquatic ecosystems to
12    acidification from sulfur and nitrogen deposition is geology (particularly surficial geology).
13    Geologic formations having low base  cation supply generally underlie the watersheds of acid-
14    sensitive lakes and streams. Other factors that contribute to the sensitivity of soils and surface
15    waters to acidifying deposition include topography, soil chemistry, land use, and hydrologic
16    flowpath. Episodic and chronic acidification tends to occur at relatively high elevation in areas
17    that have base-poor bedrock,  high relief, and shallow soils (ISA 3.2.4.1).

18           2.1.3  What is the magnitude of ecosystem responses to acidifying deposition?
19           Terrestrial and aquatic ecosystems differ in their response to acidifying deposition.
20    Therefore the magnitude of ecosystem response is described separately for aquatic and terrestrial
21    ecosystems in the following sections.  The magnitude of response refers to both the severity of
22    effects and the spatial extent of the U.S. which is affected.

23           2.1.3.1  Aquatic
24           Freshwater ecosystem surveys and monitoring in the eastern United States have been
25    conducted by many programs since the mid-1980s, including EPA's Environmental Monitoring
26    and Assessment Program (EMAP), National Surface Water Survey  (NSWS), Temporally
27    Integrated Monitoring of Ecosystems  (TIME) (Stoddard, 1990), and Long-term Monitoring
28    (LTM) (Ford et al., 1993;  Stoddard et al., 1996) programs. Based on analyses of surface water
29    data from these programs, New England, the Adirondack Mountains, the Appalachian Mountains

      March 2010                                26                 Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    (northern Appalachian Plateau and Ridge/Blue Ridge region), and the Upper Midwest contain
 2    the most sensitive lakes and streams (i.e., ANC less than about 50 ueq/L) since the 1980s.
 3    Portions of northern Florida also contain many acidic and low-ANC lakes and streams, although
 4    the role of acidifying deposition in these areas is less clear. The western U.S. contains many of
 5    the surface waters most sensitive to potential acidification effects, but with the exception of the
 6    Los Angeles Basin and surrounding areas, the levels of acidifying deposition are low in most
 7    areas. Therefore acidic surface waters are uncommon in the western U.S., and the extent of
 8    chronic surface water acidification that has occurred in that region to date has likely been very
 9    limited (ISA 3.2.4.2 and REA 4.2.2).
10          There are a number of species including fish, aquatic insects, other invertebrates and
11    algae that are sensitive to acidification and cannot survive, compete, or reproduce in acidic
12    waters (ISA 3.2.3.3). Decreases in ANC and pH have been shown to contribute to declines in
13    species richness and abundance of zooplankton, macroinvertebrates, and fish (Keller and Gunn
14    1995; Schindler et al., 1985).  Reduced growth rates have been attributed to acid stress in a
15    number offish species including Atlantic salmon (Salmo salar), Chinook salmon (Oncorhynchus
16    tshawytscha\ lake trout (Salvelinus namaycush\ rainbow trout  (Oncorhynchis mykiss), brook
17    trout (Salvelinus Fontinalis\  and brown trout (Salmo trutta)  (Baker et  al., 1990). In response to
18    small to moderate changes in  acidity, acid-sensitive species are  often replaced by other more
19    acid-tolerant species, resulting in changes in community composition and richness. The effects of
20    acidification are continuous, with more species being affected at higher degrees of acidification.
21    At a point, typically a pH <4.5 and an ANC <0 ueq/L, complete to near-complete loss of many
22    classes of organisms occur, including fish and aquatic insect populations, whereas others are
23    reduced to only a few acidophilic forms. These changes in species integrity are because energy
24    cost in maintaining  physiological homeostasis, growth, and reproduction is high at low ANC
25    levels (Schreck, 1981, 1982; Wedemeger et al., 1990; REA appendix  2.3). Decreases in species
26    richness related to acidification have been observed in the Adirondack Mountains and Catskill
27    Mountains of New York (Baker et al., 1996), New England and Pennsylvania (Haines and Baker,
28    1986), and Virginia (Bulger et al., 2000).
29          From the sensitive areas identified by the ISA, further "case study" analyses on aquatic
30    ecosystems in the Adirondack Mountains and Shenandoah National Park were conducted to
31    better characterize ecological  risk associated with acidification (REA Chapter 4).

      March 2010                               27                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX


 1           In the literature, ANC is the most widely used indicator of acid sensitivity and has been

 2    found in various studies to be the best single indicator of the biological response and health of

 3    aquatic communities in acid-sensitive systems (Lien et al., 1992; Sullivan et al., 2006; ISA). In

 4    the REA, surface water trends in SC>42" and NCV concentrations and ANC levels were  analyzed

 5    to affirm the understanding that reductions in deposition could influence the risk of acidification.

 6    ANC values were categorized according to their effects on biota, as shown in Figure 2-1.

 7    Monitoring data from the EPA-administered TEVIE/LTM and EMAP programs were assessed for

 8    the years 1990 to 2006, and past, present, and future water quality levels were estimated by both

 9    steady-state and dynamic biogeochemical models.

10
11
12
13

14

15

16
           Category Label ANC Levels' Expected Ecological Effects
           Acute
           Concern
            Severe
            Concern
           Elevated
           Concern
           Moderate
           Concern
           Low
           Concern
                   <0 |.ieq.'L
                   (Acidic)
                   0-20
                   20-50 ueq/L
                   50-100
                    :-100 ueq/L
Near complete loss offish population? is expected. Planktonic
communities have extremely low diversity and are dominated by
acidophihc forms. The number of individuals in plankton species that
are present is greatly reduced.
Highly sensitive to episodic acidification. During episodes of high
acidifying deposition, brook trout populations may experience lethal
effects. Diversity and distribution of zooplankton communities decline
sharply.
Fish species richness is greatly reduced (i.e., more than half of expected
species can be missing). On average, brook trout populations
experience sublethal effects, including loss of health, reproduction
capacity, and fitness. Diversity and distribution of zooplankton
communities decline.
Fish species richness begins to decline (i.e., sensitive species are lost
from lakes). Brook trout populations are sensitive and variable, with
possible sublethal effects. Diversity and distribution of zooplankton
communities also begin to decline as species that are sensitive to
acidifying deposition are affected.
Fish species richness may be unaffected, Reproducing brook trout
populations are expected where habitat is suitable. Zooplankton
communities are unaffected and exhibit expected diversity and
distribution.
       Figure 2-1. Ecological Effects Associated with Alternative Levels of Acid
       Neutralizing Capacity (ANC)

       The analyses of the Adirondack Case Study Area indicated that although wet deposition

rates for 862 and NOX have been reduced since the mid-1990s, current concentrations are still

well above pre-acidification (1860) conditions. Modeling predicts NCV and SO42" are 17- and 5-
      March  2010
                                               28
                                    Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
 1
 2
 3
 4
 5
 6
 7
 9
10
11
12
fold higher today, respectively. The estimated average ANC across the 44 lakes in the
Adirondack Case Study Area is 62.1 ueq/L (± 15.7 ueq/L); 78 % of all monitored lakes in the
Adirondack Case Study Area have a current risk of Elevated, Severe, or Acute. Of the 78%, 31%
experience episodic acidification, and 18% are chronically acidic today (REA 4.2.4.2).
       Based on a deposition scenario that maintains current emission levels to 2020 and 2050,
the simulation forecast indicates no improvement in water quality in the Adirondack Case Study
Area. The percentage of lakes within the Elevated to Acute Concern classes remains the same in
2020 and 2050.
                    o    140
                    ^    120
                    Id100
                         80
                    o
                    40
                    20
                     0
                           1850
                                  1900
1950
2000
2050
       Figure 2-2. Average N(V concentrations (orange), SC>42" concentrations (red),
       and ANC (blue) across the 44 lakes in the Adirondack Case Study Area modeled
       using MAGIC for the period 1850 to 2050.
     March 2010
                                          29
               Draft - Do Not Quote or Cite

-------
          Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
1
2
3
4
5
                        ANC Preacidification (1860) and Current Condition (2006)
                         Preacidification (1860)
                                           ANC
                  Source: EPA 2009
                                                >0
                                                0-20
                                                20-50
                                                50-100
                                                >100
                                               Current (2006)
Figure 2-3. ANC concentrations of preacidification (1860) and current (2006)
conditions based on hindcasts of 44 lakes in the Adirondack Case Study Area
modeled using MAGIC. [Note: in this map, the symbol for red is reversed and
should be < 0. The figure will be revised in the next draft.]
    March  2010
                                    30
Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
                              Current Condition  of Acidity
                                       and Sensitivity
                             Criticial Load
                               meq/m2/yr
                             •  Highly Sensitive: < 50
                                Moderately Sensitive: 51 -100
                                Low Sensitivity: 101 -200
                             •  Not Sensitive: > 201
                             |  | Adirondack Boundary
           Source: EPA 2009
 1
 2          Figure 2-4. Critical loads of acidifying deposition that each surface water location
 3          can receive in the Adirondack Case Study Area while maintaining or exceeding
 4          an ANC concentration of 50 ueq/L based on 2002 data. Watersheds with critical
 5          load values <100 meq/m2/yr (red and orange circles) are most sensitive to surface
 6          water acidification, whereas watersheds with values >100 meq/m2/yr (yellow and
 7          green circles) are the least sensitive sites.
 8          It is important to note that studies on fish species richness in the Adirondacks Case Study
 9   Area demonstrated the effect of acidification; of the 53 fish species recorded in Adirondack Case
10   Study Area lakes, only 27 species were found in lakes with a pH <6.0. The 26 species missing
11   from lakes with a pH <6.0 include important recreational species, such as Atlantic salmon, tiger
12   trout (Salmo trutta X Salvelinusfontinalis), redbreast sunfish (Lepomis auritus), bluegill
13   (Lepomis macrochims), tiger musky (Esox masquinongy X Indus), walleye (Sander vitreus),
14   alewife (Alosapseudoharengus), and kokanee (Oncorhynchus nerkd) (Kretser et al., 1989), as
15   well as ecologically important minnows that are commonly eaten by sport fish.  A survey of
16   1,469 lakes in the late 1980s found 346 lakes to be devoid  offish. Among lakes with fish, there
17   was a relationship between the number offish species and  lake pH, ranging from about one
18   species per lake for lakes having a pH <4.5 to about six species per lake for lakes having a pH
     March 2010
31
Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
 1
 2
 3
 4
 5
 6
 7
22
23
24
25
      >6.5 (Driscoll et al., 2001; Kretser et al., 1989). In the Adirondacks, a positive relationship exists
      between the pH and ANC in lakes and the number offish species present in those lakes (ISA
      3.2.3.4).
             Since the mid-1990s, streams in the Shenandoah Case Study Area have shown slight
      declines in N(V and SC>4 2" concentrations in surface waters. Current concentrations are still
      above pre-acidification (1860) conditions. MAGIC modeling predicts surface water
      concentrations of NCV and SC>42" arelO- and 32-fold higher today, respectively. The estimated
 8    average ANC across 60 streams in the Shenandoah Case Study Area is 57.9 ueq/L (± 4.5 ueq/L).
 9    55% of all monitored streams in the Shenandoah Case Study Area have a current risk of
10    Elevated, Severe, or Acute.  Of the 55%, 18% experience episodic acidification, and 18% are
11    chronically acidic today (REA 4.2.4.3)
12           Based on a deposition scenario that maintains current emission levels to 2020 and 2050,
13    the simulation forecast indicates that a large number of streams still have Elevated to Acute
14    problems with acidity. In fact, from 2006 to 2050, the percentage of streams with Acute Concern
15    increases by 5%, while the percentage of streams in Moderate Concern decreases by 5%.
16           Biological effects of increased acidification documented in the Shenandoah Case Study
17    Area include a reduction in the condition factor in Blacknose Dace (Dennis and Bulgar 1995,
18    Bulgar et al., 1999) and a decrease in fish biodiversity associated with decreasing stream ANC
19    (Bulger et al., 1995; Dennis and Bulger, 1995; Dennis et al., 1995; MacAvoy  and Bulger, 1995,
20    Bulgar et al., 1999). On average, the fish species richness is lower by one fish species for every
21    21 ueq/L decrease in ANC in Shenandoah National Park streams  (ISA 3.2.3.4).
                         120
                           1850
                                        1900
 1950
Years
2000
2050
                                                              2-
            Figure 2-5. Average NOs" concentrations orange), SO4 "concentrations (red), and
            ANC (blue) levels for the 60 streams in the Shenandoah Case Study Area
            modeled using MAGIC for the period 1850 to 2050.
      March  2010
                                                32
               Draft - Do Not Quote or Cite

-------
          Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
2
3
4
5
                      ANC Preacidification (1860) and Current Condition (2006)

                        Pre-acidification (1860)               Current (2006)
                Source: EPA 2009
                                            ANC
                                     <0
                                     0-20
                                     20-50
                                     50 - 100
                                     >100
Figure 2-6. ANC levels of 1860 (preacidification) and 2006 (current) conditions
based on hindcasts of 60 streams in the Shenandoah Case Study Area modeled
using MAGIC.
    March  2010
                                    33
Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
                               Current Condition of Acidity
                                        and Sensitivity
                               Criticial Load
                                 meq/m2/yr
                                •  Highly Sensitive: < 50
                                   Moderately Sensitive: 51-100
                                   Low Sensitivity: 101 -200
                                •  Not Sensitive: > 201
                                                                Source: EPA 2009
 1
 2           Figure 2-7. Critical loads of surface water acidity for an ANC of 50 ueq/L for
 3           Shenandoah Case Study Area streams. Each dot represents an estimated amount
 4           of acidifying deposition (i.e., critical load) that each stream's watershed can
 5           receive and still maintain a surface water ANC >50 ueq/L. Watersheds with
 6           critical load values <100 meq/m2/yr (red and orange circles) are most sensitive to
 7           surface water acidification, whereas watersheds with values >100 meq/m2/yr
 8           (yellow and green circles) are the least sensitive sites.
 9           2.1.3.2  Terrestrial Acidification
10           The ISA identified a variety of indicators that can be used to measure the effects of
11    acidification in soils. Tree health has been linked to base cations (Be) in soil (such as Ca2+, Mg2+
12    and potassium), as well as soil Al content. Tree species  show similar sensitivities to Ca/Al and
13    Bc/Al soil solution ratios, therefore these  are good chemical indicators because they directly
14    relate to the biological effects. Critical Bc/Al ratios for a large variety of tree species ranged
15    from 0.2 to 0.8 (Sverdrup and Warfvinge, 1993, a meta-data analysis of laboratory and field
16    studies). This range is similar to critical ratios of Ca/Al. Plant toxicity or nutrient antagonism
17    was reported to occur at Ca/Al ratios ranging from 0.2 to 2.5  (Cronan and Grigal, 1995; meta-
18    data assessment) (REA pg 4-54, REA Appendix 5).
      March 2010
34
Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1           There has been no systematic national survey of terrestrial ecosystems to determine the
 2    extent and distribution of terrestrial ecosystem sensitivity to the effects of acidifying deposition.
 3    However, one preliminary national evaluation estimated that -15% of forest ecosystems in the
 4    U.S. exceeds the estimated critical load based on soil chemistry for S and N deposition by >250
 5    eq ha"1 yr"1 (McNulty et al., 2007). Forests of the Adirondack Mountains of New York, Green
 6    Mountains of Vermont, White Mountains of New Hampshire, the Allegheny Plateau of
 7    Pennsylvania, and high-elevation forest ecosystems in the southern Appalachians are the regions
 8    most sensitive to terrestrial acidification effects from acidifying deposition (ISA 3.2.4.2). While
 9    studies show some recovery of surface waters, there are widespread measurements of ongoing
10    depletion of exchangeable base cations in forest soils in the northeastern U.S. despite recent
11    decreases in acidifying deposition, indicating a slow recovery time.
12           In the REA, a critical load analysis was performed for sugar maple and red spruce forests
13    in the eastern United States by using Bc/Al ratio in acidified forest soils as an indicator to assess
14    the impact of nitrogen and sulfur deposition on tree health. These are the two most commonly
15    studied species in North America for effects of acidification. At a Bc/Al ratio of 1.2, red spruce
16    growth can be reduced by 20%. Sugar maple growth can be reduced by 20% at a Bc/Al ratio of
17    0.6. The REA analysis determined the health of at least a portion of the sugar maple and red
18    spruce growing in the United States may have been compromised with acidifying total nitrogen
19    and sulfur deposition in 2002. Specifically, total nitrogen and sulfur deposition levels exceeded
20    three selected critical loads for tree growth in 3% to 75% of all sugar maple plots across 24
21    states. For Red Spruce, total nitrogen and sulfur deposition levels exceeded three selected critical
22    loads in 3% to 36% of all red spruce plots across eight states.

23           2.1.4  What are the key uncertainties associated with acidification?
24           There are different levels of uncertainty associated with relationships between deposition,
25    ecological effects and ecological indicators. In Chapter 7 of the REA, key uncertainties are
26    characterized as follows to evaluate the strength of the scientific basis for setting a national
27    standard to  protect against a given effect (REA 7.0):
28      •   Data Availability: high, medium or low quality. This criterion is based on the availability
29          and robustness of data sets, monitoring networks, availability of data that allows for
30          extrapolation to larger assessment areas, and input parameters for modeling and
      March 2010                                35                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1          developing the ecological effect function. The scientific basis for the ecological indicator
 2          selected is also incorporated into this criterion.
 3      •   Modeling Approach: high, fairly high, intermediate, or low confidence. This value is
 4          based on the strengths and limitations of the models used in the analysis and how accepted
 5          they are by the scientific community for their application in this analysis.
 6      •   Ecological Effect Function: high, fairly high, intermediate, or low confidence. This
 7          ranking is based on how well the ecological effect function describes the relationship
 8          between atmospheric deposition and the ecological indicator of an effect.

 9          2.1.4.1  Aquatic Acidification
10          The REA concludes that the available data are robust and considered high quality. There
11    is high confidence about the use of these data and their value for extrapolating to a larger
12    regional population of lakes. The EPA TIME/LTM network represents a source of long-term,
13    representative sampling. Data on sulfate concentrations,  nitrate concentrations and ANC from
14    1990 to 2006 used for this analysis as well as EPA EMAP and REMAP surveys, provide
15    considerable data on surface water trends.
16          There is fairly high confidence associated with modeling and input parameters.
17    Uncertainties in water quality  estimates (.i.e. ANC) from MAGIC was derived from multiple site
18    calibrations. The 95% confidence interval  for pre-acidification  of lakes was an average of 15
19    |j,eq/L difference in ANC  concentrations or 10% and 8 |j,eq/L or 5% for streams (REA 7.1.2) The
20    use of the critical load model used to estimate aquatic critical loads is limited by the uncertainties
21    associated with  runoff and surface water measurements and in estimating the catchment supply
22    of base cations from the weathering of bedrock and soils (McNulty et al., 2007). To propagate
23    uncertainty in the model parameters, Monte Carlo methods were employed to develop an inverse
24    function of exceedences. There is high confidence associated with the ecological effect function
25    developed for aquatic acidification. In calculating the ANC function, the depositional load for N
26    or S is fixed by the deposition of the other, so deposition for either will never be zero (Figure
27    7.1-6 REA).
28          Terrestrial Acidification
29          The available data used to quantify the targeted effect of terrestrial acidification are
30    robust and considered high quality. The USFS-Kane experimental forest and  significant amounts

      March 2010                                36                 Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    of research work in the Allegheny Plateau have produced extensive, peer-reviewed datasets. A
 2    meta-analysis of laboratory studies showed that tree growth was reduced by 20% relative to
 3    controls for BC/A1 ratios (ISA 7.2.1 and Figure 7.2-1). Sugar maple and red spruce were the
 4    focus of the REA since they are demonstrated to be negatively affected by Ca2+ depletion and
 5    high concentrations of available Al, and occur in areas that receive high acidifying deposition,
 6    There is high confidence about the use of the REA terrestrial acidification data and their value
 7    for extrapolating to a larger regional population of forests.
 8          There is high confidence associated with the models, input parameters, and assessment of
 9    uncertainty used in the case study for terrestrial acidification. The Simple Mass Balance (8MB)
10    model, a commonly used and widely applied approach for estimating critical loads, was used in
11    the REA analysis (ISA 7.2.2). There is fairly high confidence associated with the ecological
12    effect function developed for terrestrial acidification (REA 7.2.3).

13    2.2   NITROGEN ENRICHMENT: EVIDENCE OF EFFECTS ON
14          STRUCTURE AND FUNCTION OF TERRESTRIAL AND
15          FRESHWATER ECOSYSTEMS
16          The following summary is a concise overview of the known or anticipated effects caused
17    by nitrogen nutrient enrichment to ecosystems within the United States. Nutrient-enrichment
18    affects terrestrial, freshwater and estuarine ecosystems. Nitrogen deposition is often the main
19    source of anthropogenic nitrogen in terrestrial and freshwater ecosystems. In contrast, nitrogen
20    deposition often contributes to nitrogen-enrichment effects in estuaries, but does not drive the
21    effects. Both oxides of nitrogen and reduced forms of nitrogen, e.g. NHX, contribute to nitrogen
22    deposition. For the most part, nitrogen effects on ecosystems do not depend on whether the
23    nitrogen is in oxidized or reduced form. Thus, this summary focuses on the effects of nitrogen
24    deposition in total. We address the issue of incorporating the relative contributions of oxidized
25    and reduced nitrogen into the standards in Chapters 5, 6, and 8.

26          2.2.1   What is the nature of terrestrial and freshwater ecosystem responses to
27                 reactive nitrogen and/ sulfur deposition?
28          The ISA found that deposition of nitrogen, including NOX and NHX leads to the nitrogen
29    enrichment of ecosystems  (EPA 2008). In the process of nitrogen enrichment, geochemical

      March  2010                               37                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    components of terrestrial and freshwater aquatic ecosystems are altered in a way that leads to
 2    effects on biological organisms.
 3          The evidence is sufficient to infer a causal relationship between N deposition and the
 4    alteration of biogeochemical cycling in terrestrial ecosystems (ISA 4.3.1.1 and 3.3.2.1). This is
 5    supported by numerous observational, deposition gradient and field addition experiments.
 6    Stoddard (1994) identified the leaching of N(V in soil drainage waters and the export of N(V in
 7    steam water as two of the primary indictors of N enrichment.  Several N-addition studies indicate
 8    that MV leaching is induced by chronic additional of N (Edwards et al., 2002b; Kahl et al.,
 9    1999; Peterjohn et al., 1996; Norton et al., 1999). Aber et al. (2003) found that surface water
10    MV concentrations exceeded 1 |j,eq/L in watersheds receiving about 9 to 13 kg N/ha/yr of
11    atmospheric N deposition. N deposition disrupts the nutrient balance of ecosystems with
12    numerous biogeochemical effects. The chemical indicators that are typically measured include
13    NO3- leaching, C:N ratio, N mineralization, nitrification, denitrification, foliar N concentration,
14    and soil water NOs - and NH4+ concentrations. Note that N saturation (N leaching from
15    ecosystems) does not need to occur to cause effects. Substantial leaching of NOs- from forest
16    soils to stream water can acidify downstream waters, leading to effects described in the previous
17    section on aquatic acidification. Due to the complexity of interactions between the N and C
18    cycling, the effects of N on C budgets (quantified input and output of C to the ecosystem) are
19    variable. Regional trends in net ecosystem productivity (NEP) of forests (not managed for
20    silviculture) have been estimated through models based on gradient studies and meta-analysis.
21    Atmospheric N deposition has been shown to cause increased litter accumulation and carbon
22    storage in above-ground woody biomass. In the West, this has lead to increased susceptibility to
23    more severe fires. Less is known regarding the effects of N deposition on C  budgets of non-
24    forest ecosystems.
25          The evidence is sufficient to infer a causal relationship between N deposition on the
26    alteration of species richness, species composition and biodiversity in terrestrial ecosystems (ISA
27    4.3.1.2). The most sensitive terrestrial taxa are lichens. Empirical evidence indicates that lichens
28    in the U.S. are affected by deposition levels as low as 3 kg N/ha/yr. Alpine ecosystems are also
29    sensitive to N deposition, changes in an individual species (Carex rupestris) were estimated to
30    occur at deposition levels near 4 kg /ha/yr and modeling indicates that deposition levels near 10
31    kg N/ha/yr alter plant community assemblages. In several grassland ecosystems, reduced species

      March 2010                                38                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    diversity and an increase in non-native, invasive species are associated with N deposition (Clark
 2    and Tillman, 2008; Schwinning et al., 2005).
 3          In freshwater ecosystems, the evidence is sufficient to infer a causal relationship between
 4    N deposition and the alteration of biogeochemical cycling in freshwater aquatic ecosystems (ISA
 5    3.3.2.3). N deposition is the main source of N enrichment to headwater streams, lower order
 6    streams and high elevation lakes. The most common chemical indicators that were studied
 7    included NOs- and dissolved inorganic nitrogen (DIN) concentration in surface waters as well as
 8    Chi a:total P ratio. Elevated surface water NOs- concentrations occur in both the eastern and
 9    western U.S. Bergstrom and Jansson (2006) report a significant correlation between N deposition
10    and lake biogeochemistry by identifying a correlation between wet deposition and [DIN] and Chi
11    a: Total P. Recent evidence provides examples of lakes and streams that are limited by N and
12    show signs of eutrophication in response to N addition.
13          The evidence is sufficient to infer a causal relationship between N deposition and the
14    alteration of species richness, species composition and biodiversity in freshwater aquatic
15    ecosystems (ISA 3.3.5.3). Increased N deposition can cause a shift in community composition
16    and reduce algal biodiversity, especially in sensitive oligotrophic lakes.

17          2.2.2  What types of ecosystems are sensitive to such effects? How are these
18                 responses affected by atmospheric, ecological, and landscape factors
19          The numerous ecosystem types that occur across the U.S. have a broad range of
20    sensitivity to N deposition. Organisms in their natural environment are commonly adapted to a
21    specific regime of nutrient availability. Change in the availability of one important nutrient, such
22    as N, may result in imbalance in ecological stoichiometry, with effects on ecosystem processes,
23    structure and function (Sterner and Elser, 2002). In general, N deposition to terrestrial
24    ecosystems causes accelerated growth rates in some species,  which may lead to altered
25    competitive interactions among species and nutrient imbalances, ultimately affecting
26    biodiversity. The onset of these effects occurs with N deposition levels as low as 3 kg N/ha/yr in
27    sensitive terrestrial ecosystems. In aquatic ecosystems, N that is both leached from the soil and
28    directly deposited can pollute surface water. This causes alteration of the diatom community at
29    levels as low as  1.5 kg N/ha/yr in sensitive freshwater ecosystems.
      March 2010                                39                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1           The degree of ecosystem effects lies at the intersection of N loading and N-sensitivity. N-
 2    sensitivity is predominately driven by the degree to which growth is limited by nitrogen
 3    availability. Grasslands in the western United States are typically N-limited ecosystems
 4    dominated by a diverse mix of perennial forbs and grass species (Clark and Tilman, 2008;
 5    Suding et al., 2005). A meta-analysis by Lebauer and Treseder (2008) indicated that N
 6    fertilization increased aboveground growth in all non-forest ecosystems except for deserts. In
 7    other words, almost all terrestrial ecosystems are N-limited and will be altered by the addition of
 8    anthropogenic nitrogen. Likewise, a freshwater lake or stream must be N-limited to be sensitive
 9    to N-mediated eutrophication. There are many examples of fresh waters that are N-limited or N
10    and P co-limited (ISA 3.3.3.2). In a meta-analysis that included 653 datasets, Elser et al. (2007)
11    found that N-limitation occurred as frequently as P-limitation in freshwater ecosystems.
12    Additional factors that govern the sensitivity of ecosystems to nutrient enrichment from N
13    deposition include rates and form of N deposition, elevation, climate, species composition,
14    length of growing season, and soil N retention capacity. (ISA 4.3). Less is known about the
15    extent and distribution of the terrestrial ecosystems in the U.S. that are most sensitive to the
16    effects of nutrient enrichment from atmospheric N deposition compared to acidification.

17           2.2.3  What is the magnitude of ecosystem responses to  nitrogen deposition?

18           2.2.3.1  Terrestrial
19           Little is known about the full extent and distribution of the terrestrial ecosystems in the
20    U.S. that are most sensitive to impacts caused by nutrient enrichment from atmospheric N
21    deposition. As previously stated, most terrestrial ecosystems are N-limited, therefore they are
22    sensitive to perturbation caused by N additions (LeBauer and Treseder, 2008). Effects are most
23    likely to occur where areas of relatively high atmospheric N deposition intersect with N-limited
24    plant communities. The alpine ecosystems of the Colorado Front Range, chaparral watersheds of
25    the Sierra Nevada, lichen and vascular plant communities in the San Bernardino Mountains and
26    the Pacific Northwest, and the southern California coastal sage scrub (CSS) community are
27    among the most sensitive terrestrial ecosystems. There is growing evidence that existing
28    grassland ecosystems in the western United States are being altered by elevated levels of N
29    inputs, including inputs from atmospheric deposition (Clark and Tilman, 2008; Suding et al.,
30    2005).

      March 2010                                40                 Draft  - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1           In the eastern U.S., the degree of N saturation of the terrestrial ecosystem is often
 2    assessed in terms of the degree of NOs- leaching from watershed soils into ground water or
 3    surface water. Stoddard (1994) estimated the number of surface waters at different stages of
 4    saturation across several regions in the eastern U.S. Of the 85 northeastern watersheds examined
 5    60% were in Stage 1 or Stage 2 of N saturation on a scale of 0 (background or pretreatment) to 3
 6    (visible decline). Of the northeastern sites for which adequate data were available for assessment,
 7    those in Stage 1 or 2 were most prevalent in the Adirondack and Catskill Mountains. Effects on
 8    individual plant species have not been well studied in the U.S. More is known about the
 9    sensitivity of particular plant communities. Based largely on results obtained in more extensive
10    studies conducted in Europe, it is expected that the more sensitive terrestrial ecosystems include
11    hardwood forests, alpine meadows, arid and semi-arid lands, and grassland ecosystems (ISA
12    3.8.2).
13           The REA used published research results (REA 5.3.1 and ISA Table 4.4) to identify
14    meaningful ecological benchmarks associated with different levels of atmospheric nitrogen
15    deposition.  These  are given by figure 2-8. The sensitive areas and ecological indicators identified
16    by the ISA were analyzed further in the REA to create a national map that illustrates effects
17    observed from ambient and experimental atmospheric nitrogen deposition loads in relation to
18    CMAQ 2002 modeling results and NADP monitoring data. This map, reproduced in Figure 2-9,
19    depicts the sites where empirical effects of terrestrial nutrient enrichment have been observed
20    and site proximity to elevated atmospheric N deposition.
21
      March 2010                                41                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
                          Rocky Mountain alpine lakes: shift in diatom community dominance (Baron, 2006)


                           •  Southern California: CSS loss (Wood et al., 2006)
                           •  San Bernardino Mountains and Sierra Nevada Mountains: acidophytic lichen
                              decline in MCF (Fenn et al., 2008)

                          •  Eastern Rocky Mountain Slope: low carbon:nitrogen; low lignin:nitrogen (Baron et
                             al.,2000)
                          •  Eastern Rocky Mountain Slope: increased foliar nitrogen; increased mineralization
                             (Baron et al., 2000)

                             •   San Bernardino Mountains and Sierra Nevada Mountains: shift from acidophytic
                                to neutral or nitrogen-tolerant lichen in MCF (Fenn et al., 2008)
                             •   Minnesota grasslands: decreased plant species (Clark and Tilman, 2008)

                               • Northeast U.S.: NO3 leaching (Aber et al., 2003)
                                    Bay Area, CA: Increased cover of nonnative grasses; decreased native
                                    grasses (Weiss, 1999)

                                    San Bernardino Mountains and Sierra Nevada Mountains: loss of acidophytic
                                    lichen in MCF (Fenn et al., 2008)
                                    Southern California: shift in mycorrhizal species in CSS (Egerton-Warburton
                                    and Allen, 2000)
                                    Southern California: shift from native species to invasive grasses in CSS (Allen,
                                    2008)
                                    •  San Bernardino Mountains: high dissolved organic nitrogen (Meixner
                                       and Fenn, 2004)
                                    •  San Bernardino Mountains: nitrogen saturation (Fenn et al., 2000)

                                    •  Increased nitrogen in lichen (Fenn et al., 2007)
                                                  MCF: N03 leaching (Fenn et al., 2008)
                                                  MCF: 25% decrease in fine-root biomass (Fenn et al., 2008)


                                                  •  Southern California: NO3" leaching (Fenn et al., 2003)
                                                  •  Southern California: high foliar nitrogen (Bytnerowicz and
                                                     Fenn, 1996)
                                                  •  Los Angeles Basin, California: High NO emissions
                                                     (Bytnerowicz and Fenn,  1996)
                                                                    Fraser Experimental Forest, CO:
                                                                    increased foliar nitrogen; increased
                                                                    mineralization (Rueth et al., 2003)
1
2
3
0246  8  10  12  14  16  18 20  22  24 26  28  30 32  34  36 38  40  42  44 46  48  50

                         Nitrogen  Deposition, kg/ha/yr

Figure 2-8. Benchmarks of atmospheric nitrogen deposition for several
ecosystem indicators with the inclusion of the diatom changes in the Rocky
Mountain lakes (REA 5.3.1.2)
     March  2010
                                          42
Draft - Do Not Quote or Cite

-------
             Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
      2.
      3.
      4.
      5.

      6.
      7.
      9.
      10.
      11.

      12.
      13.

      14.
      15.

      16.
      17.
      18.
      19.

      20.
                       Legend
                Total N Deposition
                  High S6M7

                  Low 0.761
                  ) Loutont

                  I Mitonil P»*i
Nitrogen enrichment or eutrophication of lakes (Loch Vale, CO: 0.5 to1.5 kg/ha/yr; Niwot Ridge, CO: 4.71
kg/ha/yr)
Alpine lakes increase shift in diatom species (Rocky Mountains, CO: 2 kg/ha/yr)
Alpine meadows' elevated NOr levels in runoff (Colorado Front Range: 20, 40, 60 kg/ha/yr)
Alpine meadows' shift toward hairgrass  (Niwot Ridge, CO: 25 kg/ha/yr)
Nitrogen enrichment or nitrogen saturation (e.g., soil and foliar nitrogen concentration) (eastern slope of Rocky
Mountains: 1.2, 3.6 kg/ha/yr; Fraser Forest, CO: 3.2 to 5.5 kg/ha/yr)
Increased nitrogen mineralization rates and nitrification (Loch Vale, CO (spruce): 1.7 kg/ha/yr)
Alpine tundra with increased plant foliage and decreased species richness (Niwot Ridge, CO: 50 kg/ha/yr)
Nitrogen saturation, high N0s~ in streamwater, soil, leaves; high nitric oxide (NO) emissions (Los Angeles, CA, air
basin: saturation at 24 to 25 kg/ha/yr (dry) and at 0.8 to 45 kg/ha/yr (wet); northeastern U.S.: 3.3 to 12.7 kg/ha/yr)
Nitrogen saturation, high NOs- in streamwater (San  Bernardino Mountains, CA (coniferous): 2.9 and 18.8 kg/ha/yr)
 NOs- leaching (New England; Adirondack lakes: 8 to10 kg/ha/yr)
Nitrogen saturation, high dissolved inorganic nitrogen (San Bernardino Mountains, San Gabriel Mountains, CA,
chaparral, hardwood, coniferous):  11 to  40 kg/ha/yr)
Increased tree mortality and beetle activity (San Bernardino Mountains, CA (Ponderosa): 8 and 82 kg/ha/yr)
Enhanced growth of black cherry and yellow poplar; possible decline in red maple vigor; increased foliar nitrogen
(Fernow Forest, WV: 35.5 kg/ha/yr)
Impacts on lichen communities (California MCF: 3.1 kg/ha/yr; Columbia R. Gorge, OR/WA: 11/5 to 25.4)
Evidence that threatened and endangered species impacted San Francisco Bay, CA (checkerspot butterfly and
serpentinitic grass invasion: 10 to15 kg/ha/yr; Jasper Ridge, CA: 70 kg/ha/yr)
Decreased diversity of mycorrhizal communities (Southern California: -10 kg/ha/yr)
Decreased abundance of CSS (Southern California: 3.3 kg/ha/yr)
Loss of grasslands (Cedar Creek,  MN: 5.3 [1.3 to 9.8] kg/ha/yr)
Decrease in abundance of desert creosote bush, increase in nonnative grasses (Mojave Desert and Chihuahuan
Desert, CA: 1.7 kg/ha/yr and up)
Decrease in pitcher plant population growth rate (Hawley Bog, MA and Molly Bog, VA: 10 to14 kg/ha/yr)	
1
2
3
4
    Figure 2-9 (from REA figure 5.3-9). Observed effects from ambient and
    experimental atmospheric nitrogen deposition loads in relation to using CMAQ
    2002 modeling results and NADP monitoring data. Citations for effect results are
    from the ISA, Table 4.4 (U.S. EPA, 2008).
      March  2010
                                                  43
Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1          Based on information in the ISA and initial analysis in the REA, further case study
 2    analyses on terrestrial nutrient enrichment of ecosystems were developed for the CCS
 3    community and Mixed Conifer Forest (MCF) (EPA 2009). Geographic information systems
 4    (GIS) analysis supported a qualitative review of past field research to identify ecological
 5    benchmarks associated with CSS and mycorrhizal communities, as well as MCF's nutrient-
 6    sensitive acidophyte lichen communities, fine-root biomass in Ponderosa pine, and leached
 7    nitrate in receiving waters.
 8          The ecological benchmarks that were identified for the CSS and the MCF are included in
 9    the suite of benchmarks  identified in the ISA (ISA 3.3). There are sufficient data to confidently
10    relate the ecological effect to a loading of atmospheric nitrogen. For the CSS community, the
11    following ecological benchmarks were identified:
12      •   3.3 kg N/ha/yr - the amount of nitrogen uptake by a vigorous stand of CSS; above this
13         level, nitrogen may no longer be limiting
14      •   10 kg N/ha/yr - mycorrhizal community changes
15          For the MCF community, the following ecological benchmarks were identified:
16      •   3.1 kg N/ha/yr - shift from sensitive to tolerant lichen species
17      •   5.2 kg N/ha/yr-dominance of the tolerant lichen species
18      •   10.2 kg N/ha/yr-loss of sensitive lichen species
19      •   17 kg N/ha/yr - leaching of nitrate into streams.
20          These benchmarks, ranging from 3.1 to 17 kg N/ha/yr, were compared to 2002
21    CMAQ/NADP data to discern any associations between atmospheric deposition and changing
22    communities. Evidence supports the finding that nitrogen  alters CSS and MCF. Key findings
23    include the following: 2002 CMAQ/NADP nitrogen deposition data show that the 3.3 kg N/ha/yr
24    benchmark has been exceeded in more than 93% of CSS areas (654,048 ha). These deposition
25    levels are a driving force in the degradation of CSS communities.  Although  CSS decline has
26    been observed in the absence of fire, the contributions of deposition and fire to the CSS decline
27    require further research.  CSS is fragmented into many small parcels, and the 2002
28    CMAQ/NADP 12-km grid data are not fine enough to fully validate the relationship between
29    CSS distribution, nitrogen deposition, and fire. 2002 CMAQ/NADP nitrogen deposition data
30    exceeds the 3.1 kg N/ha/yr benchmark in more than 38% (1,099,133 ha)  of MCF areas, and


      March 2010                               44                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    nitrate leaching has been observed in surface waters. Ozone effects confound nitrogen effects on
 2    MCF acidophyte lichen, and the interrelationship between fire and nitrogen cycling requires
 3    additional research.

 4          2.2.3.2  Freshwater
 5          The magnitude of ecosystem response may be thought of on two time scales, current
 6    conditions and how ecosystems have been altered since the onset of anthropogenic N deposition.
 7    As noted previously, Elser et al. (2008) found that N-limitation occurs as frequently as P-
 8    limitation in freshwater ecosystems  (ISA 3.3.3.2). Recently, a comprehensive study of available
 9    data from the northern hemisphere surveys of lakes along gradients of N deposition show
10    increased inorganic N concentration and productivity to be correlated with atmospheric N
11    deposition (Bergstrom and Jansson 2006). The results are unequivocal evidence of N limitation
12    in lakes with low ambient inputs of N, and increased N concentrations in lakes receiving N
13    solely from atmospheric N deposition (Bergstrom and Jansson, 2006). These authors suggested
14    that most lakes in the northern hemisphere may have originally been N-limited, and that
15    atmospheric N deposition has changed the balance of N and P in lakes.
16          Available data suggest that the increases in total N deposition do not have to be large to
17    elicit an ecological effect. For example, a hindcasting exercise determined that the change in
18    Rocky Mountain National Park lake algae that occurred between 1850 and 1964 was associated
19    with an increase in wet N deposition that was only about 1.5 kg N/ha (Baron, 2006). Similar
20    changes inferred from lake sediment cores of the Beartooth Mountains of Wyoming also
21    occurred at about 1.5  kg N/ha deposition (Saros et al., 2003). Pre-industrial  inorganic N
22    deposition is estimated to have been only 0.1 to 0.7 kg N/ha based on measurements from remote
23    parts of the world (Galloway  et al., 1995; Holland et al., 1999). In the western U.S., pre-
24    industrial, or background, inorganic N deposition was estimated by (Holland et al., 1999) to
25    range from 0.4 to 0.7  kg/ha/yr.
26          Eutrophication effects from N deposition are most likely to be manifested in undisturbed,
27    low nutrient surface waters such as those found in the higher elevation areas of the western U.S.
28    The most severe eutrophication from N deposition effects is expected downwind of major urban
29    and agricultural centers. High concentrations of lake or streamwaterNO3-, indicative of
30    ecosystem saturation, have been found  at a variety of locations throughout the U.S., including the


      March 2010                                45                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    San Bernardino and San Gabriel Mountains within the Los Angeles Air Basin (Fenn et al., 1996),
 2    the Front Range of Colorado (Baron et al., 1994; Williams et al., 1996), the Allegheny mountains
 3    of West Virginia (Gilliam et al., 1996), the Catskill Mountains of New York (Murdoch and
 4    Stoddard, 1992; Stoddard, 1994), the Adirondack Mountains of New York (Wigington et al.,
 5    1996), and the Great Smoky Mountains in Tennessee (Cook et al.,  1994) (ISA 3.3.8).

 6          2.2.3.3  Nitrogen Enrichment: Evidence of Effects on Estuaries
 1          In contrast to terrestrial and freshwater systems, atmospheric N load to estuaries
 8    contributes to the total load but does not necessarily drive the effects. In estuaries, N-loading
 9    from multiple anthropogenic and non-anthropogenic pathways leads to water quality
10    deterioration, resulting in numerous effects including hypoxic zones, species mortality, changes
11    in community composition and harmful algal blooms that are indicative of eutrophication. The
12    following summary is a concise overview of the known or anticipated effects of nitrogen
13    enrichment on estuaries within the United States.
14          2.2.3.3.1   What is the nature of estuary responses to reactive nitrogen andsulfur
15                     deposition?
16          In the ISA, the evidence is sufficient to infer a causal relationship between Nr deposition
17    and the biogeochemical cycling of N and C in estuaries (ISA 4.3.4.1 and 3.3.2.3). In general,
18    estuaries tend to be nitrogen-limited, and many currently receive high levels of nitrogen input
19    from human  activities (REA 5.1.1). It is unknown if atmospheric deposition alone is sufficient to
20    cause eutrophication, however, the contribution of atmospheric nitrogen deposition to total
21    nitrogen load is calculated for some estuaries and can be >40% (REA 5.1.1).
22          The evidence is sufficient to infer a causal relationship between N deposition and the
23    alteration of  species richness, species composition and biodiversity in estuarine ecosystems (ISA
24    4.3.4.2 and 3.3.5.4). Atmospheric and non-atmospheric sources of N contribute to increased
25    phytoplankton and algal productivity, leading to eutrophication. Shifts in  community
26    composition, reduced hypolimnetic DO, reduced biodiversity, and mortality of submerged
27    aquatic vegetation are associated with increased N deposition in estuarine systems.
      March 2010                                46                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1          2.2.3.3.2  What types of ecosystems are sensitive to such effects? How are these
 2                    responses affected by atmospheric, ecological, and landscape factors?
 3          Because the productivity of estuarine and near shore marine ecosystems is generally
 4    limited by the availability of N, they are susceptible to the eutrophication effect of N deposition
 5    (ISA 4.3.4.1). A recent national assessment of eutrophic conditions in estuaries found the most
 6    eutrophic estuaries were generally those that had large watershed-to-estuarine surface area, high
 7    human population density, high rainfall and runoff, low dilution, and low flushing rates (Bricker
 8    et al., 2007). In the REA, the National Oceanic and Atmospheric Administration's (NOAA)
 9    National Estuarine Eutrophi cation Assessment (NEEA) assessment tool, Assessment of
10    Estuarine Tropic Status (ASSETS) categorical Eutrophication Index (El) (Bricker et al., 2007)
11    was used to evaluate eutrophi cation due to  atmospheric loading of nitrogen. ASSETS El is an
12    estimation of the likelihood that an estuary  is experiencing eutrophi cation or will experience
13    eutrophi cation based on five ecological indicators: chlorophyll a, macroalgae, dissolved oxygen,
14    nuisance/toxic algal blooms and submerged aquatic vegetation (SAV) (Bricker et al., 2007).
15          In the REA, two regions were selected for case  study analysis using ASSETS El, the
16    Chesapeake Bay and Pamlico Sound. Both  regions received an ASSETS El rating of Bad
17    indicating that the estuary had moderate to  high pressure due to overall human influence and  a
18    moderate high to high eutrophic condition (REA 5.2.4.1 and 5.2.4.2). These results were then
19    considered with SPAtially Referenced Regression (SPARROW) modeling to develop a response
20    curve to examine the role of atmospheric nitrogen deposition in achieving desired reduction load.
21    To change the Neuse River Estuary' s El score from Bad to Poor not only must 100% of the total
22    atmospheric nitrogen deposition be eliminated, but considerably more nitrogen from other
23    sources as well must be reduced (REA section 5.2.7.2). In the Potomac River estuary, a 78%
24    reduction of total nitrogen could move the El score from Bad to Poor (REA 5.2.7.1). The results
25    of this analysis indicated reductions in atmospheric deposition alone could not solve coastal
26    eutrophi cation problems due to multiple non-atmospheric nitrogen inputs (REA 7.3.3). However,
27    by reducing atmospheric contributions, it may help avoid the need for more costly controls on
28    nitrogen from other sources.
29          In general, estuaries tend to be N-limited (Elser et al., 2008), and many currently receive
30    high levels of N input from human activities to cause eutrophi cation (Howarth et al., 1996;
31    Vitousek and Howarth, 1991).  Atmospheric N loads to estuaries in the U.S. are estimated to

      March 2010                               47                 Draft - Do Not Quote or  Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    range from 2-8% for Guadalupe Bay, TX on the lowest end to as high as 72% for St Catherines-
 2    Sapelo estuary, GA (Castro et al., 2003). The Chesapeake Bay is an example of a large, well-
 3    studied and severely eutrophic estuary that is calculated to receive as much as 30% of its total N
 4    load from the atmosphere.
 5          2.2.3.3.3   What is the magnitude of ecosystem responses to eutrophication?
 6          There is a scientific consensus that nitrogen-driven eutrophication in shallow estuaries
 7    has increased over the past several decades and that the environmental degradation of coastal
 8    ecosystems due to nitrogen, phosphorus, and other inputs is now a widespread occurrence (Paerl
 9    et al., 2001). For example, the frequency of phytoplankton blooms and the extent and severity of
10    hypoxia have increased in the Chesapeake Bay (Officer et al., 1984) and Pamlico estuaries in
11    North Carolina (Paerl et al., 1998) and along the continental shelf adjacent to the Mississippi and
12    Atchafalaya rivers' discharges to the Gulf of Mexico (Eadie et al.,  1994).
13          A recent national assessment of eutrophic conditions in estuaries found that 65% of the
14    assessed systems had moderate to high overall eutrophic conditions and generally received the
15    greatest N loads from all sources, including atmospheric and land-based sources (Bricker et al.,
16    2007). Most eutrophic estuaries occurred in the mid-Atlantic region and the estuaries with the
17    lowest degree of eutrophication were in the North Atlantic (Bricker et al., 2007). Other regions
18    had mixtures of low, moderate, and high degree of eutrophication (ISA 4.3.4.3).
19          The mid-Atlantic region is the most heavily impacted area in terms of moderate or high
20    loss of submerged aquatic vegetation due to  eutrophication (ISA 4.3.4.2). Submerged aquatic
21    vegetation is important to the quality of estuarine ecosystem habitats because it provides habitat
22    for a variety of aquatic organisms, absorbs excess nutrients, and traps  sediments (ISA 4.3.4.2). It
23    is partly because many estuaries and near-coastal marine waters are degraded by nutrient
24    enrichment that they are highly sensitive to potential negative impacts from nitrogen addition
25    from atmospheric deposition.

26          2.2.4  What are the key uncertainties associated with nutrient enrichment?
27          There are different levels of uncertainty associated with relationships between deposition,
28    ecological effects  and ecological indicators.  The criteria used in the REA to evaluate the degree
29    of confidence in the data, modeling and ecological effect function are  detailed in Chapter 7 of the
30    REA and summarized in section 2.1.4 of this chapter (REA 7.0).

      March 2010                                48                 Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1          Aquatic
 2          The approach for assessing atmospheric contributions to total nitrogen loading in the
 3   REA, was to consider the main-stem river to an estuary (including the estuary) rather than an
 4   entire estuary system or bay. The biological indicators used in the NOAA ASSETS El required
 5   the evaluation of many national databases including the USGS NAWQA files, EPA's STORET
 6   database, NOAA's Estuarine Drainage Areas data, and EPA's water quality standards nutrient
 7   criteria for rivers and lakes (REA Appendix 6, Table 1.2.-1). Both the SPARROW modeling for
 8   nitrogen loads and assessment of estuary  conditions under NOAA ASSETS El, have been
 9   applied on a national scale. The REA concludes that the available data are medium quality with
10   intermediate confidence about the use of these  data and their values for extrapolating to a larger
11   regional area (REA 7.3.1). Intermediate confidence is associated with the modeling approach
12   using ASSETS El and SPARROW. The REA states there is low confidence with the ecological
13   effect function due to the results of the analysis which indicated that reductions in atmospheric
14   deposition alone could not solve coastal eutrophication problems due to multiple non-
15   atmospheric nitrogen inputs (REA 7.3.3).

16          Terrestrial
17          Ecological thresholds are  identified for CSS and MCF and these data are considered to be
18   of high quality, however, the ability to extrapolate these data to larger regional areas is limited
19   (REA 7.4.1). No quantitative modeling was conducted or ecological effect function developed
20   for terrestrial nutrient enrichment reflecting the uncertainties associated with these depositional
21   effects.

22   2.3   WHAT ECOLOGICAL  EFFECTS ARE ASSOCIATED WITH GAS-
23          PHASE NOX AND SOX?
24          Acidifying deposition and nitrogen enrichment are the main focus of this policy
25   assessment; however, there are other known ecological effects are attributed to gas-phase NOX
26   and SOX. Acute and chronic exposures to gaseous pollutants such as sulfur dioxide (802),
27   nitrogen dioxide (NO2), nitric oxide (NO), nitric acid  (HNO3) and peroxyacetyl nitrite (PAN) are
28   associated with negative impacts  to vegetation. The current secondary NAAQS were set to
29   protect against direct damage to vegetation by exposure to gas-phase NOX or SOX, such as foliar

     March 2010                               49                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    injury, decreased photosynthesis, and decreased growth. The following summary is a concise
 2    overview of the known or anticipated effects to vegetation caused by gas phase N and S.

 3          2.3.1  What is the nature of ecosystem responses to gas-phase nitrogen and sulfur?
 4          The 2008 ISA found that gas phase N and S are associated with direct phytotoxic effects
 5    (ISA 4.4). The evidence is sufficient to  infer a causal relationship between exposure to SC>2 and
 6    injury to vegetation (ISA 4.4.1 and 3.4.2.1). Acute foliar injury to vegetation from SC>2 may
 7    occur at levels above the current secondary standard (3-h average of 0.50 ppm). Effects on
 8    growth, reduced photosynthesis and decreased yield of vegetation are also associated with
 9    increased SO2 exposure concentration and time of exposure.
10          The evidence is sufficient to infer a causal relationship between exposure to NO, NO2
11    and PAN and injury to vegetation (ISA 4.4.2 and 3.4.2.2). In sufficient concentrations,  NO, NO2
12    and PAN can decrease photosynthesis and induce visible foliar injury to plants. Evidence is also
13    sufficient to infer a causal relationship between exposure  to HNOs and changes to vegetation
14    (ISA 4.4.3 and 3.4.2.3). Phytotoxic effects  of this pollutant include damage to the leaf cuticle in
15    vascular plants and disappearance of some sensitive  lichen species.

16          2.3.2  What types of ecosystems are sensitive to such effects? How are these
17                 responses affected by atmospheric, ecological, and landscape factors?
18          Vegetation in ecosystems near sources of gaseous NOX and SOX or where ambient
19    concentrations of SC>2, NO, NO2, PAN  and  HNOs are higher are more likely to be impacted by
20    these pollutants. Uptake of these pollutants  in a plant canopy is a complex process involving
21    adsorption to surfaces (leaves, stems and soil) and absorption into leaves (ISA 3.4.2). The
22    functional relationship between ambient concentrations of gas phase NOX and SOX and specific
23    plant response are impacted by internal factors such  as rate of stomatal conductance and plant
24    detoxtification mechanisms, and external factors including plant water status, light, temperature,
25    humidity, and pollutant exposure regime (ISA 3.4.2).
26          Entry of gases into a leaf is dependent upon physical and chemical processes of gas phase
27    as well as to stomatal aperature. The aperature of the stomata is controlled largely by  the
28    prevailing environmental conditions, such as humidity, temperature, and light intensity. When
29    the stomata are closed, resistance to gas uptake is high and the plant has a very low degree of
30    susceptibility to injury. Mosses and lichens do not have a protective cuticle barrier to gaseous

      March 2010                                50                 Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    pollutants or stomata and are generally more sensitive to gaseous sulfur and nitrogen than
 2    vascular plants (ISA 3.4.2).
 3          The appearance of foliar injury can vary significantly across species and growth
 4    conditions affecting stomatal conductance in vascular plants (REA 6.4.1). For example, damage
 5    to lichens from SC>2 exposure includes reduced photosynthesis and respiration, damage to the
 6    algal component of the lichen, leakage of electrolytes, inhibition of nitrogen fixation, reduced K+
 7    absorption, and structural changes (Belnap et al., 1993; Farmer et al., 1992, Hutchinson et  al.,
 8    1996).

 9          2.3.3   What is the magnitude of ecosystem responses to gas phase effects of NOX
10                 and SOX?
11          The phytotoxic effects of gas phase NOX and SOX are dependent on the exposure
12    concentration and duration and species sensitivity to these pollutants. Effects to vegetation
13    associated with NOX and SOX, are therefore, variable across the U.S. and tend to be higher  near
14    sources of photochemical smog. For example, 862 is considered to be the primary factor
15    contributing to the death of lichens in many urban and industrial areas, with fruticose lichens
16    being more susceptible to 862 than many foliose and crustose species (Hutchinson et al., 1996).
17          The ISA states  there is very limited new research on phytotoxic effects of NO, NC>2, PAN
18    and FINOs at concentrations currently  observed in the United States with the exception of some
19    lichen species (ISA 4.4). Past and current HNOs concentrations may be contributing to the
20    decline in lichen species in the Los Angeles basin (Boonpragob and Nash 1991; Nash and  Sigal,
21    1999; Riddell et al., 2008). PAN is a very small component of nitrogen deposition in most areas
22    of the United States (REA 6.4.2). Current deposition of FINOs is contributing to N saturation of
23    some ecosystems close to sources of photochemical smog  (Fenn et al., 1998) such as the MCF's
24    of the Los Angeles basin mountain (Bytnerowicz et al.,  1999).

25    2.4   SUMMARY
26          In summary, NOX and SOX in the atmosphere contribute to effects on individual species
27    and ecosystems through direct contact with vegetation, and more significantly through deposition
28    to sensitive ecosystems. The ISA concludes that the evidence is sufficient to conclude causal
29    relationships between acidifying deposition of N and S and effects on freshwater aquatic
30    ecosystems and terrestrial ecosystems, and between nitrogen nutrient enrichment and effects on

      March 2010                                51                 Draft - Do Not Quote  or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    sensitive terrestrial and freshwater aquatic ecosystems. The ISA also concludes that a causal
 2    relationship is supported between nitrogen nutrient enrichment and effects on estuarine
 3    ecosystems; however, the contribution of atmospheric oxidized nitrogen relative to reduced
 4    nitrogen and non-atmospheric nitrogen is more difficult to determine.
 5          The REA provides additional support that under recent conditions, deposition levels have
 6    exceeded benchmarks for ecological indicators of acidification and nutrient enrichment that
 7    indicate that effects are likely to be occurring in significant numbers of lakes and streams within
 8    sensitive ecosystems.

 9    2.5   REFERENCES
10    Aber JD; Goodale CL; Ollinger SV; Smith ML; Magill AH; Martin ME; Hallett RA; Stoddard
11          JL. (2003). Is nitrogen deposition altering the nitrogen status of northeastern forests?
12          Bioscience, 53, 375-389.

13    Bailey SW; Horsley  SB; Long RP; Hallett RA. (2004). Influence of edaphic factors on sugar
14          maple nutrition and health on the Allegheny Plateau. Soil Sci Soc Am J, 68, 243-252.

15    Baker JP; Bernard DP; Christensen SW; Sale MJ. (1990). Biological effects of changes in
16          surface water acid-base chemistry. (State of science / technology report #13).Washington
17          DC: National Acid Precipitation Assessment Program  (NAPAP).

18    Baker JP; Van Sickle J;  Gagen CJ; DeWalle DR; Sharpe WE; Carline RF; Baldigo BP; Murdoch
19          PS; Bath DW; Kretser WA; Simonin HA; Wigington PJ Jr. (1996). Episodic acidification
20          of small streams in the northeastern United States: Effects on fish populations. Ecol Appl,
21          6,423-437.

22    Baron JS; Ojima DS; Holland EA; Parton WJ. (1994). Analysis of nitrogen saturation potential
23          in Rocky Mountain tundra and forest: Implications for aquatic systems. Biogeochemistry,
24          27,61-82.

25    Baron JS. (2006). Hindcasting nitrogen deposition to determine ecological critical load. Ecol
26          Appl, 16, 433-439.
      March  2010                               52                 Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    Belnap J; Sigal L; Moir W; Eversman S. (1993). Identification of sensitive species, in lichens as
 2          bioindicators of air quality. In: Huckaby LS (Ed.), Lichens as bioindicators of air quality
 3          (pp. 67-88). Fort Collins, Colorado: Rocky Mountain Forest and Range
 4          Experimental Station, U.S. Forest Service, U. S. Department of Agriculture.

 5    Bergstrom A; Jansson M. (2006). Atmospheric nitrogen deposition has caused nitrogen
 6          enrichment and eutrophication of lakes in the northern hemisphere. Glob Chang Biol, 12,
 7          635-643.

 8    Blake L; Goulding KWT; Mott CJB; Johnston AE. (1999). Changes in soil chemistry
 9          accompanying acidification over more than 100 years under woodland and grass at
10          Rothamsted Experimental Station, UK. Eur J Soil  Sci, 50, 401-412.

11    Boonpragob K; Nash THI. (1991). Physiological responses of the lichen Ramalina menziesii
12          Tayl.  to the Los Angeles urban environment. Environ Exp Bot, 31, 229-238.

13    Bricker S; Longstaff B; Dennison W; Jones A; Boicourt K; Wicks C; Woerner J. (2007). Effects
14          of nutrient enrichment in the nation's estuaries: A decade of change.
15          http://ccmaserver.nos.noaa.gov/publications/eutroupdate/. (NOAA Coastal Ocean
16          Program Decision Analysis Series No. 26).  Silver  Spring, MD: National  Centers for
17          Coastal Ocean Science, National Oceanic and Atmospheric Administration(NOAA).

18    Bulger AJ; Dolloff CA; Cosby BJ; Eshleman KN; Webb JR; Galloway JN. (1995). The
19          "Shenandoah National Park: fish in sensitive habitats" (SNP: FISH) Project. An
20          integrated assessment offish community responses to stream acidification. Water Air
21          SoilPollut, 85,  309-314.

22    Bulger AJ; Cosby, BJ;  Dolloff, CA; Eshleman, KN; Webb, JR; Galloway, JN. (1999).
23          SNP:FISH, Shenandoah National Park: Fish in sensitive habitats. Project final report -
24          Volume I: Project description and summary of results; Volume II: Stream water
25          chemistry and discharge, and synoptic water quality  surveys. Volume III: Basin-wide
26          habitat and population inventories, and behavioral responses to acid in a  laboratory
27          stream. Volume IV: Stream bioassays, aluminum toxicity, species richness and stream
28          chemistry, and models of susceptibility to acidification. (Project Completion Report to

      March 2010                               53                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1          the National Park Service). Charlottesville, VA, Department of Environmental Sciences;
 2          University of Virginia.

 3    Bulger AJ; Cosby BJ; Webb JR. (2000). Current, reconstructed past, and projected future status
 4          of brook trout (Salvelinus fontinalis) streams in Virginia. Can J Fish Aquat Sci, 57, 1515-
 5          1523.

 6    Bytnerowicz A; Padgett P; Percy K; Krywult M; Riechers G; Horn J. (1999). Direct effects of
 7          nitric acid on forest vegetation. In: Miller PR, McBride J (Eds.), Oxidant air pollution
 8          impacts in the Montane forests of Southern California: The San Bernardino Mountains
 9          case study, Ecological Series 134. (pp. 270-287). New York: Springer.

10    Castro MS; Driscoll CT; Jordan TE; Reay WG; Boynton WR. (2003). Sources of nitrogen to
11          estuaries in the United States. Estuaries, 26, 803-814.

12    Clark CM; Tilman D. (2008). Loss of plant species after chronic low-level nitrogen deposition to
13          prairie grasslands. Nature, 451, 712-715.

14    Cook RB; Elwood JW; Turner RR; Bogle MA; Mulholland PJ; Palumbo AV. (1994). Acid-base
15          chemistry of high-elevation streams in the Great Smoky Mountains. Water Air Soil
16          Pollut, 72, 331-356.

17    Cronan CS; Grigal DF. (1995). Use of calcium/aluminum ratios as indicators of stress in forest
18          ecosystems.  J Environ Qual, 24, 209-226.

19    DeHayes DH; Schaberg PG; Hawley GJ; Strimbeck GR. (1999). Acid rain impacts on calcium
20          nutrition and forest health. Bioscience, 49, 789-800.

21    Dennis TD; MacAvoy SE; Steg MB; Bulger AJ. (1995). The association of water chemistry
22          variables and fish condition in streams of Shenandoah National Park (USA). Water Air
23          Soil Pollut, 85, 365-370.

24    Dennis TE; Bulger AJ.  (1999). The susceptibility of blacknose dace (Rhinichthys atratulus).to
25          acidification in Shenandoah National Park. In: Bulger AJ; Cosby BJ; Dolloff CA;
26          Eshleman KN; Galloway JN; Webb. JR (Eds.), Shenandoah National Park: Fish in

      March 2010                               54                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1          sensitive habitats. Project final report, Volume IV: Stream bioassays, aluminum toxicity,
 2          species richness and stream chemistry, and models of susceptibility to acidification.
 3          Chapter 6B. Project completion report to the National Park Service. Charlottesville, VA;
 4          Department of Environmental Sciences, University of Virginia.

 5    Driscoll CT; Lawrence GB; Bulger AJ; Butler TJ; Cronan CS; Eagar C; Lambert KF; Likens
 6          GE; Stoddard JL; Weather KC. (200Ib). Acidic deposition in the northeastern United
 7          States: Sources and inputs, ecosystem effects, and management strategies. Bioscience,
 8          51, 180-198.

 9    Eadie BJ; McKee BA; Lansing MB; Robbins JA; Metz S; Trefry JH. (1994). Records of
10          nutrient-enhanced coastal productivity in sediments from the Louisiana continental shelf.
11          Estuaries, 17, 754-765.

12    Edwards PJ; Kochenderfer JN; Coble DW; Adams MB. (2002). Soil leachate responses during
13          10 years of induced whole-watershed acidification. Water Air Soil Pollut, 140, 99-118.

14    Elser JJ; Bracken MES;  Cleland EE; Gruner DS; Harpole WS; Hillebrand IIH; Ngai JT;
15          Seabloom EW; Shurin JB;  Smith JE. (2007). Global analysis of nitrogen  and phosphorus
16          limitation of primary producers in freshwater, marine, and terrestrial ecosystems. Ecol
17          Lett, 10, 1135-1142

18    Farmer AM; Bates JW; Bell JNB.  (1992). Ecophysiological effects of acid rain on bryophytes
19          and lichens. In: Bates JW; Farmer AM (Eds.), Bryophytes and lichens in  a changing
20          environment. Oxford, UK:  Claredon Press.

21    Fenn ME; Poth MA; Johnson DW. (1996). Evidence for nitrogen saturation in the San
22          Bernardino Mountains in southern California. For Ecol Manage, 82, 211-230.

23    Ford J; Stoddard JL; Powers CF. (1993). Perspectives in environmental monitoring: an
24          introduction to the U.S. EPA long-term monitonring (LTM) project. Water Air Soil
25          Pollut, 67, 247-255.
      March 2010                               55                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    Galloway JN; Schlesinger WH; Levy H; Michaels AF; Schnoor JL. (1995). Nitrogen-fixation:
 2          anthropogenic enhancement, environmental response. Glob Biogeochem Cycles, 9, 235-
 3          252.

 4    Gilliam FS; Adams MB; Yurish BM. (1996). Ecosystem nutrient responses to chronic nutrient
 5          inputs at Fernow Experimental Forest, West Virginia. Can J For Res, 26, 196-205.

 6    Haines TA; Baker JP. (1986). Evidence offish population responses to acidification in the
 7          eastern United States. Water Air Soil Pollut, 31, 605-629.

 8    Holland EA; Dentener FJ; Braswell BH; Sulzman JM. (1999). Contemporary and pre-industrial
 9          global reactive nitrogen budgets. Biogeochemistry, 46, 7-43.

10    Horsley SB; Long RP; Bailey SW; Hallett RA; Hall TJ. (2000). Factors associated with the
11          decline disease of sugar maple on the Allegheny Plateau. Can J For Res, 30, 1365-1378.

12    Howarth RW; Billen G; Swaney D; Townsend A; Jaworski N; Lajtha K; Downing JA; Elmgren
13          R; Caraco N; Jordan T; Berendse F; Freney J; Kudeyarov V; Murdoch PS; Zhao-Liang Z.
14          (1996). Regional nitrogen  budgets and riverine N & P fluxes for the drainages to the
15          North Atlantic Ocean: natural and human influences. Biogeochemistry, 35, 75-139.

16    Hutchinson J; Maynard D; Geiser L. (1996). Air quality  and lichens - a literature review
17          emphasizing the Pacific Northwest, USA. Pacific Northwest Region Air Resource
18          Management Program; U.S. Forest Service; U.S. Department of Agriculture (USDA).

19    Kahl J; Norton S; Fernandez I; Rustad L; Handley M. (1999). Nitrogen and sulfur input-output
20          budgets in the experimental and reference watersheds, Bear Brook Watershed in Maine
21          (BBWM). Environ Monit Assess, 55, 113-131

22    Keller W; Gunn JM. (1995). Lake water quality improvements and recovering aquatic
23          communities. In: Gunn JM (Ed.), Restoration and recovery of an industrial region:
24          progress in restoring the smelter-damaged landscape near Sudbury, Canada (pp. 67-80).
25          New York, NY: Springer-Verlag.
     March 2010                              56                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    Kochy M; Wilson SD. (2001). Nitrogen deposition and forest expansion in the northern Great
 2          Plains. JEcol, 89, 807-817.

 3    Kretser W; Gallagher J; Nicolette J. (1989). Adirondack Lakes Study, 1984-1987: An Evaluation
 4          of Fish Communities and Water Chemistry: Ray Brook, NY; prepared for: Adirondack
 5          Lakes Survey (ALS) Corporation.

 6    LeBauer DS; Treseder KK. (2008). Nitrogen limitation of net primary productivity in terrestrial
 7          ecosystems is globally distributed. Ecology, 89,371-379.

 8    Lien L; Raddum GG; Fjellheim A.  (1992). Critical loads for surface waters: invertebrates and
 9          fish. (Acid rain research report no 21). Oslo, Norway: Norwegian Institute for Water
10          Research.

11    Longcore JR; Gill JD (Eds.), (1993). Acidic depositions: effects on wildlife and habitats.
12          (Wildlife Society technical review no 93-1). Bethesda, MD: The Wildlife Society.

13    MacAvoy SW; Bulger AJ. (1995). Survival of brook trout (Salvelinus fontinalis) embryos and
14          fry in streams of different acid sensitivity in Shenandoah National Park, USA. Water Air
15          Soil Pollut, 85, 445-450.

16    McNulty  SG; Cohen EC; Myers JAM; Sullivan TJ; Li H. (2007). Estimates of critical acid loads
17          and exceedances for forest soils across the conterminous United States. Environ Pollut,
18          149,281-292.

19    Murdoch  PS; Stoddard JL. (1992). The role of nitrate in the acidification of streams in the
20          Catskill Mountains of New York. Water Resour Res, 28, 2707-2720

21    Nash TH; Sigal LL. (1999). Epiphytic lichens in the San Bernardino mountains in relation to
22          oxidant gradients. In: Miller PR, McBride JR (Eds.), Oxidant air pollution impacts on the
23          montane forests  of southern California: A case study of the San Bernardino mountains.
24          Ecological Studies,  134, (pp. 223-234). New York, NY: Springer-Verlag.
      March  2010                               57                 Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1   Norton S; Kahl J; Fernandez I. (1999). Altered soil-soil water interactions inferred from stream
 2          water chemistry at an artificially acidified watershed at Bear Brook Watershed, Maine
 3          USA. Environ Monit Assess, 55, 97-111.

 4   Officer CB; Biggs RB; Taft JL; Cronin LE; Tyler MA; Boynton WR. (1984). Chesapeake Bay
 5          anoxa. Origin, development and significance. Science, 223, 22-27'.

 6   Paerl HW; Boynton WR; Dennis RL; Driscoll CT; Greening HS;  Kremer JN; Rabalais NN;
 7          Seitzinger SP. (2001). Atmospheric deposition of nitrogen in coastal waters:
 8          Biogeochemical and ecological implications. In:  Valigura RW; Alexander RB; Castro
 9          MS; Meyers TP; Paerl HW;  Stacey PE; Turner RE (Eds.), Nitrogen loading in coastal
10          water bodies: an  atmospheric perspective. (Coastal and estuarine series; Volume 57, pp
11          11-52). Washington, DC: American Geophysical Union.

12   Paerl H; Pinckney J; Fear J; Peierls  B. (1998). Ecosystem responses to internal and watershed
13          organic matter loading: Consequences for hypoxia in the eutrophying Neusse River
14          Estuary, NC, USA. Mar Ecol Prog Ser, 166, 17-25.

15   Parker KE. (1988). Common loon reproduction and chick feeding on acidified lakes in the
16          Adirondack Park, New York. Canadian Journal of Zoology, 66, 804-810.
     Parker DR; Zelazny LW; Kinraide TB. (1989). Chemical speciation and plant toxicity of
            aqueous aluminum. In: Lewis TE (Ed.), Environmental chemistry and toxicology of
19          aluminum (pp. 117-145). American Chemical Society.
17
18
20   Peterjohn WT; Adams MB; Gilliam FS. (1996). Symptoms of nitrogen saturation in two central
21          Appalachian hardwood forest ecosystems.Biogeochemistry, 35, 507-522.

22   Reuss JO. (1983). Implications of the calcium-aluminum exchange system for the effect of acid
23          precipitation on soils. J Environ Qual, 12, 591-595.

24   Riddell J; Nash lii TH; Padgett P. (2008). The effect of HNO3 gas on the lichen Ramalina
25          menziesii. Flora - Morphology, Distribution, Functional Ecology of Plants, 203, 47-54.
     March 2010                               58                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    Ross DS; Lawrence GB; Fredriksen G. (2004). Mineralization and nitrification patterns at eight
 2          northeastern USA forested research sites. For Ecol Manage, 188, 317-335.

 3    Saros JE; Interlandi SJ; Wolfe AP; Engstrom DR. (2003). Recent changes in the diatom
 4          community structure of lakes in the Beartooth Mountain Range, USA. Arct Antarct Alp
 5          Res, 35, 18-23.

 6    Schindler DW; Mills KH; Malley DF; Findlay MS; Schearer JA; Davies U; Turner MA; Lindsey
 7          GA; Cruikshank DR. (1985). Long-term ecosystem stress: Effects of years of
 8          experimental acidification. Science, 228, 1395-1401.

 9    Schreck CB.  (1981). Stress and rearing of salmonids. Aquaculture, 28, 241-249.

10    Schreck CB.  (1982). Stress and compensation in teleostean fishes: response to social and
11          physical factors. In: Pickering AD (Ed.),  Stress and fish (pp. 295-321). London:
12          Academic Press.

13    Schwinning S; Starr BI; Wojcik NJ; Miller ME; Ehleringer JE; Sanford RL Jr. (2005). Effects of
14          nitrogen deposition on an arid grassland in theColorado Plateau cold desert. Journal of
15          Rangeland Ecology and Management, 58, 565-574.

16    Sterner RW;  Elser JJ. (2002). Ecological stoichiometry: the biology of elements from molecules
17          to the biosphere. Princeton, NJ: Princeton University Press.

18    Stoddard JL. (1990). Plan for converting the NAPAP aquatic effects long-term monitoring
19          (LTM) project to the temporally integrated monitoring of ecosystems (TIME) project.
20          (International Report). Corvallis, OR; U.S. Environmental Protection Agency.

21    Stoddard JL. (1994). Long-term changes in watershed retention of nitrogen: its causes and
22          aquatic consequences. In Baker LA (Ed.), Environmental chemistry of lakes  and
23          reservoirs, (pp. 223-284). Washington, D.C.: American Chemical Society.

24    Stoddard JL; Urquhart NS; Newell AD; Kugler D. (1996). The Temporally Interated Monitoring
25          of Ecosystems (TIME) project design 2. Detection of regional acidification trends. Water
26          Resour Res, 32, 2529-2538.

      March  2010                               59                 Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    Suding KN; Collins SL; Gough L; Clark C; Cleland EE; Gross KL; Milchunas DG; Pennings S.
 2          (2005). Functional- and abundance-based mechanisms explain diversity loss due to N
 3          fertlization. Proc Natl Acad Sci USA, 102, 4387-4392.

 4    Sullivan TJ; Driscoll CT; Cosby BJ; Fernandez IJ; Herlihy AT; Zhai J; Stemberger R; Snyder
 5          KU; Sutherland JW; Nierzwicki-Bauer SA; Boylen CW; McDonnell TC; Nowicki NA.
 6          (2006). Assessment of the extent to which intensively studied lakes are representative of
 7          the Adirondack Mountain region. (Final Report no 06-17).Corvallis, OR; prepared by
 8          Environmental Chemistry, Inc. for: Albany, NY; Environmental Monitoring Evaluation
 9          and Protection Program of the New York State Energy Research and Development
10          Authority (NYSERDA).

11    Sverdrup H; Warfvinge P. (1993). The effect of soil acidification on the growth of trees, grass
12          and herbs as expressed by the (Ca+ Mg+ K)/A1  ratio. Rep in Ecol & Eng, 2, 1993.

13    US EPA (2008) U.S. EPA. Integrated  Science Assessment (ISA) for Oxides of Nitrogen and
14          Sulfur Ecological Criteria (Final Report). U.S. Environmental Protection Agency,
15          Washington, D.C., EPA/600/R-08/082F, 2008.

16    US EPA (2009) Risk and Exposure Assessment for Review of the Secondary National Ambient
17          Air Quality Standards for Oxides of Nitrogen and Oxides of Sulfur-Main Content  - Final
18          Report. U.S. Environmental Protection Agency, Washington, D.C., EPA-452/R-09-008a

19    Vitousek PM; Howarth RW. (1991). Nitrogen limitation on land and in the sea: how can it
20          occur? Biogeochemistry, 13, 87-115.

21    Williams MW; Baron  JS; Caine N; Sommerfeld R; Sanford JR. (1996). Nitrogen saturation in
22          the Rocky Mountains. Environ Sci Technol,  30, 640-646.

23    Wigington PJ Jr; DeWalle DR; Murdoch PS; Kretser WA; Simonin HA; Van Sickle J; Baker JP.
24          (1996b). Episodic acidification of small streams in the northeastern United States: Ionic
25          controls of episodes. Ecol Appl, 6, 389-407.
     March 2010                              60                Draft - Do Not Quote or Cite

-------
          Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

1   Wedemeyer GA; Barton BA; MeLeay DJ. (1990). Stress and acclimation. In: Schreck CB,
2          Moyle PB (Eds.), Methods for fish biology (pp. 178-196). Bethesda, MD: American
3          Fisheries Society.
    March  2010                                61                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 i          3.     CONSIDERATIONS OF ADVERSITY TO PUBLIC
 2                                      WELFARE

 3   3.1    HOW DO WE CHARACTERIZE ADVERSITY TO PUBLIC
 4          WELFARE? WHAT ARE THE RELEVANT FACTORS AND HOW
 5          ARE THEY ADDRESSED IN THIS DOCUMENT?
 6          The paradigm of looking at adversity to public welfare as deriving from disruptions in
 7   ecosystem structure and function has been used broadly by EPA to categorize effects from the
 8   cellular to the ecosystem level. An evaluation of adversity to public welfare might consider the
 9   type, intensity, and scale of the effect as well as the potential for recovery.
10          Similar concepts were used in past reviews of secondary NAAQS for ozone, PM relating
11   to visibility as well as initial reviews of effects from lead deposition. Because NOX and SOX are
12   deposited from ambient sources into ecosystems where they affect changes to organisms,
13   populations and ecosystems, the concept of adversity to public welfare as related to impacts on
14   the public from alterations in structure and function of ecosystems is appropriate for this review.
15   Other information that may be helpful to consider includes the role of critical loads and
16   ecosystem service impacts as benchmarks or measures of impacts on ecosystems that may affect
17   public welfare. Ecosystem services can be related directly to concepts of public welfare to
18   inform discussions of societal adverse impacts. Subsequent sections will discuss each of these
19   concepts as they relate to adversity.

20          3.1.1  What are the benchmarks for adversity from other sources?

21          3.1.1.1  Ozone and PM NAAQS Reviews
22          The evaluation  of adversity from a public welfare perspective in the context of ozone  and
23   particulate matter (PM) are relevant to this current review. Both ozone and PM have documented
24   effects on ecological receptors. These criteria pollutants are being reviewed on a schedule as part
25   of the NAAQS process. The ozone secondary standard is currently under reconsideration from
26   the 2008 ruling with a proposal  due on January 6, 2010. A draft Policy Assessment for PM is
27   being developed for CASAC and public consultation.
     March 2010                               62               Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1          3.1.1.1.1   Ozone
 2          Welfare effects of ozone are primarily limited to vegetation. These effects begin at the
 3    level of the individual cell and accumulate up to the level of whole leaves and plants. If effects
 4    occur on enough individual plants within the population, communities and ecosystems may be
 5    impacted. Prior to the 2008 ozone review, Ozone vegetation effects were classified as either
 6    "injury" or "damage" (FR 72 37889). "Injury" was defined as; encompassing all plant reactions,
 7    including reversible changes or changes in plant metabolism, quality  or reduced growth that does
 8    not impair the intended use of the plant while "damage" includes those injury effects that reach
 9    sufficient magnitude as to reduce  or impair the intended use of the plant (FR 72 37890). The
10    "intended use" of the plant was imbedded with the concept of adversity to public welfare.
11    Ozone-associated "damage" was considered adverse if the intended use of the plant was
12    compromised (i.e. crops, ornamentals, plants located in Class I areas). Effects of ozone on single
13    plants or species grown in monocultures such as agricultural crops and managed forests were
14    evaluated without consideration of potential effects on natural forests or entire ecosystems.
15          In the 2008 rulemaking, EPA expanded the characterization of adversity to go beyond the
16    individual plant level and this language is continued in the 2010 ozone reconsideration. The 2008
17    final rule and 2010 proposal conclude that a determination of what constitutes an "adverse"
18    welfare effect in the context of secondary NAAQS review can appropriately occur by
19    considering effects at higher ecological levels (populations, communities, ecosystems) as
20    supported by recent literature. The ozone review uses the example of the construct presented  in
21    Hogsett et al. (1997) as a model for assessing risks to forests. This study suggests that adverse
22    effects could be classified into one or more of the following categories: (1) economic production,
23    (2) ecological structure, (3) genetic resources, and (4) cultural values". Another recent
24    publication, "A Framework for Assessing and Reporting on Ecological Condition: an SAB
25    report" (Young and Sanzone, 2002) provides additional support for expanding the consideration
26    of adversity beyond the species level and at higher levels by making explicit the linkages
27    between stress-related effects at the species level and at higher levels within an ecosystem
28    hierarchy (See Figure 3.1.1).
      March 2010                                63                Draft - Do Not Quote or Cite

-------
       Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
•           Hydrologic alteration
           Habitat conversion
           Habitat fragmentation
       t/K  Climate change
       Ij  Invasive non-native species
       |§  Turbidity/sedimentation
       uj  Pesticides
       5  Disease/pest otabHktitt
           Nutrient pukes
           Metals
           Dissolved oxygen depletion
           Ozone ftropospberic)
                                    1
                                        Hydroiogic aiteratit
                                        Habita f conversion
                                              Habit* t fragment^ tion
                                              Climate chanty
                                              Over-harvestmg of vegetation
                                              Large-scale invasive
                                               species introductions
                                              Large-scale disease/pes t outbreaks
        L
Landscape
Condition
                                       Biotic
                                     Condition
           Hydrologic alteration
           Habitat conversion
           Climate change
           Turbidity/seaimentation
           Pesticides
           Nutrient pukes
           Metals
           Dissolved oxygen depletion
           Ozone (tropospherif)
           Nitroget) oxides
           Natural
         Disturbance
                                                                               Hydrologic alteration
                                                                                 Habitat conversion  (3,
                                                                                   Climate change  Q
                                                                        Over-harvesting of vegetation  tu
                                                                              Disease/pest outbreaks  g
                                                                                 Altered fire regime  <-n
                                                                                Altered flood regime
                                                                         -	,	1
                                                       Hydrology/
                                                      Geomorphology.
                                                    Ecological
                                                    Processes
                           Hydrol&gic alteration
                            Habitat conversion  ^
                          Habitat fragmentation  P
                               CMmate change  §
                         Turbidity/sedimentation
                                            I  Hydroiogic alteration
                                          ff.  Habitat conversion
                                          g  Climate chanty
                                          Q  Pesticides
                                              Disease/pest outbreaks
                                          5  Nutrient pulses
                                              Dissohea oxygen depletion
 -.                                          |  Nitrogen oxides

 2            Figure 3-1. Common anthropogenic stressors and the essential ecological
 3            attributes they affect. Modified from Young and Sanzone (2002)

 4            In the 2008 ozone NAAQS review and current ozone NAAQS proposal, the

 5    interpretation of what constitutes an adverse effect on public welfare can vary depending on the

 6    location and intended use of the plant. The degree to which Os-related effects are considered

 7    adverse to public welfare depends on the intended use of the vegetation and its significance to

 8    public welfare (73 FR 16496). Therefore, effects on vegetation (e.g., biomass loss, foliar injury,

 9    impairment of intended use) may be judged to have a different degree of impact on public

10    welfare depending, for example, on whether that effect occurs in a Class I area, a city park,

11    commercial cropland or private  land.

12            In the proposed ozone reconsideration in 2010 the Administrator has found that the types

13    of information most useful in informing the selection of an appropriate range of protective levels
March 2010
                                                       64
                        Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    is appropriately focused on information regarding exposures and responses of sensitive trees and
 2    other native species known or anticipated to occur in protected areas such as Class I areas or on
 3    lands set aside by States, Tribes and public interest groups to provide similar benefits to the
 4    public welfare, for residents on those lands, as well as visitors to those areas. She further notes
 5    that while direct links between Os induced visible foliar injury symptoms and other adverse
 6    effects (e.g., biomass loss) are not always found, visible foliar injury in itself is considered  by the
 7    National Park Service (NFS) to affect adversely air quality related values (AQRV) in Class I
 8    areas, while the Administrator recognizes that uncertainty remains as to what level of annual tree
 9    seedling biomass loss when compounded over multiple years should be judged adverse to the
10    public welfare, she believes that the potential for such anticipated effects should be considered in
11    judging  to what degree a standard should be precautionary  (73 FR 16496). The range of
12    proposed levels from 7-15 ppb includes at the maximum level of 15 ppb protection of
13    approximately 75% of seedlings from more than 10% biomass loss.
14          3.1.1.1.2   PM
15          [To be added in the second draft policy assessment based on the draft PM policy
16    assessment]

17          3.1.2   Other EPA Programs and Federal Agencies
18          Various federal laws and policies exist to protect ecosystem health. How other federal
19    agencies and EPA offices consider ecosystem effects in carrying out their programs can help
20    inform the Administrator when she evaluates the adversity  of ecosystem impacts on public
21    welfare.  For example, an effect may be considered adverse to public welfare if it contributes to
22    the inability of areas to meet water quality objectives as defined by the Clean Water Act. The
23    following federal statutes and policies may prove helpful to consider.

24          EPA Office of Water
25          Section 101 of the Clean Water Act (CWA) (Declaration of Goals and Policy) states that
26    the objective of the CWA is to restore and maintain the chemical, physical, and biological
27    integrity of the Nation's waters and to attain, where possible, water quality that protects fish,
28    shellfish, wildlife and provides for water-based recreation.
      March 2010                                65                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1          The CWA also authorizes EPA to develop water quality criteria as a guide for the states
 2    to set water quality standards to protect aquatic life. In consideration of acidification effects,
 3    EPA's Redbook, Quality Criteria for Water, published originally in 1976, recommends that
 4    alkalinity be 20 mg/1 or more as CaCO3 for freshwater aquatic life except where natural
 5    concentrations are less. Alkalinity is the sum total of components in the water that tend to elevate
 6    the pH of the water above a value of about 4.5.
 7          As mentioned in the Redbook, alkalinity is expressed as CaCO3 in mg/1. Alkalinity
 8    differs slightly from ANC in that ANC includes other buffering compounds (Na, Mg, and K) as
 9    well and includes buffering capacity of particulates in the water sample. Since alkalinity is
10    expressed as mg/1 and ANC is expressed as ueq/1, alkalinity must be multiplied by 20 to be
11    converted to ueq/1. Thus a recommended criterion of 20 mg/1 alkalinity is roughly equivalent to
12    an ANC of 400 ueq/1.
13          The Clean Air Act's Prevention of Significant Deterioration (PSD) program (42
14    U.S.C. 7470) purposes include to "preserve, protect and enhance the air quality in national parks,
15    wilderness areas and other areas of natural, recreational, scenic or historic value .  . . ." Also, the
16    PSD program charges the Federal Land Managers, including the NFS, with ". .  . an affirmative
17    responsibility to protect the air quality related values . . . "within federal Class I lands. (42 U.S.C.
18    7475(d)(2)(B)).

19          National Park Service
20          The National Park Service (NFS) is responsible for the protection  of all resources within
21    the national park system. These resources include those that are related to and/or dependent upon
22    good air quality, such as whole ecosystems and ecosystem components. The NFS, in its Organic
23    Act (16 U.S.C. 1), is directed to conserve the scenery, natural and historic objects and wildlife
24    and to provide for the enjoyment of these resources unimpaired for current and future
25    generations.
26          The Wilderness Act of 1964 asserts wilderness areas will be administered in such a
27    manner as to leave them unimpaired and preserve them for the enjoyment of future generations.
28          NFS Management Policies (2006) guide all NFS actions including natural resources
29    management.  In general, the NFS Management Policies reiterate the NFS Organic Act's mandate
30    to manage the resources "unimpaired."
      March  2010                                66                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1           U.S. Fish and Wildlife Service
 2           On endangered species, Title 16 USC Chapter 35 Section  1531 states "The Congress
 3    finds and declares that— these species offish, wildlife, and plants are of esthetic, ecological,
 4    educational, historical, recreational, and scientific value to the Nation and its people and that all
 5    Federal departments and agencies will use their authorities to conserve threatened and
 6    endangered species.
 7           The United States Fish and Wildlife Service (FWS) manages the National Wildlife
 8    Refuge System lands to "...ensure that the biological integrity, diversity, and environmental
 9    health of the Systems are maintained for the benefit of present and future generations of
10    Americans." 16 U.S.C. Section 668dd(a)(4)(B)(1997).

11           U.S. Forest Service
12           The National Forest units are managed consistent with Land and Resource Management
13    Plans (LRMPs) under the provisions of the National Forest Management Act (NFMA).  16
14    §U.S.C. 1604 (1997). LRMPs are, in part, specifically based on recognition that the National
15    Forests are ecosystems and their management for goods and services requires an awareness and
16    consideration of the interrelationships among plants, animals, soil, water, air, and other
17    environmental factors within such ecosystems.  36 C.F.R. §219.1(b)(3)
18           Any measures addressing Air Quality Related Values (AQRV) on National Forest
19    System lands will be implemented through, and be  consistent with, the provisions of an
20    applicable LRMP or its revision (16 U.S.C. §1604(i)). Additionally, the  Secretary of Agriculture
21    must prepare a Renewable Resource Program that recognizes the  need to protect and, if
22    necessary, improve the quality of air resources. 16 U.S.C. §1602(5)(C).
23           AQRVs in Wilderness areas may receive  further protection by the previously mentioned
24    1964 Wilderness Act. For Wilderness  Areas in the National Forest System, the Act's
25    implementing regulations are found at 36 C.F.R.  §293 requiring these Wilderness Areas be
26    administered to preserve and protect [their] wilderness character.

27           Chesapeake Bay Total Maximum Daily Loads
28           Under section 303(d) of the Clean Water Act, states, territories, and authorized tribes are
29    required to develop lists of impaired waters. These  are waters that are too polluted or otherwise
30    degraded to meet the water quality standards set by states, territories, or  authorized tribes. The

      March 2010                                67                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    law requires that these jurisdictions establish priority rankings for waters on the lists and develop
 2    TMDLs for these waters. A Total Maximum Daily Load, or TMDL, is a calculation of the
 3    maximum amount of a pollutant that a waterbody can receive and still safely meet water quality
 4    standards. EPA is developing a TMDL for the Chesapeake Bay and its tributaries. The
 5    Chesapeake Bay Program has modeled the level of nitrogen that can reach the Bay and still meet
 6    the Bay's water quality standards. The TMDL, with full public participation, will set waste load
 7    allocations for point source discharges and load allocations for nonpoint sources of nitrogen. Air
 8    deposition to the Bay and its watershed, as a source category, will have a specific allocation. The
 9    allocation can be used to calculate the level of ambient air concentrations of reactive nitrogen
10    that are likely to meet the deposition allocation. To find the NOX portion of the allocation one
11    would subtract the reduced forms from the total allocation. If the total load to the Bay of nitrogen
12    from all the allocated source categories remains below the allocations, then the Bay is  expected
13    to meet the water quality standards, which are set to protect the designated uses of the Bay. Since
14    the designated uses are set by the states with public input, not meeting the designated uses can be
15    seen as having an adverse effect.

16           United Nations Economic Commission for Europe (UNECE)
17           [This information will be included in the second draft.]

18           Critical Loads
19           The term critical load is used to describe the threshold of air pollution deposition that
20    causes a specified level of harm to sensitive resources in an ecosystem. A critical load is
21    technically defined as "the quantitative estimate of an exposure to one or more pollutants below
22    which significant harmful effects on specified sensitive elements of the environment are not
23    expected to occur according to present knowledge" (Nilsson  and Grennfelt, 1988). The
24    determination of when a harmful effect becomes  "significant" may be in the view of a researcher
25    or through a policy development process. Researchers  often use the term "critical loads" to
26    describe when particular detrimental effects are realized, as is the case in Figure 2-1. In many
27    European countries a critical loads framework is used to determine a level of damages to
28    ecosystem services from pollution that are legally allowed. These critical loads are determined
29    through a policy process.
      March 2010                                68                 Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
            Harmful effects due to acidification have been defined here as those that occur below a
      given ANC for aquatic systems and below a given Be: Al ratio for terrestrial systems. However,
 3    the level at which an effect becomes harmful in that it causes adverse effects on public welfare is
      H(^t(^rmm(^H \\\j tVi£> A Hmim ctrcitr\r
1
2   given
4    determined by the Administrator.
 5    3.2   WHAT ARE ECOSYSTEM SERVICES AND HOW DOES THIS
 6          CONCEPT RELATE TO PUBLIC WELFARE?
 7          An additional concept that may be useful in considering the issue of adversity to public
 8    welfare is ecosystem services. In the next section the concept of ecosystem services, its
 9    relationship to adversity and public welfare within the context of this review are explained.
10          Characterizing a known or anticipated adverse effect to public welfare is an important
11    component of developing any secondary NAAQS. According to the Clean Air Act, welfare
12    effects include:
13          effects on soils, water, crops, vegetation, manmade materials, animals, wildlife,
14          weather, visibility, and climate, damage to and deterioration of property, and
15          hazards to transportation, as well as effect on economic values and on personal
16          comfort and well-being, whether caused by transformation, conversion, or
17          combination  with other air pollutants (CAA, Section 302(h)).
18          While the text above lists a number of welfare effects, these effects are not an effect on
19    public welfare in and of themselves.
20          Ecosystem services can be generally defined as the benefits individuals  and organizations
21    obtain from ecosystems. Ecosystem services can be classified as provisioning (food and water),
22    regulating (control of climate and disease), cultural (recreational), and  supporting (nutrient
23    cycling) (MEA 2005). Conceptually, changes in ecosystem services may be used to aid in
24    characterizing a known or anticipated adverse effect to public welfare.  In the  context of this
25    review, ecosystem services may also aid in assessing the magnitude and significance to the
26    public of a resource and in assessing how NOX and SOX concentrations and deposition may
27    impact that resource. The relationship between ecosystem services and public welfare effects is
28    illustrated in Table 3.2.1.
      March  2010                                69                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
     Table 3-1. Crosswalk between Ecosystem Services and Public Welfare Effects
Public Welfare Effect
Soils
Water
Crops
Vegetation
Wildlife
Climate
* Personal Comfort and
Wellbeing
Ecosystem Service
Nutrient Cycling
Drinking water, Recreation,
Aesthetic
Food, Fuel Production
Food, Recreation, Aesthetic,
Nonuse
Recreation, Food, Nonuse
Climate Control

Service Category
Supporting
Provisioning, Cultural
Provisioning
Provisioning, Cultural
Cultural, Provisioning
Regulating

 1   *A11 ecosystem services contribute to personal comfort and wellbeing.
 2          EPA has defined ecological goods and services for the purposes of a Regulatory Impact
 3   Analysis as the "outputs of ecological functions or processes that directly or indirectly contribute
 4   to social welfare or have the potential to do so in the future. Some outputs may be bought and
 5   sold, but most are not marketed" (US EPA 2006). Though this is not a definition specifically for
 6   use in the NAAQS process it may be a useful one in considering the scope of ecosystem services
 7   and the effects of air pollutants upon those services. Especially important is the
 8   acknowledgement that most of the goods and services supplied by ecosystems cannot be fully
 9   measured or monetized. Valuing ecological benefits, or the contributions to social welfare
10   derived from ecosystems, can be challenging as noted in EPA's Ecological Benefits Assessment
11   Strategic Plan (US EPA 2006) and the Science Advisory Board report "Valuing the Protection of
12   Ecological Systems and Services" (US EPA, 2009). It can be informative in characterizing
13   adversity to public welfare to  attempt to place an economic valuation on the set of goods and
14   services that have been identified with respect to a change in policy however it must be noted
15   that this valuation will be incomplete and illustrative only. The stepwise concept leading to the
16   valuation of ecosystem services is graphically depicted in Figure 3-2.
     March 2010
70
Draft - Do Not Quote or Cite

-------
      Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
                              EPA action
                                ^^•^H
                               Ecosyste
                         Ecological goods and services
                             affected by the policy
                       Planning and problem formulation
                       '
                           Goods and services
                                identified
                             Ecological analysis
                          Goods and services
                              quantified

                             Economic analysis
                           Goods and
                            services
                           monetized
                                                                 Goods and
                                                                 services not
                                                                  identified
                                                              Identified
                                                              goods and
                                                             services not
                                                              quantified
                                                           Quantified
                                                           goods and
                                                          services not
                                                           monetized
 2           Figure 3-2. RepresemauoTI 01 me oenenis assessment process indicating where
 3           some ecological benefits may remain unrecognized, unquantified, or
 4           unmonetized. (Source: EBASP USEPA 2006).
 5           A conceptual model integrating the role of ecosystem services in characterizing known or
 6    anticipated adverse effects to public welfare is shown in Figure 3-3. Under Section 109 of the
 7    CAA, the secondary standard is to specify a level  of air quality that is requisite to protect public
 8    welfare. For this review, the relevant air quality indicator is interpreted as ambient NOX and SOX
 9    concentrations that can be linked to levels of deposition for which there are ecological effects
10    that are adverse to public welfare. The case study  analyses (described in Chapters 4 and 5 of the
11    REA and summarized in Chapter 2 of this document) link deposition in sensitive ecosystems
12    (e.g., the exposure pathway) to changes in a given ecological indicator (e.g., for aquatic
13    acidification, changes in acid neutralizing capacity [ANC]) and then to changes in ecosystems
14    and the services they provide (e.g., fish species richness and its influence on recreational
15    fishing). To the extent possible for each targeted effect area, ambient concentrations of nitrogen
16    and sulfur (i.e., ambient air quality indicators) were linked to deposition in sensitive ecosystems
17    (i.e., exposure  pathways), and then deposition was linked to system response as measured by a
18    given ecological indicator (e.g., lake and stream acidification as measured by ANC). The
March 2010
                                                  71
Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    ecological effect (e.g., changes in fish species richness, etc.) was then, where possible, associated
 2    with changes in ecosystem services and their public welfare effects (e.g., recreational fishing).
 3           Knowledge about the relationships linking ambient concentrations and ecosystem
 4    services can be used to inform a policy judgment on a known or anticipated adverse public
 5    welfare effect. The conceptual model outlined for aquatic acidification in Figure 3-3 can be
 6    modified for any targeted effect area where sufficient data and models are available. For
 7    example, a change in an ecosystem structure and process, such as foliar injury, would be
 8    classified as an ecological effect, with the associated changes in ecosystem services, such as
 9    primary productivity, food availability, and aesthetics (e.g., scenic viewing), classified as public
10    welfare effects. Additionally,  changes in biodiversity would be classified as an ecological effect,
11    and the associated changes in ecosystem services—productivity, recreational viewing and
12    aesthetics—would be classified as public welfare effects. This information can then be used by
13    the Administrator to determine whether or not the changes described are adverse to public
14    welfare. In subsequent sections these concepts are applied to characterize the ecosystem services
15    potentially affected by nitrogen and/or sulfur for each of the effect areas assessed in the REA.
      March 2010                                 72                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
                           Ambient Air Quality
                                 Indicator
                            Exposure Pathway
                           Affected Ecosystem
                           Ecological Response
                           (ecological indicator  )
                                                        NOX/SOX
                                                     Concentrations
                                                   Atmospheric N & S
                                                       Deposition
                                                         Aquatic
                                                       ~
                                                      Acidification
                                                   (lake/stream ANC )
                             Ecological Effect
                                                  Change in Ecosystem
                                                  Structure & Processes
                                                  (fish species richness  )
 1
 2
 3
 4
 5
 6
 9

10

11

12
                            Ecological Benefit
                              Welfare Effect
                                                       Change in
                                                   Ecosystem Services
                                                   (recreational fishing )
       Figure 3-3. Conceptual model showing the relationships among ambient air
       quality indicators and exposure pathways and the resulting impacts on
       ecosystems, ecological responses, effects and benefits to characterize known or
       anticipated adverse effects to public welfare.  [This figure to be revised for Second
       Draft Policy Assessment Document]

       These  concepts can also be applied to the programs described in section 3.1. National

parks represent areas of nationally recognized ecological and public welfare significance, which

are afforded a higher level of protection. Therefore, staff has also focused on air quality and

deposition in the subset of national park sites and important natural areas. Figures 3-4 and 3-5

illustrate the spatial relationships between sensitive regions, Class 1 areas and nitrogen
deposition levels.
      March 2010
                                            73
Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
1
2
3
                               *   *;••,••'
                                  '•^-.j^
                             |    | Class 1 Areas
                               J Sensitive Aquatic Areas
                             Combined N and S (wet and dry)
                             Value
                               •I High : 1 50003e»008


                               • Low: 1 01593e+006
                                           ••
                                            ~\
                                                ^
Figure 3-4. Locations of Eastern U.S. National Parks (Class I areas) relative to
deposition of Nitrogen and Sulfur in sensitive aquatic areas
     March 2010
                                      74
Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
                         	| Class 1 Areas
                       Combined N and S (wet and dry)
                       Value
                         • High : 150.003 kg/ha/yr

                         | Low : 1.016 kg/ha/yr
 1
 2          Figure 3-5. Location of Western U.S. National Parks (Class I areas) relative to
 3          deposition of Nitrogen and Sulfur
 4          [Figures 3-4 and 3-5 will be revised for Second Draft policy Assessment Document]

 5   3.3   WHAT IS THE ROLE OF ECONOMICS?
 6          As discussed earlier in this document, a secondary NAAQS is required to be set at the
 7   "level(s) of air quality necessary to protect the public welfare from any known or anticipated
 8   adverse effects". As part of the effort to determine the standard, EPA linked the changes in the
 9   ambient air concentrations of NOX and SOX to the changes in ecosystem services and ultimately
10   to changes in public welfare (U.S. EPA, 2009). As previously mentioned most ecosystem
     March  2010
75
Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    services are not amenable to monetization a small subset of changes in services can be described
 2    by economic valuation methods. And although economics on its own cannot determine which
 3    impact on public welfare is "adverse", economics could be helpful in the context of a secondary
 4    NAAQS for determining the degree to which improvements are beneficial to public welfare and
 5    illustrating  and aggregating those impacts.7
 6           The Role of Economics in Defining "Adversity" There is neither an economic definition
 7    of how much loss in public welfare is adverse nor an economic definition of adversity. While an
 8    economist might consider a particular scenario adverse because it might imply some harm or
 9    potential for improvement, there is no specific threshold level when a loss in welfare (e.g. loss in
10    dollars) becomes adverse. An individual might be willing to give up some of their resources to
11    avoid a threat or negative outcome (i.e., willing to pay to avoid a particular outcome). According
12    to economic theory, if an individual is willing to give up something to avoid the outcome, then
13    imposing the outcome on the individual must make them worse off, at which point an economist
14    might colloquially describe the outcome as adverse. However, the amount they would have been
15    willing to pay to avoid the outcome might be quite small, and might not rise to a level  of harm
16    that the Administrator interprets as "adverse" to public welfare. In summary, economics provides
17    little guidance as to how the Administrator should interpret the word "adverse" in the context of
18    public welfare.
19           Ecosystem Services and Links to Public Welfare An ecosystem service framework
20    provides a structure to measure changes in public welfare from changes  in ecosystem functions
21    affected by air pollution. EPA's Risk Assessment for this rulemaking defines  ecosystem services
22    as "the ecological processes or functions having monetary or nonmonetary value to individuals
23    or society at large" (EPA 2009.) The discipline of economics provides a useful approach for
24    summarizing how the public values changes in the services provided by the environment. An
25    ecosystem services framework (with or without valuation) can provide measures of changes in
26    public welfare.
      7 Section 109 of the Clean Air Act forbids consideration of the compliance costs of reducing pollution when setting
      a NAAQS. However, there is no prohibition regarding the consideration of the monetized impacts of welfare
      effects occurring due to levels of pollution above alternative standards in evaluating the adversity of the impacts to
      public welfare. Ecosystem services can be characterized as a method of monetizing the impacts of the air pollution.
      Although a separate regulatory document quantifying the costs and benefits of attaining a NAAQS is prepared
      simultaneously, this document is not considered when selecting a standard.
      March 2010                                 76                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1           Economics as a Framework to Illustrate Changes in Public Welfare Economics can
 2    provide a framework to illustrate how public welfare8 changes in response to changes in
 3    environmental quality by quantitatively linking changes in ecosystem services to preferences.
 4    Economics assumes that the choices that individuals make reflect their preferences over certain
 5    outcomes and that, generally speaking, they will make choices that, in expectation, will make
 6    them as well off as possible given their resources. In economics revealed and stated preference
 7    methods are used to observe the choices individuals make to understand the outcomes
 8    individuals prefer. What individuals are willing to give up for an outcome is their willingness-to-
 9    pay (WTP) for that outcome. An example of an outcome is  an improvement in an ecosystem
10    service. Often, to provide comparability to other goods and services,  in economics these
11    tradeoffs are framed  relative to dollars for convenience.9
12           Economics could inform the Administrator by valuing and characterizing the  changes in
13    public welfare from changes in the quantity and quality of ecosystem services. Overall, this
14    assessment intends to characterize changes in ecosystem services from a scientific perspective
15    using effects on ecosystem structures and functions or ecosystem integrity. Economics then
16    estimates the  effect on public welfare of these changes in the quantity and quality of ecosystem
17    services. For example, a decrease in a particular bird species can be characterized by  its effect on
18    the ecosystem's structure and function, while from an economic perspective, the effects would
19    be based on the impact on public welfare or the value  the public places on that species. A simple
20    example is a comparison between a decrease in a bird species that is relatively unknown
21    compared to a decrease in a very prominent species (e.g. Bald Eagle). The public is likely to
22    have a higher WTP to avoid the latter, and thus the decrease would affect the public welfare
23    more.
24           There are important complications with using preferences to understand the effect of
25    pollution on public welfare. For example, while the field of economics generally assumes that
26    public preferences are the paramount consideration; these preferences may change when the
27    public receives new information. Therefore, if individuals do not understand how pollution will
       [A discussion of economic interpretation of "Public Welfare" will be included in the second draft]
      9 Often groups collectively make choices to engage in activities that improve the collective welfare of the group. For
      example, a community around an acidified lake might purchase lime and use it to reduce the acidity of the lake.  The
      collective decisions can also be used to understand how people value improvements to ecosystem services.
      [Additional discussion will be included in the second draft related to collective actions that reveal preferences for
      improvements in relevant ecosystem services and how these collective actions, and the absence of these actions, can
      be interpreted.]

      March 2010                                77                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    affect ecosystem services, or even how those ecosystem services affect their quality of life, then
 2    they will have a difficult time valuing changes in those services. Similarly, it may be very costly
 3    for individuals to learn and understand how changes in particular ecosystem services may affect
 4    them, in part because typically there are significant interdependences within an ecosystem.
 5    Because of this complexity, individuals may implicitly value a species, or habitat, or ecosystem
 6    function because it supports an ecosystem service that they do clearly value. Furthermore, the
 7    public also has limited understanding regarding irreversibilities, tipping points, and other more
 8    complex aspects of ecosystems, which limits the ability to adequately value these ecosystems.10
 9    In addition, where and when a change in an ecosystem takes places is crucial for characterizing
10    the associated change in an ecosystem service, and will also affect the value the public places on
11    that change.

12    3.4    WHAT IS THE EVIDENCE  FOR EFFECTS ON ECOSYSTEM
13           SERVICES? HOW DO WE LINK ECOLOGICAL INDICATORS TO
14           SERVICES?
15           The process used to link ecological indicators to ecosystem services is discussed
16    extensively in Appendix 8 of the REA. In brief, for each effect area assessed the ecological
17    indicators were linked to an ecological response that was subsequently linked, to the extent
18    possible, to associated services. For example in the case study for aquatic acidification the
19    chosen ecological indicator is ANC which can be linked to the ecosystem service of recreational
20    fishing as  illustrated in the conceptual model shown in Figure 3-6. Although recreational fishing
21    losses are  the only service effects that can be quantified or monetized at this time, there are, as
22    can be seen in the Figure, numerous other ecosystem services that  may be related to the
23    ecological effects of acidification.
      10 While the public may not fully appreciate the interdependencies within ecosystems, they can learn them, but again
      it may be costly to do so. It is possible for individuals to value outcomes that are irreversible or result in discrete
      changes (i.e., tipping points) in the quality and quantity of ecosystem services. Avoiding irreversible outcomes
      should be and are more valued by individuals than outcomes that are not irreversible (Arrow and Fischer,  1974).

      March 2010                                78                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
              Acidifying Inputs
              to Surface Wat&r
                                               Impacts on Ecosystem Endpoinrs
                                                                           Affected Ecosystem Services
              N+S Deposition


Surface Water
Acidification:
Low pH and
ANC



Declines in
Aquatic Biota

Declines in
•1 Aquatic Biota:
I Reduced
[ Species
J Abundance,
Diversity, and
Richness

• Declines in
Terrestnai
Nearshore
Biota

^^^^^H
^^>
*

Provisioning Services
•production f of commercial
and subsistence fishing
Cultural Services
•recreational fishing
•waterfowl hunting
•aesthetic enjoyment
•no nuse services

                                                                            Regulating services
                                                                            •biological control
 2           Figure 3-6. Conceptual model linking ecological indicator (ANC) to affected
 3           ecosystem services.
 4           The next four sections summarize the current levels of certain ecosystem services for
 5    each of the effect areas analyzed in the REA and present results of analyses that have attempted
 6    to quantify and monetize the harms to public welfare, as represented by ecosystem services, due
 7    to nitrogen and sulfur deposition.
 8           Evidence for Adversity Related to Aquatic Acidification
 9           Acidification primarily affects the ecosystem services that are derived from the fish and
10    other aquatic life found in these surface waters (REA, Section 5.2.1.3). Food is generally the
11    most important provisioning services provided by inland surface waters (MEA, 2005). In the
12    northeastern United States, the surface waters affected by acidification are not a major source of
13    commercially raised or caught fish; however, they are a source of food for some recreational and
14    subsistence fishers and for other consumers. Although data and models are available for
15    examining the effects on recreational fishing, relatively little data are available for measuring the
16    effects on subsistence and other consumers. For example, although there is evidence that certain
17    population subgroups in the Northeastern United States, such as the Hmong and Chippewa ethnic
18    groups, have particularly high rates of self-caught fish consumption (Hutchison and Kraft, 1994;
19    Peterson et al.,  1994), it is not known if and how their consumption patterns are affected by the
20    reductions in available fish populations caused by surface water acidification.
21           Inland surface waters support several cultural services, such as aesthetic and educational
22    services; however, the type of service that is likely to be most widely and significantly affected
      March 2010
79
Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    by aquatic acidification is recreational fishing11. Recreational fishing in lakes and streams is
 2    among the most popular outdoor recreational activities in the northeastern United States. Data
 3    from the 2006 National Survey of Fishing, Hunting, and Wildlife Associated Recreation
 4    (FHWAR) indicate that more than 9% of adults in this part of the country participate annually in
 5    freshwater fishing with 140 million freshwater fishing days. Based on studies conducted in the
 6    northeastern United States, Kaval and Loomis (2003) estimated average consumer surplus values
 7    per day of $35 for recreational fishing (in 2007 dollars). Therefore, the implied total annual value
 8    of freshwater fishing in the northeastern United States was $5 billion in 2006.
 9           In general, inland surface waters such as lakes, rivers, and streams provide a number of
10    regulating services, such as hydrological regime regulation and climate regulation. There is little
11    evidence that acidification of freshwaters in the northeastern United States has significantly
12    degraded these specific services; however, freshwater ecosystems also provide biological control
13    services by providing environments that sustain delicate aquatic food chains.
14           The toxic effects of acidification on fish and other aquatic life impair these services by
15    disrupting the trophic structure of surface waters (Driscoll et al., 2001). Although it is difficult to
16    quantify these services and how they are affected by acidification, it is worth noting that some of
17    these services may be captured through measures of provisioning and cultural services. For
18    example, these biological control services may serve as "intermediate" inputs that support the
19    production of "final" recreational fishing and other cultural services.
20           What is the value of the impaired recreational fishing services?
21           The previous section describes the ecosystem services that are most likely to be affected
22    by N and S deposition, and it summarizes evidence regarding the  current magnitude and values
23    of recreational fishing services; however, it does not measure the degree to which these services
24    are impaired by existing NOX/SOX levels.
25           To address this limitation, the REA (Appendix 8) provides insights into the magnitude of
26    ecosystem service impairments.
27           Specifically, the REA focuses on measuring the benefits of ecosystem service
28    enhancements resulting from the elimination of anthropogenic sources of NOX/SOX. Rather than
29    asking how much public welfare is currently adversely affected relative to a scenario without
      11 Banzhaf et al (2006) has shown that non-use services are arguably a more significant source of benefits from
      reduced acidification than recreational fishing.

      March 2010                                80                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    anthropogenic NOX/SOX, it asks a similar question of how much public welfare would improve if
 2    the emissions were eliminated. The REA provides quantitative estimates of selected ecosystem
 3    services impairments or enhancements for three main categories of ecosystem effects - aquatic
 4    acidification, terrestrial acidification, and aquatic nutrient enrichment12. Within these three
 5    categories, the selection of specific ecosystem services for more in-depth analysis depended
 6    primarily on the expected magnitude of impairments and on the availability of appropriate data
 7    and modeling tools.
 8           The analysis of ecosystem service impairments due to aquatic acidification builds on the
 9    case study analysis of lakes in the New York Adirondacks. It estimates changes in recreational
10    fishing services, as well as changes more broadly in "cultural" ecosystem services (including
11    recreational, aesthetic, and nonuse services). First, the MAGIC model was applied to 44 lakes to
12    predict what ANC levels would be under both "business as usual" conditions (i.e., allowing for
13    some decline in deposition due to existing regulations) and pre-emission (i.e., background)
14    conditions. When these model runs were initiated staff were interested in a prospective analysis
15    of conditions assuming a 2010 implementation of "zero-out" emissions with a projected lag time
16    to improvement of 10 years thus results were calculated for the year 2020. These predictions
17    were then extrapolated to the full universe of Adirondack lakes. Second, to estimate the
18    recreational fishing impacts of aquatic acidification in these lakes, an existing model of
19    recreational fishing demand and site choice was applied. This model predicts how recreational
20    fishing patterns in the Adirondacks would differ and how much higher the average annual value
21    of recreational fishing services would be for New York residents if lake ANC levels
22    corresponded to background (rather than business as usual) conditions. Aggregating these values
23    across all NY residents implies that acidification of Adirondack lakes due to anthropogenic
24    sources of NOX/SOX would impair annual recreational  fishing services of NY residents by $6
25    million to $11 million in 2020. Current annual impairments are most  likely of a similar
26    magnitude because, although current NOX/SOX levels are somewhat higher than those expected in
27    2020 (under business as usual - given expected emissions  controls associated with Title IV
28    regulations but no additional nitrogen or sulfur controls), the affected NY population is also
29    somewhat smaller (based on U.S. Census Bureau projections).
      12 Estimates for terrestrial nutrient enrichments were not generated due to the limited availability of necessary data
      and models for this effect category.

      March 2010                                81                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1           Third, to estimate impacts on a broader category of cultural ecosystem services, results
 2    from an existing valuation survey of NY residents were adapted and applied to this context. The
 3    survey used a contingent valuation approach to estimate the average annual household WTP for
 4    future reductions (20% and 45%) in the percent of Adirondack lakes impaired by acidification.
 5    These WTP estimates were then (1) rescaled to reflect predicted changes between business-as-
 6    usual and background conditions in 2020 (MAGIC lake modeling results indicate that the
 7    percentage of impaired lakes would be 22 to 31  points lower under background conditions), and
 8    (2) aggregated across NY households. The aggregate annual value to NY residents in 2010 for a
 9    reduction in lake acidification to background levels by 2020 was estimated to range $4 million to
10    $300 million in 2007 dollars. For comparison the previous section estimated the value of
11    recreational fishing in the Northeastern states at approximately $5 billion in 2006. These results
12    suggest that the value of avoiding current impairments to ecosystem services from Adirondack
13    lakes are even higher than the estimate, because they occur today rather than in 2020 (i.e.,  no
14    delayed effect) and because the percent of impaired lakes is slightly higher today than expected
15    in 2020 under business-as-usual. These results imply significant value to the public derived from
16    recreational fishing services.  The analysis especially illustrates what may be the scale of all
17    impacts to public welfare when viewed as a subset of all services impacted by acidification.
18           Evidence for Terrestrial Acidification
19           A similar model to Figure 3-6 can be drawn for terrestrial acidification that links Bc:Al
20    ratio to reduced tree growth to decreases in timber harvest although we have less confidence in
21    the significance of this linkage than we do for aquatic acidification. There are numerous services
22    expected to be affected, but the means to adequately describe those losses does not as yet exist.
23    These services include effects to forest health, water quality, and habitat, including decline in
24    habitat for threatened and endangered species, decline in forest aesthetics, decline in forest
25    productivity, increases in forest soil erosion and decreases in water retention (ISA, 2009; REA,
26    2009; Krieger, 2001).
27           Forests in the Northeastern United States provide several important and valuable
28    provisioning services, which are reflected in the production and sales of tree products.
29           Sugar maples are a particularly important commercial hardwood tree species in the
30    United States, producing wood products like timber and maple syrup that provide hundreds of
      March 2010                                82                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    millions of dollars in economic value annually (NASS, 2008). Red spruce is also used in a
 2    variety of wood products and provides up to $100 million in economic value annually.
 3           Forests in the Northeastern United States are also an important source of cultural
 4    ecosystem services, including nonuse (existence value for threatened and endangered species),
 5    recreational, and aesthetic services (ISA, 2009; REA, 2009). Red spruce forests are home to two
 6    federally listed species.
 7           Although we do not have the data to link acidification damages directly to economic
 8    values of lost recreational services in forests, these resources are valuable to the public. A recent
 9    study  suggests that the total annual value of off-road driving recreation was more than $9 billion,
10    total and value of hunting and wildlife viewing was more than $4 billion each in the Northeastern
11    United States in 2006(Kaval and Loomis, 2003). In addition, fall color viewing is a recreational
12    activity that is directly dependent on forest conditions. Sugar maple trees, in particular, are
13    known for their bright colors and are, therefore, an essential aesthetic component of most fall
14    color landscapes. Statistics on fall color viewing are much less available than for the other
15    recreational and tourism activities; however, a few studies have documented the extent and
16    significance of this activity. For example, Spencer and Holecek (2007) found that roughly 30%
17    of residents reported at least one trip in the previous year involving fall color viewing. In a
18    separate study conducted in Vermont, Brown (2002) reported that more than 22% of households
19    visiting Vermont in 2001 made the trip primarily for the purpose of viewing fall colors.
20           Two studies that have estimated values for protecting high-elevation spruce forests in the
21    Southern Appalachians. Kramer et al. (2003) conducted a contingent valuation study estimating
22    households' WTP for programs to protect remaining high-elevation spruce forests from damages
23    associated with air pollution and insect infestation (Haefele et al., 1991; Holmes and Kramer,
24    1995). Median household WTP was estimated to be roughly $29 (in 2007 dollars) for the
25    minimal program and $44 for the more extensive program. Another study by Jenkins, Sullivan,
26    and Amacher (2002) estimated an aggregate annual value of $3.4 billion for avoiding a
27    significant decline in the health of high-elevation spruce forests in the Southern Appalachian
28    region.
29           Forests in the Northeastern United States also support and provide a wide variety of
30    valuable regulating services, including soil stabilization and erosion control, water regulation,
31    and climate regulation (Krieger, 2001). Forest vegetation plays an important role in maintaining

      March 2010                                83                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    soils in order to reduce erosion, runoff, and sedimentation that can adversely impact surface
 2    waters. In addition to protecting the quality of water in this way, forests also help store and
 3    regulate the quantity and flows of water in watersheds. Finally, forests help regulate climate
 4    locally by trapping moisture and globally by sequestering carbon. The total value of these
 5    ecosystem services is very difficult to quantify and the magnitude of these impacts is currently
 6    very uncertain.
 7           What is the value of current ecosystem service impairments?
 8          The analysis of ecosystem service impairments associated with terrestrial acidification
 9    specifically addresses impacts on the forest product provisioning services from two
10    commercially important tree species - sugar maple and red spruce—that are particularly sensitive
11    to the  effects of acidification. Using data from the USFS Forest Inventory and Analysis (FIA)
12    database, an  exposure-response relationship was estimated for each species to measure the
13    average negative effect of critical load exceedances (CLEs) of nitrogen and sulfur deposition on
14    annual tree growth. These estimated relationships were then applied to sugar maple and red
15    spruce stocks in the Northeast and North central regions to estimate the average percent increase
16    in annual tree growth that would occur if all CLEs were eliminated. To estimate the  aggregate-
17    level forest market impacts of eliminating CLEs starting in the year 2000, the tree-level growth
18    adjustments were applied using the Forest and Agricultural Sector Optimization Model
19    (FASOM), which is a dynamic optimization model of the U.S. forest and agricultural sectors.
20    The public welfare gains linked to these markets from eliminating CLEs was estimated to be
21    $0.69  million per year. These estimates can also be interpreted as the current value of
22    impairments to forest provisioning services due to forest acidification effects from nitrogen and
23    sulfur deposition.
24          Nutrient Enrichment
25          For the purposes of the following sections nutrient enrichment refers only to  that due to
26    NOy deposition. Additionally these sections focus on the detrimental effects of that deposition.
27    Staff acknowledges that a certain amount of NOX deposition in managed terrestrial ecosystems
28    may have a beneficial effect. However no attempt has been made to quantify those beneficial
29    effects.
      March 2010                                84                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1          Evidence for Aquatic Nutrient Enrichment
 2          Estuaries in the eastern United States are an important source of food production, in
 3    particular fish and shellfish production.  The estuaries are capable of supporting large stocks of
 4    resident commercial species, and they serve as the  breeding grounds and interim habitat for
 5    several migratory species (U.S. EPA, 2009). To provide an indication of the magnitude of
 6    provisioning services associated with coastal fisheries, from 2005 to 2007, the average value of
 7    total catch was $1.5 billion per year in 15 East Coast states. It is not known, however, what
 8    percentage of this value is directly attributable to or dependent upon the estuaries in these states.
 9    Based on commercial landings in Maryland and Virginia, the values for three key species—blue
10    crab, striped bass, and menhaden- totaled nearly $69 million in 2007 in the Chesapeake Bay
11    alone.
12          Assessing how eutrophication in estuaries affects fishery resources requires bioeconomic
13    models (i.e., models that combine biological models offish population dynamics with economic
14    models describing fish harvesting and consumption decisions), but relatively few exist (Knowler,
15    2002). Kahn and Kemp (1985) estimated that a 50% reduction in SAV from levels would
16    decrease the net social benefits from striped bass by $16 million (in 2007 dollars). In a separate
17    analysis, Anderson (1989) modeled blue crab harvests under baseline conditions and under
18    conditions with "full restoration" of SAV. In equilibrium, the increase in annual producer surplus
19    and consumer surplus with full restoration of SAV was estimated to be $7.9 million (in 2007
20    dollars). Mistiaen, Strand, and Lipton (2003) found that reductions in DO cause a statistically
21    significant reduction in commercial harvest and revenues crab harvests. For the Patuxent River
22    alone, a simulated reduction of DO from 5.6 to 4.0 mg/L was estimated to reduce crab harvests
23    by 49%  and reduce total annual earnings in the fishery by $275,000 (in 2007 dollars).
24          In addition, eutrophi cation in estuaries may also affect the demand for seafood. For
25    example, a well-publicized toxic pfiesteria bloom in the Maryland Eastern Shore in 1997 led to
26    an estimated $56 million (in 2007 dollars) in lost seafood sales for 360 seafood firms in
27    Maryland in the months following the outbreak (Lipton, 1999). Surveys by Whitehead, Haab,
28    and Parsons (2003) and Parsons et al. (2006) indicated a reduction in consumer surplus due to
29    eutrophication-related fish kills ranging from $2 to $5 per seafood meal.13 As a result, they
      13 Surprisingly, these estimates were not sensitive to whether the fish kill was described as major or minor or to the
      different types of information included in the survey.

      March 2010                                85                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    estimated aggregate consumer surplus losses of $43 million to $84 million (in 2007 dollars) in
 2    the month after a fish kill.
 3           As mentioned in the REA (5.2.1.3), estuaries in the eastern United States also provide an
 4    important and substantial variety of cultural ecosystem services, including water-based
 5    recreational and aesthetic services. For example, FHWAR data indicate that 4.8% of the
 6    population in coastal states from North Carolina to Massachusetts participated in saltwater
 7    fishing, in 26 million saltwater fishing days in 2006 (U.S. DOT, 2007). Based on estimates in
 8    Section 5.2.1.3 of the REA, total recreational consumer surplus value from these saltwater
 9    fishing days was approximately $1.3 billion (in 2007 dollars). Recreational participation
10    estimates for several other coastal recreational activities are also available for 1999-2000 from
11    the NSRE. Almost 6 million individuals participated in motorboating in coastal states from North
12    Carolina to Massachusetts. Again, based on analysis in the REA,  the aggregate value of these
13    coastal motorboating outings was $2billion per year. Almost 7 million participated in
14    birdwatching, for a total of almost 175 million days per year, and more than 3 million
15    participated in visits to nonbeach coastal waterside areas, for a total of more than 35 million days
16    per year.
17    Estuaries and marshes have the potential to support a wide range  of regulating services,
18    including climate, biological, and water regulation; pollution detoxification; erosion prevention;
19    and protection against natural hazards (MEA, 2005c). The relative lack of empirical models and
20    valuation studies imposes obstacles to the estimation of ecosystem services affected by nitrogen
21    deposition.  While atmospheric deposition contributes to eutrophication there is uncertainty in
22    separating the effects of atmospheric nitrogen from nitrogen reaching the estuaries from many
23    other sources.
24           What is the value of current ecosystem service impairments?
25           The aquatic nutrient enrichment case study relied on the NOAA Eutrophi cation Index as
26    the indicator, which includes dissolved oxygen, HABs, loss of SAV and loss of water clarity.
27    There are methods available to link some of the components to ecosystem services, most notably
28    loss of SAV and reductions in DO. The REA analysis estimates the change in several ecosystem
29    services including recreational fishing, boating, beach use, aesthetic services and nonuse
30    services. The REA focuses on two major East Coast estuaries - the Chesapeake Bay and the
31    Neuse River. Both estuaries receive between 20%-30% percent of their annual nitrogen loadings

      March 2010                                86                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    through air deposition and both are showing symptoms of eutrophication. The analysis uses and
 2    adapts results from several existing studies to approximate effects on several ecosystem services,
 3    including commercial fishing, recreation, aesthetic enjoyment, and nonuse values. For example,
 4    it is estimated that atmospheric nitrogen reduces the annual benefits of recreational fishing,
 5    boating, and beach use in the Chesapeake Bay by $43-$217 million, $3-8 million, and $124
 6    million respectively, and reduces annual aesthetic benefits to nearshore residents by $39-102
 7    million. In the Neuse River, the value of annual commercial crab fishing services would be
 8    between $0.1-1 million higher without the contribution of atmospheric nitrogen, and recreation
 9    fishing services in the larger Albermarle Pamlico Sound estuary system (which includes the
10    Neuse) would be $ 1 -8 million greater per year.
11           Evidence for Terrestrial Nutrient Enrichment
12           The ecosystem service impacts of terrestrial nutrient enrichment include primarily
13    cultural and regulating services. In CSS areas, concerns focus on a decline in CSS and an
14    increase in nonnative grasses and other species, impacts on the viability of threatened and
15    endangered species associated with CSS, and an increase in fire frequency. Changes in MCF
16    include changes in habitat suitability and increased tree mortality, increased fire intensity, and a
17    change in the forest's nutrient cycling that may affect surface water quality through nitrate
18    leaching (EPA, 2008).
19           The value that California residents and the U.S. population as a whole place on CSS and
20    MCF habitats is reflected in the various federal, state, and local government measures that have
21    been put in place to protect these habitats. Threatened and endangered species are protected by
22    the Endangered Species Act.  The State of California passed the Natural Communities
23    Conservation Planning Program (NCCP) in 1991, and CSS was the first habitat identified for
24    protection under the program (see www.dfg.ca.gov/habcon/nccp). Private organizations such as
25    The Nature Conservancy, the Audubon Society, and local land trusts also protect and restore
26    CSS and MCF habitat.
27           CSS and MCF are found in numerous recreation areas in California. Three national parks
28    and monuments in California contain CSS, including Cabrillo National Monument, Channel
29    Islands National Park, and Santa Monica National Recreation Area. All three parks showcase
30    CSS habitat with educational programs and information provided to visitors, guided hikes, and
31    research projects focused on understanding and preserving CSS. Over a million visitors traveled

      March 2010                                87                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    through these three parks in 2008. MCF is highlighted in Sequoia and Kings Canyon National
 2    Park, Yosemite National Park, and Lassen Volcanic National Park, where more than 5 million
 3    people visited in 2008.
 4           The 2006 FHWAR for California (DOT, 2007) reports on the number of individuals
 5    involved in fishing, hunting, and wildlife viewing in California. Millions of people are involved
 6    in just these three activities each year. The quality of these trips depends in part on the health of
 7    the ecosystems and their ability to support the diversity of plants and animals found  in important
 8    habitats found in CSS or MCF ecosystems and the parks associated with those ecosystems.
 9    Based on analyses in Section 5.3.1.3 of the REA (U.S.EPA, 2009), average values of the total
10    benefits in 2006 from fishing, hunting, and wildlife viewing away from home in California were
11    approximately  $947 million, $169 million, and $3.59 billion, respectively. In addition, data from
12    California State Parks (2003) indicate that in 2002, 68.7% of adult residents participated in trail
13    hiking for an average of 24.1 days per year. The analyses  in the REA (U.S.EPA, 2009) indicate
14    that the aggregate annual benefit for California residents from trail hiking in 2007 was $11.59
15    billion.
16           CSS and MCF are home to a number of important and rare species and habitat types. CSS
17    displays richness in biodiversity with more than 550 herbaceous annual and perennial species. Of
18    these herbs, nearly half are endangered, sensitive, or of special status (Burger et al., 2003).
19    Additionally, avian, arthropod, herpetofauna, and mammalian species live in CSS habitat or use
20    the habitat for breeding or foraging. Communities of CSS are home to three important federally
21    endangered species. MCF is home to one federally endangered species and a number of state-
22    level sensitive species. The Audubon Society lists 28 important bird areas in CSS habitat and at
23    least 5 in MCF in California (http://ca.audubon.org/iba/index.shtml).14
24           The terrestrial enrichment case study in Section 5.3.1.3 of the REA and Section 3.3.5 of
25    the ISA identified fire regulation as a service that could be affected by nutrient enrichment of the
26    CSS and MCF  ecosystems by encouraging growth of more flammable grasses, increasing fuel
27    loads, and altering the fire cycle. Over the 5-year period from 2004 to 2008,  Southern California
28    experienced, on average, over 4,000 fires per year burning, on average, over 400,000 acres per
29    year (National Association of State Foresters [NASF], 2009). It is not possible at this time to
30    quantify the contribution of nitrogen depositio, among many other factors, to increased fire risk.
      14 Important Bird Areas are sites that provide essential habitat for one or more species of bird.

      March 2010                                88                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1          The CSS and MCF were selected as case studies for terrestrial enrichment because of the
 2    potential that these areas could be adversely affected by excessive N deposition. To date, the
 3    detailed studies needed to identify the magnitude of the adverse impacts due to N deposition
 4    have not been completed. Based on available data, this report provides a qualitative discussion of
 5    the services offered by CSS and MCF and a sense of the scale of benefits  associated with these
 6    services. California is famous for  its recreational opportunities and beautiful landscapes. CSS
 7    and MCF are an integral part of the California landscape, and together the ranges of these
 8    habitats include the densely populated and valuable coastline and the mountain areas.  Through
 9    recreation and scenic value, these  habitats affect the lives of millions of California residents and
10    tourists. Numerous threatened and endangered species at both the state and federal levels reside
11    in CSS and MCF. Both habitats may play an important role in wildfire frequency and  intensity,
12    an extremely important problem for California. The potentially high value of the ecosystem
13    services provided by CSS and MCF justify careful attention  to the long-term viability of these
14    habitats.
15          The terrestrial nutrient enrichment case study relies on benchmark deposition levels for
16    various species and ecosystems as indicators of ecosystem response. While it would be expected
17    that deposition above those levels would have deleterious effects on the provision of ecosystem
18    services in those areas, at this time it is possible only to describe the magnitude of the some of
19    the services currently being provided. Methods are not yet available to allow estimation of
20    changes in services due to nitrogen deposition.

21    3.5   REFERENCES
22    Adams, D., R. Alig, B.A. McCarl, and B.C. Murray. February 2005. FASOMGHG
23          Conceptual Structure, and Specification: Documentation. Available at
24          http://agecon2.tamu.edu/people/faculty/mccarl-bruce/FASOM.html. Accessed on
25          October 22, 2008.

26    Anderson, E. 1989. "Economic Benefits of Habitat Restoration: Seagrass  and the Virginia Hard-
27          Shell Blue Crab Fishery." North American Journal of Fisheries Management 9:140-149.

28    Acharya,  G., and L.L. Bennett. 2001. "Valuing Open  Space  and Land-Use Patterns in Urban
29          Watersheds." Journal of Real Estate Finance and Economics 22(2/3):221-237. Arrow K.

      March 2010                                89                 Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1          and A.C. Fisher. 1974. "Environmental Preservation, Uncertainty, and Irreversibility."
 2          The Quarterly Journal of Economics., 88(2):312-319.

 3    Arrow K. and A.C. Fisher. 1974. "Environmental Preservation, Uncertainty, and Irreversibility."
 4          The Quarterly Journal of Economics, 88(2):312-319.

 5    Banzhaf, S., D. Burtraw, D. Evans, and A. Krupnick. 2006. "Valuation of Natural Resource
 6          Improvements in the Adirondacks." Land Economics 82:445-464.

 7    Bartik, TJ. 1988. "Measuring the Benefits of Amenity Improvements in Hedonic Price
 8          Models "LandEconomics 64(2): 172-183.

 9    Bockstael, N.E., K.E. McConnell, and I.E. Strand. 1988. "Benefits from Improvements in
10          Chesapeake Bay Water Quality, Volume III." Environmental Protection Agency
11          Cooperative Agreement CR-811043-01-0.

12    Bockstael, N.E., K.E. McConnell, and I.E. Strand. 1989. "Measuring the Benefits of
13          Improvements in Water Quality: The Chesapeake Bay." Marine Resource Economics
14          6:1-18.

15    Boyd, J., and S. Banzhaf. 2007. "What are Ecosystem Services? The Need for Standardized
16          Environmental Accounting Units." Ecological Economics 63:616-626.

17    Bricker, S., D. Lipton, A. Mason, M. Dionne, D. Keeley, C. Krahforst, J. Latimer, and
18          J.Pennock. 2006. Improving Methods and Indicators for Evaluating Coastal Water
19          Eutrophication: A Pilot Study in the  Gulf of Maine. NOAA Technical Memorandum
20          NOS NCCOS 20. Silver Spring, MD: National Ocean Service, National Centers for
21          Coastal Ocean Science, Center for Coastal Monitoring and Assessment.

22    Bricker, S., Longstaff, B., Dennison, W., Jones, A., Boicourt, K., Wicks, C., and Woerner, J.
23          2007. Effects of Nutrient Enrichment in the Nation's Estuaries: A Decade of Change.
24          NOAA Coastal Ocean Program Decision Analysis Series No. 26. National Centers for
25          Coastal Ocean Science, Silver Spring, MD.
     March 2010                              90                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1   Bricker, S.B., J.G. Ferreira, and T. Simas. 2003. "An Integrated Methodology for Assessment of
 2          Estuarine Trophic Status." Ecological Modelling 169:39-60.

 3   Brown, L.H. 2002. Profile of the Annual Fall Foliage Tourist in Vermont: Travel Year
 4          2007.Report prepared for the Vermont Department of Tourism and Marketing.
 5          Burlington, VT: Vermont Tourism Data Center, University of Vermont.

 6   Burger, J.C., R.A. Redak, E.B. Allen, J.T. Rotenberry, and M.F. Allen. 2003. "Restoring
 7          Arthropod Communities in Coastal Sage Scrub." Conservation Biology 17(2):460-467.

 8   California Department of Fish and Game. n.d.
 9          http://www.dfg.ca.gov/habcon/nccp/images/region.gif.

10   Chen, L., and C.T. Driscoll. 2004. "Modeling the Response of Soil and Surface Waters in the
11          Adirondack and Catskill Regions of New York to Changes in Atmospheric Deposition
12          and Historical Land Disturbance." Atmospheric Environment 38:4099-4109.

13   Chesapeake Bay Program, n.d. "Chesapeake Bay Program: A Watershed Partnership." Available
14          at http://www.chesapeakebay.net/.

15   Chesapeake Bay Program. "Bay Grass Abundance (Baywide)." Available at
16          http://www.chesapeakebay.net/ status_baygrasses.aspx?menuitem=19652. Last updated
17          May 21, 2008.

18   Chesapeake Eco-Check. 2007. "Indicator Details: 2007. Water Quality Index." Available at
19          http://www.eco-check.org/reportcard/chesapeake/2007/indicators/water_quality_index/.
20          Environmental Systems Research Institute, Inc. (ESRI). 2002. "Census Block Groups:
21          DC, DE, MD, and VA." [CD-ROM]. ESRI Data & Maps 2002.

22   Cordell, H.K., CJ. Betz, G. Green, and M. Owens. 2005. Off-Highway Vehicle Recreation in the
23          United States, Regions and States: A National Report from the National Survey on
24          Recreation and the Environment (NSRE). Report prepared for the Forest Service's
25          National OHV Policy & Implementation Teams. USDA Forest Service.
     March 2010                              91                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    Cordell, K., B. Leeworthy, G.T. Green, C. Betz, and B. Stephens, n.d. The National Survey
 2          onRecreation & the Environment. Research Work Unit 4953.  Athens, GA: Pioneering
 3          Research on Changing Forest Values in the South and Nation USDA Forest Service
 4          Southern Research Station. Available at www.srs.fs.fed.us/trends.

 5    Daily, G.C., S. Alexander, P.R. Ehrlich, L. Goulder, J. Lubchenco, P.A. Matson, H.A. Mooney,
 6          S. Postel, S.H. Schneider, D. Tilman, and G.M. Woodwell.  1997. "Ecosystem
 7          Services:Benefits Supplied to Human Societies by Natural Ecosystems." Issues in
 8          Ecology 2:1-16

 9    DeHayes, D.H., P.G. Schaberg, GJ. Hawley, and G.R. Strimbeck,  1999. "Acidic Rain Impacts
10          on Calcium Nutrition and Forest Health." BioScience 49:789-800.

11    Driscoll, C.T. et al. 2003. "Effects of Acidic Deposition on Forest and Aquatic Ecosystems in
12          New York State." Environmental Pollution 123:327-336.

13    Driscoll, C.T., G.B. Lawrence, AJ. Bulger, TJ. Butler, C.S. Cronan, C. Eagar, K.F. Lamber,
14          G.E. Likens, J.L. Stoddard, and K.C. Weathers. 2001. Acidic Deposition in the
15          Northeastern United States: Sources and Inputs, Ecosystem Effects and Management
16          Strategies. BioScience 51:180-198.

17    Economist.com Dictionary at:
18          http://www.economist.com/research/economics/alphabetic.cfm?letter=E

19    Federal Register (2007) National Ambient Air Quality Standards for Ozone; Proposed Rule. 40
20          CFR 50; Federal Register 72: 37818

21    Federal Register (2008) National Ambient Air Quality Standards for Ozone; Final Rule. 40 CFR
22          parts 50 and 58; Federal Register 73:16436

23    Fenn, M.E., T.G. Huntington, S.B. McLaughlin, C. Eager, A. Gomez, and R.B.  Cook.
24          2006."Status of Soil Acidification in North America." Journal of Forest Science 52
25          (special issue):3-13.

26    Greene, W.H. 2003. Econometric Analysis, 5th Ed. New Jersey: Prentice Hall.

      March  2010                               92                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1   Haefele, M., R.A. Kramer, and T.P. Holmes. 1991. Estimating the Total Value of a Forest
 2          Quality in High-Elevation Spruce-Fir Forests. The Economic Value of Wilderness:
 3          Proceedings of the Conference. Gen. Tech. Rep. SE-78 (pp.  91-96).  Southeastern For
 4          Exper. Station. Asheville, NC: USDA Forest Service.

 5   Holmes, T.P. 1992. Economic Welfare Impacts of Air Pollution Damage to Forests in the
 6          Southern United States. Asheville, NC: U.S. Dept. of Agriculture, Forest Service,
 7          Southeastern Forest Experiment Station, pp. 19-26.

 8   Holmes, T., and R. Kramer. 1995. "An Independent Sample Test of Yea-Saying and Starting
 9          Point Bias in Dichotomous-Choice Contingent Valuation. " Journal of Environmental
10          Economics and Management 28:121-132.Final

11   Hutchison, R., and C.E. Kraft. 1994. "Hmong Fishing Activity and Fish Consumption." Journal
12          of Great Lakes Research 20(2):471-487.

13   Industrial Economics, Inc. June 2008. "The Economic Impact of the Clean Air Interstate Rule on
14          Recreational Fishing in the Adirondack Region of New York State." Prepared for the
15          Clean Air Markets Division, Office of Air and Radiation, U.S. EPA.

16   Irwin, E.G. 2002. "The Effects of Open Space on Residential Property Values." Land Economics
17          78(4):465-480.

18   J. Nilsson and Grennfelt, P. (Eds) (1988) Critical loads for sulphur and nitrogen. UNECE/Nordic
19          Council workshop report, Skokloster, Sweden. March  1988.  Nordic Council of Ministers:
20          Copenhagen.

21   Jenkins, D.H., J.  Sullivan, and G.S. Amacher. 2002. "Valuing High  Altitude Spruce-Fir
22          Forestlmprovements: Importance of Forest Condition and Recreation Activity." Journal
23          of'Forest Economics 8:77'-99.

24   Johnston, R.J., E.Y. Besedin, R.  lovanna, C.J. Miller, R.F. Wardwell, and M.H. Ranson.
25          2005."Systematic Variation in Willingness to Pay for Aquatic Resource Improvements
26          and Implications for Benefit Transfer: A Meta-Analysis." Canadian Journal of
27          Agricultural Economics 53:221-248.

     March 2010                              93                 Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1   Johnston, R.J., M.H. Ranson, E. Y. Besedin, and B.C. Helm. 2006. "What Determines
 2          Willingness to Pay per Fish? A Meta-Analysis of Recreational Fishing Values." Marine
 3          Resource Economics 21:1-32.

 4   Kahn, J.R., and W.M. Kemp. 1985. "Economic Losses Associated with the Degradation of an
 5          Ecosystem: The Case of Submerged Aquatic Vegetation in Chesapeake Bay." Journal of
 6          Environmental Economics and Management 12:246-263.

 7   Kaoru, Y. 1995. "Measuring Marine Recreation Benefits of Water Quality Improvements by the
 8          Nested Random Utility Model." Resource and Energy Economics 17(2):119-136.

 9   Kaoru, Y., V. Smith, and J.L. Liu. 1995. "Using Random Utility Models to Estimate the
10          Recreational Value of Estuarine Resources." American Journal of Agricultural
11          Economics 77:141-151.

12   Kaval, P., and J. Loomis. 2003. Updated Outdoor Recreation Use Values With Emphasis On
13          National Park Recreation. Final Report October 2003, under Cooperative Agreement CA
14          1200-99-009, Project number EVIDE-02-0070.

15   Kennedy, P. 2003. A Guide to Econometrics, pp. 312-313. Cambridge, MA: MIT Press.
16          Millennium Ecosystem Assessment (MEA). 2005. Ecosystems and Human Well-being:
17          Wetlands and Water. Synthesis. A Report of the Millennium Ecosystem Assessment.
18          Washington, DC: World Resources Institute.

19   Knowler, D. 2002. "A Review of Selected Bioeconomic Models:  With Environmental Influences
20          in Fisheries." Journal of Bioeconomics 4:163-181.

21   Kramer, A., T. Holmes, and M. Haefele. 2003. "Contingent Valuation of Forest Ecosystem
22          Protection." In Forests in a Market Economy., E.O. Sills and K.L. Abt, eds., pp. 303-320.
23          Dordrecht, The Netherlands: Kluwer Academic Publishers.

24   Krieger, DJ. 2001. Economic Value of Forest Ecosystem Services: A Review. Washington, DC:
25          The Wilderness Society.
     March 2010                               94                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1   Krupnick, A. 1988. "Reducing Bay Nutrients: An Economic Perspective." Mary land Law Review
 2          47(2):453-480.

 3   Land Trust Alliance. 2006. The 2005 National Land Trust Census Report. Washington,
 4          D.C.:Land Trust Alliance, November 30, 2006.

 5   Leeworthy, V.R., and P.C. Wiley. 2001. Current Participation Patterns inMarine
 6          Recreation.Silver Spring, MD: U.S. Department of Commerce, National Oceanic and
 7          Atmospheric Administration, National Ocean Service, Special Projects.

 8   Lipton, D. W. 1999. "Pfiesteria's Economic Impact on Seafood Industry Sales and Recreational
 9          Fishing." In Proceedings of the Conference, Economics of Policy Options for Nutrient
10          Management andPfiesteria. B. L. Gardner and L. Koch, eds., pp.  35-38. College Park,
11          MD: Center for Agricultural and Natural Resource Policy, University of Maryland.

12   Lipton, D. 2004. "The Value of Improved Water Quality to Chesapeake Bay Boaters." Marine
13          Resource Economics 19:265-270.

14   Lipton, D. 2006. "Economic Impact of Maryland Boating in 2005." University of Maryland Sea
15          Grant Extension Program. Available at
16          ftp://ftp.mdsg.umd.edu/Public/MDSG/rec_boat05.pdf.

17   Lipton, D. 2008. "Economic Impact of Maryland Boating in 2007." University of Maryland Sea
18          Grant Extension Program. Available at
19          ftp://ftp.mdsg.umd.edu/Public/MDSG/rec_boat07.pdf.

20   Lipton, D.W., and R.W. Hicks.  1999. "Linking Water Quality Improvements to Recreational
21          Fishing Values: The Case of Chesapeake Bay Striped Bass." In Evaluating the Benefits of
22          Recreational Fisheries. Fisheries Centre Research Reports 7(2), TJ. Pitcher, ed., pp. 105-
23          110. Vancouver: University of British Columbia.

24   Lipton, D.W., and R.W. Hicks.  2003. "The Cost of Stress: Low Dissolved Oxygen and the
25          Economic Benefits of Recreational Striped Bass Fishing in the Patuxent River." Estuaries
26          26(2A):310-315.
     March 2010                              95                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1   Loomis, J., D. Griffin, E. Wu, and A. Gonzalez-Caban. 2002. "Estimating the Economic Value
 2          of Big Game Habitat Production from Prescribed Fire Using a Time Series Approach.
 3          Journal of Forest Economics 2:119-29.

 4   Loomis, J.B., and D.S. White. 1996. "Economic Benefits of Rare and Endangered Species:
 5          Summary and Meta-Analysis." Ecological Economics 18(3): 197-206.

 6   Luzadis, V. A. and E.R. Gossett. 1996. "Sugar Maple." In Forest Trees of the Northeast.,
 1          J.P.Lassoie, V.A. Luzadis, and D.W. Grover, eds., pp. 157-166. Cooperative Extension
 8          Bulletin 235. Cornell Media Services. Available at
 9          http://maple.dnr.cornell.edu/pubs/trees.htm.

10   Mansfield, C.A., S.K. Pattanayak, W. McDow, R. MacDonald, and P. Halpin. 2005. "Shades of
11          Green: Measuring the Value of Urban Forests in the Housing Market." Journal of Forest
12          Economics \\(3y.m-\99.

13   Massey, M., S. Newbold, and B. Gentner. 2006.  "Valuing Water Quality Changes Using a
14          Bioeconomic Model of a Coastal Recreational Fishery." Journal of
15          EnvironmentalEconomics and Management 52:482-500.

16   Millennium Ecosystem Assessment (MEA). 2005a. Ecosystems and Human Well-being: Current
17          State and Trends, Volume  1. R. Hassan, R. Scholes, and N.  Ash, eds. Washington,
18          DCisland Press. Available at
19          http://www.millenniumassessment.org/documents/document.766.aspx.pdf.

20   Millennium Ecosystem Assessment (MEA). 2005b. Ecosystems and Human Well-
21          being:Synthesis. Washington, DC: World Resources Institute.

22   Mistiaen, J.A., I.E. Strand,  and D. Lipton. 2003.  "Effects of Environmental Stress on Blue Crab
23          (Callinectes sapidus) Harvests in Chesapeake Bay Tributaries." Estuaries 26(2a): 316-
24          322.

25   Montgomery, M. and M. Needelman. 1997. "The Welfare Effects of Toxic Contamination in
26          Freshwater Fish." Land Economics 73(2):211-223.
     March 2010                              96                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    Morgan, C., and N. Owens. 2001. "Benefits of Water Quality Policies: The Chesapeake Bay."
 2          Ecological Economics 39:271-284.

 3    Murray, T., and J. Lucy. 1981. "Recreational Boating in Virginia: A Preliminary Analysis."
 4           Special Report in Applied Marine Science and Ocean Engineering No. 251. Available at
 5          http://www.vims.edu/GreyLit/VIMS/sramsoe251 .pdf

 6    National Agricultural Statistics Service (NASS). June 12, 2008. "Maple Syrup Production Up 30
 7          Percent Nationwide." New England Agricultural Statistics, NASS, USDA. Spencer, D.M.,
 8          and D.F. Holecek. 2007. "Basic Characteristics of the Fall Tourism Market." Joz/ram
 9          Management 28:491 -504

10    National Association of State Foresters (NASF). 2009. Quadrennial Fire Review
11          2009.Washington, DC: NASF. Quadrennial Fire and Fuel Review Final Report 2009.

12    National Marine Manufacturers Association (NMMA).  2008. "2007 Recreational Boating
13           Statistical Abstract." Available at
14          http://www.nmma.org/facts/boatingstats/2007/files/Abstract.pdf

15    National Oceanic and Atmospheric Administration (NOAA). (2007, August). "Annual
16          Commercial Landing Statistics." Available at
17          http://www.st.nmfs.noaa.gov/stl/commercial/landings/annual_landings.html.

18    National Oceanic and Atmospheric Administration (NOAA). "Recreational Fisheries."
19          Available at www.st.nmfs.noaa.gov/stl/recreational/overview/overview.html. Last
20          updated February 12, 2009.

21    National Research Council. 2005.  Valuing Ecosystem Services: Toward Better Environmental
22          Decision-Making. Washington, DC: National Academies Press.

23    Parsons, G., A.O. Morgan, J.C. Whitehead, and T.C. Haab. 2006. "The Welfare Effects of
24          Pfiesteria-Related Fish Kills: A Contingent Behavior Analysis of Seafood Consumers."
2 5          Agricultural and Resource Economics Review 3 5 (2): 1 -9.
      March  2010                               97                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1   Peterson, D.E., M.S. Kanarek, M.A. Kuykendall, J.M. Diedrich, H.A. Anderson, P.L.
 2          Remington, and T.B. Sheffy. 1994. "Fish Consumption Patterns and Blood Mercury

 3   Levels in Wisconsin Chippewa Indians." Archives of Environmental Health 49(1):53 58.

 4   Poor, P.I, K.L. Pessagno, and R.W. Paul. 2007. "Exploring the Hedonic Value of Ambient
 5          Water Quality: A Local Watershed-Based Study." Ecological Economics 60:797-806.

 6   Smith, M.D. 2007. "Generating Value in Habitat-Dependent Fisheries: The Importance of
 7          Fishery Management Institutions." Land Economics 83(l):59-73.

 8   Sweeney, J. 2007. "Impacts of CAMD 2020 CAIR on Nitrogen Loads to the Chesapeake Bay."
 9          University of Maryland, Chesapeake Bay Program Office.

10   U.S. Department of Agriculture, Forest Service. Forest Inventory and Analysis National
11          Program, RPA Assessment Tables. 2002. Available at
12          http://Ncrs2.Fs.Fed.Us/4801/Fiadb/Rpa_Tabler/Draft_RPA_2002_Forest_Resource_Tabl
13          es.pdf.

14   U.S. Census Bureau. 2008a. "Annual Social and Economic (ASEC) Supplement." Available at
15          http://pubdb3.census.gov/macro/032008/hhinc/new02_001.htm.

16   U.S. Census Bureau. 2008b. "Housing Unit Estimates for Counties of MD and VA: April 1/2000
17          to July 1/2007." Available at http://www.census.gov/popest/housing/HU-EST2007-
18          CO.html.

19   U.S. Department of the Interior, Fish and Wildlife Service, and U.S. Department of Commerce,
20          U.S. Census Bureau. 2007. 2006 National Survey of Fishing, Hunting, and Wildlife-
21          Associated Recreation.

22   U.S. EPA.  1998.. EPA/630/R-95/002F. Risk Assessment Forum, Washington, DC. Available
23          from: http://www.epa.gov/ncea/ecorsk.htm.
     March 2010                               98                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1   U.S. Environmental Protection Agency (EPA), Office of Air and Radiation. November 1999.The
 2          Benefits and Costs of the Clean Air Act 1990 to 2010: EPA Report to Congress.EPA-
 3          410-R-99-001. Washington, DC: U.S. Environmental Protection Agency.

 4   U.S. Environmental Protection Agency (EPA). 2002. December. Environmental and Economic
 5          Benefit Analysis of Final Revisions to the National Pollutant Discharge Elimination
 6          System Regulation and the Effluent Guidelines for Concentrated Animal Feeding
 7          Operations (EPA-821-R-03-003). Washington, DC: U.S. Environmental Protection
 8          Agency, Office of Water, Office of Science and Technology.

 9   U. S. EPA (Environmental Protection Agency). 2006. Ecological Benefits Assessment Strategic
10          Plan. EPA-240-R-06-001. Office of the Administrator. Washington, DC. Available at
11          http://www.epa.gov/economics.

12   U.S. Environmental Protection Agency (EPA). 2008. Integrated Science Assessment for Oxides
13          of Nitrogen and Sulfur-Environmental Criteria.EPA/600/R-08/082. U.S. Environmental
14          Protection Agency, Office of Research and Development, National Center for
15          Environmental Assessment - RTF Division, Research Triangle Park, NC

16   US Environmental Protection Agency (EPA). 2009. Risk and Exposure Assessment for the
17          Review of the Secondary national Ambient Air Quality  Standards for Oxides of Nitrogen
18          and Oxides of Sulfur. EPA/452/R/09/008a. U.S environmental Protection Agency, Office
19          of Air Quality planning and Standards, Health and Environmental impacts Division,
20          Research Triangle Park, NC.

21   U.S. EPA (Environmental Protection Agency). 2009. Valuing the Protection of Ecological
22          Systems and Services: A report of the EPA Science Advisory Board. EPA-SAB-09-012.
23          Office of the Administrator. Washington, DC. Available at
24          http://yosemite.epa.gov/sab/sabproduct.nsf/WebBOARD/ValProtEcolSys&Serv?

25   Valigura, R.A., R.B. Alexander, M.S. Castro, T.P. Meyers, H.W. Paerl, P.E. Stacy, and
26          R.E.Turner. 2001. Nitrogen Loading in Coastal Water Bodies: An Atmospheric
27          Perspective.Washington, DC: American Geophysical Union.
     March 2010                              99                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1   Van Houtven, G. and A. Sommer. December 2002. Recreational Fishing Benefits: A Case Study
 2          of Reductions in Nutrient Loads to the Albemarle-Pamlico Sounds Estuary. Final Report.
 3          Prepared for the U.S. Environmental Protection Agency. Research Triangle Park,
 4          NC:RTI International.

 5   Van Houtven, G.L., J. Powers, and S.K. Pattanayak. 2007. "Valuing Water Quality
 6          Improvements Using Meta-Analysis: Is the Glass Half-Full or Half-Empty for National
 7          Policy Analysis?" Resource and Energy Economics 29:206-228.

 8   Vaughan, WJ. 1986. "The Water Quality Ladder." Included as Appendix B in R.C. Mitchell,
 9          and R.T. Carson, eds. The Use of Contingent Valuation Data for Benefit/Cost Analysis in
10          Water Pollution Control. CR-810224-02. Prepared for the U.S. Environmental Protection
11          Agency, Office of Policy, Planning, and Evaluation.

12   Virginia Department of Conservation and Recreation. 2007. "2006 Virginia Outdoors
13          Survey."Available at
14          http://www.dcr. virginia.gov/recreational_planning/documents/vopsurvey06.pdf

15   Wallace, KJ. 2007. "Classification of Ecosystem Services: Problems and Solutions." Ecological
16          Conservation 139:235-246.

17   Whitehead, J.C., T.C. Haab, and G.R.  Parsons. 2003. "Economic Effects of Pfiesteria." Ocean &
18          Coastal Management 46(9-10):845-858.

19
     March 2010                              100                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    4     ADDRESSING THE ADEQUACY OF THE CURRENT STANDARDS
 2          Based on the information in Chapters 2 and 3, we conclude that there is support in the
 3    available effects-based evidence for consideration of secondary standards for NOX and SOX that
 4    are protective against adverse ecological effects associated with deposition of NOX and SOX to
 5    sensitive ecosystems. Having reached this general conclusion, we then to the extent possible
 6    evaluate the adequacy of the current NOX and SOX secondary standards by considering to what
 7    degree risks to sensitivity ecosystems would be expected to occur in areas that meet the current
 8    standards. Staff conclusions regarding the adequacy of the current standards are based on the
 9    available ecological effects, exposure and risk-based evidence. In evaluating the strength of this
10    information, staff have taken into account the uncertainties and limitations in the scientific
11    evidence. This chapter addresses key policy relevant questions that inform our determination
12    regarding the adequacy of the structure and levels of the current secondary standards. The
13    chapter begins with a discussion of the structure of the current standards, followed by a
14    presentation of information on recent air quality relative to the existing standards, recent NOX
15    and SOX deposition levels, evaluation of recent deposition levels relative to levels where adverse
16    ecological effects have been observed, and a set of conclusions regarding the adequacy of the
17    current structure and levels of the standards.
18          It is also appropriate in this review to consider whether the current standards are adequate
19    to protect against the direct effects on vegetation resulting from ambient NC>2 and 862 which
20    were the basis for the current secondary standards. We will include a discussion of this issue in
21    the second draft policy assessment.

22    4.1   ARE THE STRUCTURES OF THE CURRENT NOX AND SOX
23          SECONDARY STANDARDS BASED ON RELEVANT
24          ECOLOGICAL INDICATORS SUCH THAT THEY ARE
25          ADEQUATE TO DETERMINE AND PROTECT PUBLIC WELFARE
26          AGAINST ADVERSE EFFECTS ON ECOSYSTEMS?
27          The current secondary NOX and SOX standards are intended to protect against adverse
28    effects to public welfare. For NOX, the current secondary standard was set identical to the
     March 2010                              101                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    primary standard15, e.g. an annual standard set for NC>2 to protect against adverse effects on
 2    vegetation from direct exposure to ambient NOX. For SOX, the current secondary standard is a 3-
 3    hour standard intended to provide protection for plants from the direct foliar damage associated
 4    with atmospheric concentrations of SC>2.
 5           The ISA has established that the major effects of concern for this review of the NOX and
 6    SOX standards are associated with deposition of N and S associated with atmospheric
 7    concentrations of NOX and SOX (see Chapter 2). As such, the current secondary standards do not
 8    reflect the conclusions of the ISA in the major areas of indicator, form, or averaging times. By
 9    using atmospheric NC>2 and 862, concentrations as indicators the current standards address only
10    a fraction of total atmospheric NOX and SOX, and do not take into account the effects from
11    deposition of total atmospheric NOX and SOX. By addressing short-term concentrations of SO2,
12    the  current SC>2 standard, while protective against direct foliar effects  from gaseous SOX, does
13    not  take into account the findings of effects in the ISA, which notes the relationship between
14    annual deposition of S and acidification effects which are likely to be  more severe and
15    widespread than  phytotoxic effects under current ambient conditions.  Acidification is a process
16    which occurs over time, as the ability of an aquatic  system to counteract acidic inputs is reduced
17    as natural buffers are used more rapidly than they can be replaced through geologic weathering.
18    The relevant period of exposure for ecosystems is therefore not the exposures captured in the
19    short averaging time of the current 862 standard. In addition, the ISA has concluded that NOX
20    and SOX and their deposition products jointly affect ecosystems, and as such the current separate
21    structure of the NOX and SOX secondary standards does  not take into account the joint ecological
22    effects of the two pollutants.
23           Current standards are specified as allowable single atmospheric concentration levels for
24    NC>2 or SC>2. This type of structure does not take into account variability in the atmospheric and
25    ecological factors that may alter the effects of NOX and  SOX  on public welfare. Consistent with
26    section 108, the ISA includes in the air quality criteria consideration of how these variable
27    factors impact the effects of ambient NOX and SOX on public welfare.  Secondary standards are
28    intended to address a wide variety of effects occurring in different  types of environments and
29    ecosystems. Ecosystems are not uniformly distributed either spatially  or temporally in their
      15 The current primary NO2 standard has recently been changed to the 3 year average of the 98thpercentile of the
      annual distribution of the 1 hour daily maximum of the concentration of NO2. The current secondary standard
      remains as it was set in 1971.

      March 2010                                102                 Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    sensitivity to air pollution. Therefore,  failure to account for the major determinants of variability,
 2    especially geologic conditions related to sensitivity to acidification and atmospheric conditions
 3    which govern rates of deposition, may lead to standards that do not provide appropriate levels of
 4    protection across ecosystems. We can state with confidence the current standards were not
 5    designed to be protective against those welfare effects tied to deposition of ambient NOX and SOX
 6    and thus are not likely to be adequate to protect public welfare against known or anticipated
 7    adverse effects from deposition.
 8          Because most areas of the U.S. are in attainment with the current NO2 and SOX standards,
 9    it is possible to evaluate current conditions, and evaluate the impact on public welfare from the
10    current effects on ecosystems from NOX and SOX deposition in areas that attain the current
11    standards that use NO2 and SO2 as indicators. In addition,  this chapter qualitatively addresses the
12    adequacy of the structures of the existing standards relative to ecologically relevant standards for
13    NOX and SOX, and sets up arguments for developing an ecologically relevant structure for the
14    standards as described in Chapter 5.

15    4.2   TO WHAT EXTENT ARE THE STRUCTURES OF  THE CURRENT
16          NOX AND SOX SECONDARY STANDARDS MEANINGFULLY
17          RELATED TO RELEVANT ECOLOGICAL INDICATORS OF
18          PUBLIC WELFARE EFFECTS?
19          The current secondary standard for NOX, set in 1971, using NO2 as the atmospheric
20    indicator, is 0.053  parts per million (ppm) (100 micrograms per cubic meter of air [|jg/m3]),
21    annual arithmetic average, calculated as the arithmetic mean of the 1-hour NO2 concentrations.
22    This standard was  selected to provide  protection to the public welfare against acute injury to
23    vegetation from direct exposure and resulting phytoxicity. During the last review of the NOX
24    standards, impacts associated with chronic acidification and eutrophication from NOX deposition
25    were acknowledged, but the relationships between atmospheric concentrations of NOX and levels
26    of acidification and eutrophication and associated welfare impacts were determined to be too
27    uncertain  to be useful as a basis for setting a national secondary standard (USEPA 1995).
28          The current secondary standard for SOX, set in 1971, uses SO2 as the atmospheric
29    indicator, is a 3-hour average of 0.5 ppm, not to be exceeded more than once per year. This
30    standard was selected to provide protection to the public welfare against acute injury to

      March 2010                               103                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    vegetation. In the last review of the SOX secondary standard, impacts associated with chronic
 2    acidification were acknowledged, but the relationships between atmospheric concentrations of
 3    SOX and levels of acidification, along with the complex interactions between SOX and NOX  in
 4    acidification processes, were cited as critical uncertainties which made the setting of secondary
 5    NAAQS to protect against acidification inappropriate at that time (USEPA 1982).
 6          In the previous independent reviews of the NOX and SOX secondary standards, each
 7    review acknowledged the additional impacts of NOX and SOX on public welfare through the
 8    longer term impact of the pollutants once deposited to ecosystems. However, the previous
 9    reviews cited numerous uncertainties as the basis for not addressing those impacts in the setting
10    of the standards. In addition, these previous reviews did not consider the common pathways of
11    impact for the two pollutants acting on the same ecosystem endpoints.
12          Three issues arise that call into question the ecological relevance of the current structure
13    of the secondary standards for NOX and SOX. One issue is the exposure period that is relevant for
14    ecosystem impacts. The majority of deposition related impacts are associated with depositional
15    loads that occur over periods of months to years. This differs significantly from exposures
16    associated with hourly concentrations of NC>2 and  862 as measured by the current standards.
17    Even though the NC>2 standard uses an annual average of NC>2, it is focused on the annual
18    average of 1-hour NC>2 concentrations,  rather than a cumulative metric or an averaging metric
19    based on daily or monthly averages. A  second issue is the choice of atmospheric indicators. NC>2
20    and SO2 are used as the component of oxides of nitrogen and sulfur that are measured, but they
21    do not provide a complete link to the direct effects on ecosystems from deposition of NOX and
22    SOX as they do not capture all relevant species of oxidized nitrogen and oxidized sulfur that
23    contribute to deposition. The ISA provides evidence that deposition related effects are linked
24    with total nitrogen and total sulfur, and thus all forms of oxidized nitrogen and oxidized sulfur
25    that are deposited will contribute to effects on ecosystems. This suggests that more
26    comprehensive atmospheric indicators should be considered in designing ecologically relevant
27    standards. Further discussions of the need for more ecologically relevant atmospheric indicators
28    as well as the relative contributions to deposition from various species of NOX and SOX can be in
29    found in Chapters 5 and 6. The third issue is that the current standards reflect separate
30    assessments of the two individual pollutants, NC>2  and SC>2, rather than assessing the joint
31    impacts of deposition of NOX and SOX to ecosystems, recognizing the role that each pollutant

      March 2010                               104               Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    plays in jointly affecting ecosystem indicators, functions, and services. The clearest example of
 2    this interaction is in assessment of the impacts of acidifying deposition on aquatic ecosystems.
 3           Acidification in an aquatic ecosystem depends on the total acidifying potential of the
 4    deposition of both N and S from both atmospheric deposition of NOX and SOX as well as the
 5    inputs from other sources of N and S such as reduced nitrogen and non-atmospheric sources. It is
 6    the joint impact of the two pollutants that determines the ultimate effect on organisms within the
 7    ecosystem, and critical ecosystem functions  such as habitat provision and biodiversity. Standards
 8    that are set independently are less able to account for the contribution of the other pollutant. This
 9    suggests that interactions between NOX and SOX should be a critical element of the conceptual
10    framework for ecologically relevant standards. There are also important interactions between
11    NOX and SOX and reduced forms of nitrogen, which also contributes to acidification and nutrient
12    enrichment. While the standards do not address reduced forms of nitrogen in the atmosphere, it is
13    important that the  structure of the standards address the role of reduced nitrogen in determining
14    the ecological effects resulting from deposition of atmospheric NOX and SOX.  Consideration will
15    also have to be given to account for loadings coming from non-atmospheric sources as
16    ecosystems will respond to these sources as well.
17           In addition to the fundamental issues discussed above, the current structures of the
18    standards do not address the complexities in the responses of ecosystems to deposition of NOX
19    and SOX. Ecosystems contain complex grouping of organisms that respond in various ways to the
20    alterations of soil and water that result from  deposition of nitrogen and sulfur compounds.
21    Different ecosystems therefore respond in different ways depending on a multitude of factors
22    that control how deposition is integrated into the system. For example,  the same levels of
23    deposition falling on limestone dominated soils have a very different effect than those falling on
24    shallow glaciated soils underline with granite. One  system may over time display no obvious
25    detriment while the other may experience a catastrophic loss in fish communities. This degree of
26    sensitivity  is a function of many atmospheric factors which control rates of deposition as well as
27    ecological  factors which control how an ecosystem responds to that deposition.  The current
28    standards do not take into account spatial and seasonal variations not only in depositional
29    loadings but also in sensitivity of ecosystems exposed to those loadings.
      March 2010                                105                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    4.3   TO WHAT EXTENT DO CURRENT MONITORING NETWORKS
 2          PROVIDE A SUFFICIENT BASIS FOR DETERMINING THE
 3          ADEQUACY OF CURRENT SECONDARY NOX AND SOX
 4          STANDARDS?
 5          There are over 1000 ground level monitoring platforms (Figures 4-1 and 4-2) that provide
 6    measurements of some form of atmospheric nitrogen or sulfur. The key pollutants for this
 7    assessment are total oxidized nitrogen (NOy), total reduced nitrogen (NHX), and total sulfur (ST).
 8    Total reactive oxidized atmospheric nitrogen, NOy, is defined as NOX (NO and NO2) and all
 9    oxidized NOX products: NOy = NO2 + NO + HNO3 + PAN +2N2O5 + HONO+ NO3  + organic
10    nitrates + particulate NO3 (Finlayson-Pitts and Pitts, 2000). This definition of NOy reflects the
11    operational principles of standard measurement techniques in which all oxidized nitrogen species
12    are converted to nitrogen oxide (NO) through catalytic reduction and the resulting NO is detected
13    through luminescence. Thus, NOy is truly defined as total oxidized nitrogen as converted to NO.
14    NOy is not a strict representation of the all moles of oxidized nitrogen as the diatomic nitrogen
15    species such as N2Os yield 2 moles of NO. This definition is consistent with the relationship
16    between atmospheric nitrogen and acidification processes as the reported NOy provides a direct
17    estimate of the potential equivalents available for acidification.  Total reduced nitrogen (NHX)
18    includes ammonia, NH3, plus ammonium, NH4 (EPA, 2008). Reduced nitrogen plus oxidized
19    nitrogen is referred to as total reactive nitrogen. Total sulfur (ST) includes SO2 gas and
20    particulate sulfate, SO4. Ammonium and sulfate are components of atmospheric particulate
21    matter as well as directly measured and modeled in precipitation as direct deposition
22    components. As discussed in this section, there are only very limited routine measurements of
23    total oxidized and reduced nitrogen. In addition, existing monitoring networks do not provide
24    adequate geographic coverage to fully assess concentrations and deposition of reactive nitrogen
25    and sulfur in and near sensitive ecosystems.
26
     March 2010                              106               Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
1
2
3
                                              MAPI (All N)
                    NCore, NOY(2009), SEARCH, PA MS/SLA MS, CASTNET, IMPROVE
             •*•  NCore
             Q|  Rural NCore
             »  SEARCH
             Q  Rural SEARCH
             •  PAMS_HO- NO2- NQX-NQY_2009
             *  SLAMS_NO-N02-N OX-HOY
             »  CASTNET-NPS
                CASTNET-EPA
             •  IMPR QVE_Hrtr3te£_20Q6
Figure 4-1. Routinely operating surface monitoring stations measuring forms of
atmospheric nitrogen.
     March  2010
                                       107
Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
                                         MAP 6-1 a (AIIS)
                     NCore, SO2(2008), SEARCH, CASTNET, IMPROVE, and
                          Trends/Supplemental Speciation Sites (2008)
                            - SO2(2008) includes NAMS / SLAMS / PAMS -
           • SEARCH
           D Rural SEARCH
           , CASTNET-NPS
             CASTNET-EPA
           A Special! on_Sulfates_2003
           • IMPROVE Sulfales 2006
 1
 2          Figure 4-2. Routinely operating surface monitoring stations measuring forms of
 3          atmospheric sulfur.
 4          The principal monitoring networks include the regulatory based State and Local Air
 5   Monitoring Stations (SLAMS) providing mostly urban-based 862, NO and NOX, the PM2.5
 6   chemical speciation networks Interagency Monitoring of Protected visual Environments
 7   (IMPROVE) and EPA's Chemical Speciation Network (CSN) providing particle bound sulfate
 8   and nitrate, and the Clean Air Status and Trends Network (CASTNET) providing weekly
 9   averaged values of SO2, nitric acid, and particle bound sulfate, nitrate and ammonium. The
10   private sector supported SouthEastern Aerosol Research and Characterization (SEARCH) Study
11   network of 4-8 sites in the southeast provides the only routinely operating source of true
12   continuous NO2, ammonia, and nitric acid measurements. SEARCH also provides PM2.5 size
13   fractions of nitrate and sulfate. Collectively, the SLAMS, Photochemical Assessment
14   Measurement Stations (PAMS), SEARCH and NCore networks will provide over 100 sites
15   measuring NOy (Figure 4-3). The NCore network (Scheffe et al., 2009) is a multiple pollutant
     March 2010
108
Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1   network with co-located measurements of key trace gases (CO, SO2, 63, NO and NOy), PM2.5
 2   and PM(10-2.5) mass and PM2.5 chemical speciation. Additional air pollutants, particularly
 3   volatile organic compounds (VOCs), will be measured at those sites that are part of the existing
 4   PAMS and National Air Toxics Trends (NATTS) platforms. The NATTS (EPA, 2008) include
 5   27 stations across the U.S. that monitor for a variety of hazardous air pollutants and are intended
 6   to remain in place to provide a longe term record. Additional measurements of ammonia and
 7   possibly true NO2 are under consideration. True NO2 is noted to differentiate from the NO2
 8   determined through routine regulatory networks that have known variable positive bias for NO2.
 9   The network currently is  being deployed and expected to be operational with nearly 75 sites by
10   January, 2011.  The sites are intended to serve as central site monitors capturing broadly
11   representative (e.g., not strongly influenced by nearby  sources) air quality in a suite of major and
12   mid size cities, and approximately 20 sites are located in rural locations.
     March  2010                               109                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
                                              MAP 4
                         Currents. Planned Routine NOY Monitoring Sites
                                   NCore, NOY(2009), SEARCH
 1
 2          Figure 4-3. Anticipated network of surface based NOy stations based on 2009
 3          network design plans. The NCore stations are scheduled to be operating by
 4          January, 2011.
 5          There are significant measurement gaps for characterizing NOy, NHX and SC>2 in the
 6   nations ambient air observation networks (EPA, 2008) that lead to greater reliance on air quality
 7   modeling simulations to describe current conditions. National design of routinely operating
 8   ambient air monitoring networks is driven mostly by data uses associated with implementing
 9   primary NAAQS, with noted exceptions of the CASTNET and IMPROVE networks In addition
10   to significant spatial gaps in sensitive ecosystem areas that arise from a population oriented
11   network design, the current measurements for primary and secondary nitrogen are markedly
12   different and in some instances of negligible value for secondary  NOX and SOX standards. For
13   example, a true NOX (NO plus NO2) measurement typically would capture less than 50% (see
14   discussion below) of the total regional NOy mass in rural locations as the more aged air masses
15   contain significant oxidized nitrogen products in addition to NOX. Note that the NOX monitors
     March 2010
110
Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    used for NAAQS primary compliance purposes do capture varying amounts of transformed
 2    nitrogen species; however, the method provides biased low estimates with significant airshed
 3    induced variability relative to true NOy. With the exception of the SEARCH network in the
 4    southeast, there are virtually no routine networks that measure ammonia, although EPA is
 5    considering options for ammonia sampling in CASTNET and NCORE networks. Ammonium is
 6    reported in EPA chemical speciation networks, although the values are believed to be biased low
 7    due to ammonia volatization.
 8           CASTNET provides mostly rural measurements of 862, total nitrate, and ammonium, and
 9    affords  an existing infrastructure useful for future monitoring in support of a NOX and SOX
10    secondary standard. However, the lack of NOy, SOX and NHX measurements in sensitive
11    ecosystems will require attention in the N/S secondary standard proposal.
12           As a result of the limited monitoring networks for NOy and SOX in sensitive ecosystems,
13    we are unable to use current monitoring data to fully assess whether the current standards have
14    resulted in levels of NOy and SOX in sensitive ecosystems that would result in  deposition levels
15    that are or are not causing ecological effects adverse to public welfare. We supplement the
16    available monitoring data with the use of sophisticated atmospheric modeling  conducted using
17    EPA'sCMAQmodel.

18           4.3.1   What does the NADP monitoring network provide and what are the major
19                 limitations?
20           The National Atmospheric Deposition Program (NADP) includes approximately 250
21    sites (Figure 4-4)  across the U.S. providing annual total wet deposition based on weekly
22    averaged measures of wet deposition of nitrate, ammonium and sulfate ions based on the
23    concentrations of these ions in precipitation samples. Meteorological models have difficulty in
24    capturing the correct spatial and temporal features of precipitation events, raising the importance
25    of the NADP as a principal source of precipitation chemistry. The NADP has enabled several
26    organizations to participate in a measurement program with a centralized laboratory affording
27    measurement and analysis protocol consistency nationwide. Virtually every CASTNET site is
28    located at an NADP site and the combined NADP/CASTNET infrastructure is a starting point for
29    discussions addressing future N/S monitoring needs. The Organic bound nitrogen is not analyzed
30    routinely in NADP samples. Consideration might be given to adding NADP sites in locations
      March  2010                               111                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
 1   where ambient air monitoring is conducted to assess compliance with a secondary NOX/SOX
 2   standard.
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
                         Ammonium ion wet deposition, 2005
                                                                               Ammonium as NH,*
                                                                                   (kg/ha)
           Sites not pictured:
           AK03    0.3 kg/ha
           VI01     0.4 kgJha
       National Atmospheric Deposition Program/National Trends Network
       http://nadp.sws.uiuc.edu
                                                                                     0.5.1.0
                                                                                     1.0-1.5
                                                                                     2.5 - 3.0
                                                                                     3.0 - 3.5
                                                                                     3.5 - 4.0
                                                                                     4.0-4.6
                                                                                     >4,5
       Figure 4-4. Location of approximately 250 National Atmospheric Deposition
       Monitoring (NADP) National Trends Network (NTN) sites illustrating annual
       ammonium deposition for 2005. Weekly values of precipitation based nitrate,
       sulfate and ammonium are provided by NADP.
       4.3.2  How do we characterize deposition through Monitoring and Models?
       Routinely available directly measured precipitation to quantify wet deposition of sulfur
and nitrogen species are provided through the NADP. Dry deposition is not a directly measured
variable in routine monitoring efforts and, for all practical purposes, largely will remain a
research endeavor that supports the parameterizations used for estimating dry deposition, as
opposed to striving to develop operational methods. Estimates of dry deposition based on
observations are provided through the CASTNET program. However,  dry deposition is a
calculated value represented as the product of ambient concentration (either observed or
estimated through air quality modeling) and deposition velocity, Dep®"7 = v®"7 • C^mb
     March  2010
                                           112
Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1          Deposition velocity is modeled as a mass transfer process through resistance layers
 2    associated with the canopy, uptake by vegetation, water and soil which collectively are
 3    influenced by micrometeorology, land surface and vegetation types and species specific
 4    solubility and reactivity. Dry deposition is calculated through deposition velocity models
 5    capturing these features and using species specific ambient air concentrations. This approach
 6    conceptually is similar using either observed or modeled air concentrations. Dry deposition
 7    estimates from the Community Multi-scale Air Quality (CMAQ) model (EPA, 1999) have been
 8    used in this assessment to provide spatially more resolved and extensive estimates of dry
 9    deposition for sulfur and all reactive nitrogen (oxidized and reduced) species (CASTNET does
10    not capture important gases such as nitrogen dioxide and peroxyacetyl nitrate). All of the
11    relevant meteorological, land use, vegetation and elevation data required to estimate deposition
12    velocities are generated or accessible in the CMAQ and/or meteorological pre-processors.

13          4.3.2.1  Why are we using CMAQ to model deposition? How are we using it? Why is
14                   CMAQ the right model to use? What is the spatial and temporal resolution of
15                   CMAQ? What are the model years ? What are the limitations to CMAQ?
16          CMAQ provides a platform that allows for a consistent mass accounting approach across
17    ambient concentrations and dry and wet deposition values. Recognizing the limitations of
18    ambient air networks, CMAQ was used to estimate dry deposition to complement NADP wet
19    deposition for MAGIC modeling and for the FAB  critical  load modeling. CMAQ promotes
20    analytical consistency and efficiency across analyses of multiple pollutants. EPA's Office of
21    Research and Development continues to enhance the underlying deposition science in CMAQ.
22    For the purposes of this policy assessment, CMAQ provides a consistent platform incorporating
23    the atmospheric and  deposition species of interest over the entire United States. The caveats and
24    limitations of the use of model predictions are largely associated with the general reliance on
25    calculated values, rather than measurements. Model evaluation addressing the comparison of
26    predictions with observed values is addressed in the REA. Currently, there are efforts to improve
27    a number of nitrogen related processes in CMAQ,  recognizing comparatively less uncertainty
28    with the treatment of sulfur. Active areas  of model process improvement are in the treatment of
29    lightning generated NOX and the transference of nitrogen between atmospheric and terrestrial and
30    aquatic media, often referred to as bi-directional flux. Lightning NOX potentially provides a
      March  2010                               113                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    significant contribution to wet deposition as the resulting NOX is rapidly entrained into aqueous
 2    cloud processes. Both the thermodynamics of soil processes and mass transfer of nitrogen
 3    species across the surface-atmosphere interface is governed by an assortment of temperature,
 4    moisture, advection and concentration patterns. These processes and mass transfer relationships
 5    are coupled within the emissions, meteorological, and chemical simulation processes and
 6    associated surface/vegetation and terrain information incorporated in or accessed by the CMAQ.
 7    In addition to research activities to improve the characterization of nitrogen-related processes in
 8    CMAQ, efforts are also underway to improve the general characterization of ammonia emissions
 9    which remains as an area of large uncertainty due to limited source data and the ubiquitous
10    nature of these emissions. Another challenge for regional/national air quality modeling is
11    properly representing the effects on pollutant concentrations, precipitation and therefore
12    deposition of variable terrain features, particularly steep mountain-valley gradients and the
13    interfaces to wide open basins encountered in the Western United States.
14           The CMAQ was used in this assessment because it is the state of science model for
15    treating simulating sources, formation, and fate of nitrogen and sulfur species. In addition to
16    undergoing periodic independent scientific peer review, CMAQ bridges the scientific and
17    regulatory communities as it is used extensively by EPA for regulatory air quality assessments
18    and rules. CMAQ provides hourly estimates of the important precursor, intermediate and
19    secondarily formed species associated with atmospheric chemistry and deposition processes
20    influencing ozone, particulate matter concentrations and sulfur and nitrogen deposition.
21    Simulations based on horizontal spatial scale resolutions of 12 km and 36 km were used in this
22    PAD for 2002-2005.

23    4.4    WHAT IS OUR BEST CHARACTERIZATION OF ATMOSPHERIC
24           CONCENTRATIONS OF NOY AND SOX,  AND DEPOSITION OF N
25           AND S?
26           Air quality models and blending of model  results and observations are used to
27    characterize current environmental state conditions due to the relative sparseness of monitoring
28    coverage in sensitive ecosystems as well as gaps in coverage for specific  atmospheric species of
29    N and S most relevant to deposition, such  as NOy, in available monitoring platforms.
      March  2010                               114               Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1          4.4.1   What are the current atmospheric concentrations of reactive nitrogen, NOy,
 2                 reduced nitrogen, NHX, sulfur dioxide, SOi, and sulfate, SC>4?
 3          To provide information for use in characterizing the adequacy of the current standards,
 4    we assess the best available data for estimating the ambient concentrations of the major sources
 5    of atmospheric nitrogen and sulfur across the U.S. Acidification and nutrient enrichment
 6    processes are largely dependent on the cycling of total nitrogen and sulfur species. From an
 7    atmospheric  perspective, it is convenient and consistent with current measurement and modeling
 8    frameworks to consider the reduced and oxidized forms of atmospheric nitrogen. Virtually all
 9    atmospheric  sulfur is considered oxidized sulfur in the forms of particulate bound sulfate and
10    gaseous sulfur dioxide. In order to assess current concentrations of reactive nitrogen and sulfur
11    we evaluated data available from monitoring the existing networks as well as from the CMAQ
12    model. Regarding the monitoring data, there are a number of important issues in understanding
13    the measurements of NOy provided by different monitoring networks. In principle, measured
14    NOy is based on catalytic conversion of all oxidized species to NO followed by
15    chemiluminescence NO detection. We recognize the caveats associated with instrument
16    conversion efficiency and possible inlet losses. The CMAQ treats the dominant NOy species as
17    explicit species while the minor contributing non-PAN organic nitrogen compounds are
18    aggregated. Atmospheric nitrogen and sulfur are largely viewed as regional air quality issues due
19    to the importance of chemical conversion of primary emissions into secondarily formed species;
20    a combination of ubiquitous sources, particularly mobile source emissions of NOX, and elevated
21    emissions of NOX and SO2 that aid pollutant mass dispersal and broader physical transport over
22    large distances. In effect, the regional nature is due to both transport processes as well as the
23    relatively ubiquitous nature of sources combined with chemical processes that tend to form more
24    stable species with extended atmospheric lifetimes. This regionalized effect, particularly
25    throughout the Eastern United States, dominates the overall patterns discussed below of
26    secondarily formed species such as sulfate or NOy, which is an aggregate of species where the
27    more aged air masses consisting largely of chemically processed air is dominated by secondarily
28    formed peroxyacetyl  nitrate (PAN),  particulate nitrate and nitric acid.
29          Nationwide maps of CMAQ-predicted 2005 annual average NOy, NHX (NH3 and NH4),
30    NH3, NH4, ST, SO/t, and SO2 are provided in figures 4-5 through 4-11 respectively. Given the
31    considerable gaps in air quality observation networks as discussed in the REA and ISA (2008),

      March 2010                                115               Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    modeled concentration patterns are used here to illustrate national representations of current air
 2    quality conditions for nitrogen and sulfur. The 2005 model year reflects the most recent available
 3    simulation for inclusion in this policy assessment. In addition, figures 4-12 and 4-13 provide
 4    maps of 2005 annual average SO2 and SO/t, respectively based on CASTNET observations. Site
 5    specific annual average 2005 NOy measured concentrations at SLAMS (Figure 4-14) are
 6    typically are less than 40 ppb., The spatial patterns for the 2005  modeled and observed NOy and
 7    SC>2 concentrations are similar to the 2002 CMAQ-based maps provided in the REA., largely
 8    capturing the influence of major source regions throughout the nation. A spreading of the
 9    oxidized sulfur fields (Figures 4-5 and 4-6), relative to 862, is consistent with  sulfate
10    transformation and associated air mass aging and transport. Ammonia and ammonium
11    concentration patterns (Figure 4-4) are influenced strongly by the ammonia emissions
12    distribution, with marginal spreading associated with the addition of NFLj. The NHX fields are
13    more strongly influenced by source location, relative to sulfur, based on the fast removal of
14    atmospheric ammonia through deposition. Total  deposition for nitrogen and sulfur (Figures 4-15
15    and 4-16) basically follow the patterns of ambient air concentrations.
16           Current conditions indicate that the current 862 and NC>2 secondary standards are not
17    exceeded (Figures 4-17 and 4-18) in locations where ecological  effects have been observed, and
18    where critical loads of nitrogen and sulfur are exceeded. This is  consistent with the fact that NC>2
19    accounts for only a fraction of NOy, and thus reductions in NC>2  emissions would not be expected
20    to fully address concentrations of NOy. The map in Figure 4-19  further illustrates this point by
21    showing that the  contribution of NC>2 to NOy is often less than 50% in rural areas.
      March 2010                                116                Draft - Do Not Quote or Cite

-------
          Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
1
2
3
                                     AMAD 2005af CMAQ — NOy (ppb)
Figure 4-5. 2005 CMAQ modeled annual average NOy (ppb). These maps will be
replaced with full CONUS maps in the next draft.
    March  2010
                                    117
Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
                                                                                  Legend
                                                                                      >= 1.0to<3.0
                                                                                      >= 30 to < 5.0
                                                                                      >= 5.0 to < 7.0
                                                                                      =•= 7.0 to < 10.0
                                                                                      >= 10.0
                                       AMAD 2005af CMAQ — NHx (ug/m3)
2
3
4
Figure 4-6. 2005 CMAQ modeled annual average total reduced nitrogen (NHX)
(as ng/m3 nitrogen)
     March 2010
                                      118
Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
1




2




3
                                     AMAD 2005af CMAQ — NH3 (ug/m3)
Figure 4-7. 2005 CMAQ modeled annual average ammonia, NHs, (as ng/m N)
     March 2010
                                     119
Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
1




2




3
                                     AMAD 2005af CMAQ — NH4 (ug/m3)
Figure 4-8. 2005 CMAQ modeled annual average ammonia, NH4, (as ng/m N)
     March 2010
                                     120
Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
                                     AMAD 2005af CMAQ — ST (ug/m3)
2
3

4
5
Figure 4-9. 2005 CMAQ modeled annual average SOX, (as ng/m  S from 862 and
S04).
     March  2010
                                    121
Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
1
2

3
                                        AMAD 2005af CMAQ — SO2 (ug/m3)
                                                                                    Legend
                                                                                    I   l<05
                                                                                    I   | >=0.5lo< 1.0
                                                                                    I   |>=1.0tO<3.0
                                                                                    I   | >=• 3,0 to < 5.0
                                                                                    I   | >= 5,0 to < 7.0
                                                                                    ^H >= 7.0 to * 10.0
                                                                                    ^B ~-'= 10 ° *° *' 20 °
                                                                                    ^H =•= 20.0
Figure 4-10. 2005 CMAQ modeled annual average SO2 (as ng/m  S)
     March  2010
                                               122
Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
1




2




3
                                     AMAD 2005af CMAQ — SO4 (ug/m3)
Figure 4-11. 2005 CMAQ modeled annual average SO4 (as ng/m S).
     March 2010
                                     123
Draft - Do Not Quote or Cite

-------
          Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
1
2
3
4
5
6
7
Figure 4-12. 2005 annual average sulfur dioxide concentrations based on
CASTNET generated by the Visibility Information Exchange Web Sysytem
(VIEWS).
Figure 4-13. 2005 annual average sulfate concentrations based on CASTNET
generated by the Visibility Information Exchange Web Sysytem (VIEWS).
    March 2010
                                   124
Draft - Do Not Quote or Cite

-------
          Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
1
2
3
                            Annual Average NOY Concentrations (2005)
Figure 4-14. Annual average 2005 NOy concentrations from reporting stations in
AQS.
    March  2010
                                    125
Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
1
2
                                                                                             <2,0
                                                                                             >= 2.0 to < 3.0
                                                                                             >=3.0tO<4.0
                                                                                             >- 4.0 to < 50
                                                                                             >= 5.0 to < 7.0
                                                                                             >=7.0to<9.0
                                                                                             >=9.Qto< 14,0
                                                                                             >= 14.0 to < 20.0
                                                                                             >= 20.0
                                 AMAD 200Saf CMAO —
                          Oxidized Nitrogen Deposition ( kgN/Ha/Yr )
Figure 4-15. 2005 CMAQ modeled Oxidized Nitrogen Deposition (kgN/Ha/Yr).
     March 2010
                                        126
Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
1
2
3
                                              AMAD 2005af CMAQ —
                                        Oxidized Sulfur Deposition { kgS/Ha/Yr )       „-•
                                                                                         Legend
                                                                                 »=1.0to<2.0
                                                                                 >= 2.0 lo < 3.0
                                                                                 >=3.0(o<6.0
                                                                                 >=6.0lo< 10.0
                                                                                 >- 10.0 to < 16.0
                                                                                 >= 16.0 to < 24.0
                                                                                 >= 24.0 10 < 30.0
                                                                                 >= 30.0
Figure 4-16. 2005 CMAQ modeled Oxidized Sulfur Deposition (kgS/Ha/Yr).
     March 2010
                                         127
Draft - Do Not Quote or Cite

-------
          Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
                              3-hr Max SO2 Concentrations (2005)
1
2
3
4
5
6
7
8
Figure 4-17. Three hour average maximum 2005 SC>2 concentrations based on the
SLAMS reporting to EPA's Air Quality System (AQS) data base. The current
SC>2 secondary standard based on the maximum 3 hour average value is 500 ppb,
a value not exceeded. While there are obvious spatial gaps, the majority of these
stations are located to capture maximum values generally in proximity to major
sources and high populations. Lower relative values are expected in more remote
acid sensitive areas.
    March 2010
                                   128
Draft - Do Not Quote or Cite

-------
          Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
                             Annual Average NO2 Concentrations (2005)
1
2
3
4
5
6
7

8
Figure 4-18. Annual average 2005 NC>2 concentrations based on the SLAMS
reporting to EPA's Air Quality System (AQS) data base. The current NC>2
secondary standard is 53 ppb, a value well above those observed. While there are
obvious spatial gaps, the stations are located in areas of relatively high
concentrations in highly populated areas. Lower relative values are expected in
more remote acid sensitive areas.
    March  2010
                                    129
Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
                                   Layer 1 (NOY[1]- NO2[2])/ NOY[1]
 9
10
11
12
13
14
        231

        221

        211

        201

        191

        181

        171

        161

        151

        141

        131

       - 121

        111

        101

        91

        81

        71
 1
 2
 3
 4
 5
 6
 1   4.5
                                  December 31,0002 00:00:00 UTC
                                Mill (10, 34) = 0.175, Max (5,10) = 0.915
      Figure 4-19. 2005 CMAQ derived annual average ratio of (NOy - NO2)/NOy. The
      fraction of NO2 contributing to total NOy generally is less than 50% in the
      Adirondack and Shenandoah case study areas. The ratio reflects the relative air
      mass aging associated with transformation of oxidized nitrogen beyond NO and
      NO2 as one moves from urban to rural locations.
      ARE ADVERSE EFFECTS ON THE PUBLIC WELFARE
      OCCURRING UNDER CURRENT AIR QUALITY CONDITIONS
      FOR NO2 AND SO2 AND WOULD THEY OCCUR IF THE NATION
      MET THE CURRENT SECONDARY STANDARDS?
      The previous sections have established that almost all areas of the U.S. were at
concentrations of SO2 and NO2 below the levels of the current standards. In many locations, SO2
and NO2 concentrations are substantially below the levels of the standards. This suggests that
levels of deposition and any effects on ecosystems due to deposition of NOX and SOX under
     March 2010
                                        130
Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    recent conditions are occurring even though areas meet or are below current standards. This
 2    section focuses on summarizing the evidence of effects occurring at deposition levels consistent
 3    with recent conditions.
 4           The ISA summarizes the available studies of relative nitrogen contribution and finds that
 5    in much of the U.S., NOX contributes from 50 to 75 percent of total atmospheric deposition [ISA
 6    Section 2.8.4]. While the proportion of total nitrogen loadings associated with atmospheric
 7    deposition of nitrogen varies across locations (N deposition in the Eastern U.S. includes
 8    locations with greater than 9 kg N/ha/year, and in the central U.S. high deposition locations with
 9    values on the order of 6 to 7 kg N/ha/year), the ISA indicates that atmospheric N deposition is
10    the main source of new anthropogenic N to most headwater streams, high elevation lakes, and
11    low-order streams. Atmospheric N deposition contributes to the total N load in terrestrial,
12    wetland, freshwater, and estuarine ecosystems that receive N through multiple pathways. In
13    several large estuarine systems, including the Chesapeake Bay, atmospheric deposition accounts
14    for between 10 and 40 percent of total nitrogen loadings.
15           Atmospheric concentrations of SOX account for nearly all S deposition in the US. For the
16    period 2004-2006, mean S deposition in the U.S. was greatest east of the Mississippi River with
17    the highest deposition amount, 21.3 kg S/ha/yr, in the Ohio River Valley where most recording
18    stations reported 3 year averages >10 kg S/ha/yr. Numerous other stations in the East reported S
19    deposition >5 kg S/ha/yr. Total S deposition in the U.S. west of the 100th meridian was
20    relatively low, with all recording stations reporting <2 kg S/ha/yr and many reporting <1 kg
21    S/ha/yr. S was primarily deposited in the form of wet SC>4 2 followed in decreasing order by a
22    smaller proportion of dry SC>2 and a much smaller proportion of deposition as dry SC>42  .
23           New scientific evidence exists to address each of the areas of uncertainty raised in the
24    previous reviews (summarized above). Based on the new evidence, the current ISA concludes
25    that:
26           (1) The evidence is sufficient to infer a causal relationship between acidifying deposition
27              (to which both NOX and SOX contribute) and effects on biogeochemistry related to
28              terrestrial and aquatic ecosystems; and biota in terrestrial and aquatic ecosystems.
29           (2) The evidence is sufficient to infer a causal relationship between N deposition, to
30              which NOX and NHX contribute, and the alteration of A) biogeochemical cycling of N
31              and carbon in terrestrial, wetland, freshwater aquatic, and coastal marine ecosystems;

      March 2010                               131                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1              B) biogenic flux of methane (CH4), and N2O in terrestrial and wetland ecosystems;
 2              and C) species richness, species composition, and biodiversity in terrestrial, wetland,
 3              freshwater aquatic and coastal marine ecosystems.
 4          (3) The evidence is sufficient to infer a causal relationship between S deposition and
 5              increased Hg methylation in wetlands and aquatic environments.
 6          Subsequent to the previous review of the NOX secondary standard, a great deal of
 7    information on the contribution of atmospheric deposition associated with ambient NOX has
 8    become available. Chapter 3 of the REA provides a thorough assessment of the contribution of
 9    NOX to nitrogen deposition throughout the U. S., and the relative contributions of ambient NOX
10    and reduced forms of nitrogen. Staff concludes that based on that analysis, ambient NOX is a
11    significant component of atmospheric nitrogen deposition, even in areas with relatively high
12    rates of deposition of reduced nitrogen. In addition, staff initially concludes that atmospheric
13    deposition of oxidized nitrogen contributes significantly to total nitrogen loadings in nitrogen
14    sensitive ecosystems.
15          As discussed throughout the risk and exposure assessment document, there are several
16    key areas of risk that are associated with ambient concentrations of NOX and SOX.  In previous
17    reviews of the NOX and SOX secondary standards, the standards were designed to protect against
18    direct  exposure of plants to ambient concentrations of the pollutants.  A significant shift in
19    understanding of the effects of NOX and SOX has occurred since the last reviews, reflecting the
20    large amount of research that has been conducted on the effects of deposition of nitrogen and
21    sulfur  to ecosystems.  The most significant risks of adverse effects to public welfare are those
22    related to deposition of NOX and SOX to both terrestrial and aquatic ecosystems. These risks fall
23    into two categories: acidification and nutrient enrichment. These made up the emphasis of the
24    REA,  and are most relevant to evaluating the adequacy of the existing standards in protecting
25    public welfare from adverse ecological effects.
      March  2010                               132                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
       4.5.1   To what extent do the current NOX and SOX secondary standards provide
              protection from adverse effects associated with deposition of atmospheric
              NOX, and  SOX which results in acidification in sensitive aquatic and
              terrestrial ecosystems?
       The focus of the REA case studies was on determining whether deposition of sulfur and
oxidized nitrogen in locations where ambient NOX and SOX was at or below the current standards
was resulting in acidification and related effects. This review has focused on identifying
ecological indicators that can link atmospheric deposition to ecological effects associated with
acidification. NOX and SOX contribute to acidification in both aquatic and terrestrial ecosystems,
although the indicators of effects differ. While there are some geographic areas with both
terrestrial and aquatic ecosystems that are vulnerable to acidification, the case study areas do not
fully overlap. Figure 4-20 shows the locations of the case studies evaluated in the REA.
                                         0    250   500   750   1,000
                                               —^^™
                                               Kilometers
       Figure 4-20. National map highlighting the 9 case study areas evaluated in the
       REA.
     March  2010
                                           133
Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1          4.5.1.1   Aquatic Acidification
 2          Based on the case studies conducted for lakes in the Adirondacks and streams in
 3    Shenandoah National Park, staff concludes that there is significant risk to acid sensitive aquatic
 4    ecosystems at atmospheric concentrations of NOX and SOX at or below the current standards.
 5    This conclusion is based on application of the MAGIC model to estimate the effects of
 6    deposition at levels consistent with atmospheric NOX and SOX concentrations that are at or below
 7    the current standards. An important ecological indicator for aquatic acidification effects is ANC,
 8    measuring the acid buffering capacity of a waterbody, and the case study focused on evaluating
 9    whether locations were likely to be below critical values of ANC given deposition levels
10    associated with NOX and SOX concentrations that meet the current standards. In addition, the case
11    studies assessed the ecological effects and some of the known ecosystem services that are
12    associated with different levels  of ANC in order to associate the ecological indicator with
13    measures of public welfare that may be adversely affected by deposition levels consistent with
14    concentrations of NOX and SOX that meet the current standards.
15          Staff concludes that the  evidence and risk assessment support strongly a relationship
16    between atmospheric deposition of NOX and SOX and ANC, and that ANC is an excellent
17    indicator of aquatic acidification. Staff also  concludes that at levels of deposition associated with
18    NOX and SOX concentrations at  or below the current standards, ANC levels are expected to be
19    below benchmark values that are associated with significant losses in fish species richness (REA
20    Section 4)
21          Many locations in sensitive areas of the U.S.  have ANC  levels below benchmark levels
22    for ANC classified as severe, elevated, or moderate concern  (see Figure 2-1). The average
23    current ANC levels across 44 lakes in the Adirondack case study area is 62.1 (moderate
24    concern), however, 44 percent of lakes had deposition levels exceeding the critical load for an
25    ANC of 50, and 28 percent of lakes had deposition levels exceeding the critical load for an ANC
26    of 20 |ieq/L (REA Section 4.2.4.2). This indicates that almost half of the 44 lakes in the
27    Adirondacks case study area are at an elevated concern levels, and almost a third are at a  severe
28    concern level. These levels are associated with greatly reduced fish species diversity, and losses
29    in the health and reproductive capacity of remaining  populations. Based on assessments of the
30    relationship between number offish species and ANC level in both the Adirondacks and
31    Shenandoah areas, the number of fish species is decreased by over half at an ANC level of 20

      March 2010                               134                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    neq/L relative to an ANC level at 100 |ieq/L (REA Figure 4.2-1). At levels below 20 |ieq/L,
 2    populations of sensitive species, such as brook trout, may decline significantly during episodic
 3    acidification events. When extrapolated to the full population of lakes in the Adirondacks area
 4    using weights based on the EMAP probability survey (REA 4.2.6.1), 36 percent of lakes
 5    exceeded the critical load for an ANC of 50 jieq/L and 13 percent of lakes exceeded the critical
 6    load for an ANC of 20 jieq/L.
 7           Many streams in the Shenandoah case study area also have levels of deposition that are
 8    associated with ANC levels classified as severe, elevated, or moderate concern. The average
 9    ANC under recent conditions for the 60 streams evaluated in the Shenandoah case study area is
10    57.9, indicating moderate concern. However, 85 percent of streams had recent deposition
11    exceeding the critical load for an ANC of 50 |ieq/L, and 72 percent exceeded the critical load for
12    an ANC of 20 |ieq/L. As with the Adirondacks area, this suggests that significant numbers of
13    sensitive streams in the Shenandoah area are at risk of adverse impacts on fish populations under
14    recent conditions. Many other streams in the Shenandoah area are likely to  experience conditions
15    of elevated to severe concern based on the prevalence in the area  of bedrock geology associated
16    with increased sensitivity to acidification suggesting that effects due to stream acidification could
17    be widespread in the Shenandoah area (REA 4.2.6.2).
18           The ISA notes that large portions of the Eastern U.S. are acid sensitive, and that current
19    deposition levels exceed those that would allow recovery of the most acid sensitive lakes in the
20    Adirondacks (ISA ES). In addition, because of past loadings, areas of the Shenandoah are
21    sensitive to current deposition levels (ISA ES). Much of the West is naturally less sensitive to
22    acidification, and as such, less focus is placed on the adequacy of the existing standards in these
23    areas, with the exception of the mountainous areas of the West, which experience episodic
24    acidification due to deposition.
25           While most (99 percent) of stream kilometers in the U.S. are not chronically acidified
26    under current conditions, a recent survey found sensitive streams  in many locations in the U.S.,
27    including the Appalachian mountains, the Coastal Plain, and the Mountainous West (ISA
28    Section 4.2.2.3). In these  sensitive areas, between 1 and 6 percent of stream kilometers are
29    chronically  acidified.
30           The ISA notes that "consideration of episodic acidification greatly increases the extent
31    and degree of estimated effects for acidifying deposition on surface waters." (ISA Section

      March 2010                               135                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    3.2.1.6) Some studies show that the number of lakes that could be classified as acidified based on
 2    episodic acidification is 2 to 3 times the number of lakes classified as acidified based on chronic
 3    ANC. These episodic acidification events can have long term effects on fish populations (ISA
 4    Section 3.2.1.6). Under recent conditions, episodic acidification has been observed in locations
 5    in the Eastern U.S. and in the Mountainous Western U.S. (ISA Section 3.2.1.6).
 6           It can therefore be concluded that recent levels of NOX and SOX are associated with
 7    deposition that leads to ANC values below benchmark values known to cause ecological harm in
 8    sensitive  aquatic systems, including lakes and streams in multiple areas of the U.S. These
 9    changes are known to have impacts on ecosystem services such as reductions in recreational
10    fishing. While other ecosystem services (e.g. habitat provisioning, subsistence fishing, and
11    biological control  as well as many others) are potentially affected by reductions in ANC,
12    confidence in the specific translation of ANC values to these additional ecosystem services is
13    much lower.

14           4.5.1.2  Terrestrial Acidification
15           Based on the case studies on sugar maple and red spruce habitat, staff concludes that
16    there is significant risk to terrestrial ecosystems from acidification at atmospheric concentrations
17    of NOX and SOX at or below the current standards. This conclusion is based on application of the
18    simple mass balance model to deposition levels associated with NOX and SOX concentrations at
19    or below  the current standards. The ecological indicator  selected for terrestrial acidification is the
20    base cation to aluminum ratio (BC:A1), which has been linked to tree health and growth. The
21    results of the REA strongly support a relationship between atmospheric deposition of NOX and
22    SOX and BC:A1, and that BC:A1 is a good indicator of terrestrial acidification. At levels of
23    deposition associated with NOX and SOX concentrations at or below the current standards, BC: Al
24    levels are expected to be below benchmark values that are associated with significant losses in
25    tree health and growth. Such degradation of terrestrial ecosystems could affect ecosystem
26    services such as habitat provisioning, endangered species, goods production (timber, syrup, etc.)
27    and many others.
28           Many locations in sensitive areas  of the U.S. have Bc/Al levels below benchmark levels
29    classified as providing low to intermediate levels of protection to tree health. At a Bc/Al ratio of
30    1.2 (intermediate level of protection), red spruce growth can be reduced by 20 percent. At a
      March 2010                                136                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    Bc/Al ratio of 0.6 (low level of protection), sugar maple growth can be reduced by 20 percent.
 2    The REA did not evaluate broad sensitive regions. However, in the sugar maple case study area
 3    (Kane Experimental Forest), recent deposition levels are associated with a Bc/Al ratio below 1.2,
 4    indicating between intermediate and low level of protection, which would indicate the potential
 5    for a greater than 20 percent reduction in growth. In the red spruce case study area (Hubbard
 6    Brook Experimental Forest), recent deposition levels are associated with a Bc/Al ratio slightly
 7    above 1.2, indicating slightly better than an intermediate level of protection (REA Section
 8    4.3.5.1)
 9           Over the full range of sugar maple, 12 percent of evaluated forest plots exceeded the
10    critical load for a Bc/AL ratio of 1.2, and 3 percent exceeded the critical load for a Bc/Al ratio of
11    0.6. However, there was large variability across states. In New Jersey, 67 percent of plots
12    exceeded the critical load for a Bc/Al ratio of 1.2, while in several states on the outskirts of the
13    range for sugar maple, e.g. Arkansas, Illinois, no plots exceeded the critical load for a Bc/Al ratio
14    of 1.2. For red spruce, overall 5 percent of plots exceeded the critical  load for a Bc/Al ratio of
15    1.2, and 3 percent exceeded the critical load for a Bc/Al ratio of 0.6. In the major red spruce
16    producing states  (Maine, New Hampshire, and Vermont), critical loads for a Bc/AL ratio of 1.2
17    were exceeded in 0.5, 38, and 6 percent of plots.
18           The ISA reported one study that estimated 15 percent of U.S.  forest ecosystems exceeded
19    the critical loads  for acidity for N and S deposition by >250 eq/ha/year under current conditions
20    (ISA Section 4.2.1.3). Staff believes that this represents a significant  portion of sensitive
21    terrestrial ecosystems.
22           It can therefore be concluded that recent levels of NOX and SOX are associated with
23    deposition that leads to BC:A1 values below benchmark values that cause ecological harm in
24    some sensitive terrestrial ecosystems. While effects are more widespread for sugar maple, there
25    are locations with low to intermediate levels of protection from effects on both sugar maple and
26    red spruce.  While there are many other ecosystem services, including timber production, natural
27    habitat provision, and regulation of water, climate,  and erosion, potentially affected by
28    reductions in BC:A1,  linkages of BC:A1 values to these additional ecosystem services is on the
29    whole not well understood.
      March 2010                                137                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1           4.5.2  To what extent does the current NOX secondary standard provide protection
 2                 from adverse effects associated with deposition of atmospheric NOX, which
 3                 results in nutrient enrichment effects in sensitive aquatic and terrestrial
 4                 ecosystems?
 5           Nutrient enrichment effects are due to nitrogen loadings from both atmospheric and non-
 6    atmospheric sources. Evaluation of nutrient enrichment effects requires an understanding that
 7    nutrient inputs are essential to ecosystem health. The specific long term levels of nutrients in a
 8    system affect the types of species that occur over long periods of time. Short term additions of
 9    nutrients can affect species competition, and even small additions of nitrogen in areas that are
10    traditionally nutrient poor can have significant impacts. In certain limited situations, additions of
11    nitrogen can increase rates of growth,  and these increases can have short term benefits in certain
12    managed ecosystems. As noted earlier, this review of the standards is focused on unmanaged
13    ecosystems. As a result, in assessing adequacy of the current standards, we are focusing on the
14    adverse effects of nutrient enrichment in unmanaged ecosystems. However, the following
15    discussion provides a brief assessment of effects in managed ecosystems.
16           Impacts of nutrient enrichment in managed ecosystems may be positive or negative
17    depending on the levels of nutrients from other sources in those areas. Positive effects can occur
18    when crops or commercial forests are  not receiving enough nitrogen nutrients. Nutrients
19    deposited on crops from atmospheric sources are often referred to as passive fertilization.
20    Nitrogen is a fundamental nutrient for primary production in both managed and unmanaged
21    ecosystems. Most productive agricultural systems require external sources of nitrogen in order to
22    satisfy nutrient requirements. Nitrogen uptake by crops varies, but typical requirements for wheat
23    and corn are approximately 150 kg/ha/yr and 300 kg/ha/yr, respectively (NAPAP, 1990). These
24    rates compare to estimated rates of passive nitrogen fertilization in the range of 0 to  5.5 kg/ha/yr
25    (NAPAP, 1991).
26           Information on the effects of changes in passive nitrogen deposition on forestlands and
27    other terrestrial ecosystems is very limited. The multiplicity of factors affecting forests, including
28    other potential stressors such as ozone, and limiting factors such as moisture and other nutrients,
29    confound assessments of marginal changes in any one stressor or nutrient in forest ecosystems.
30    The ISA notes that only a fraction of the deposited nitrogen is taken up by the forests, most of
31    the nitrogen is retained in the soils (ISA 3.3.2.1). In addition, the ISA indicates that forest

      March 2010                               138                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    management practices can significantly affect the nitrogen cycling within a forest ecosystem, and
 2    as such, the response of managed forests to NOX deposition will be variable depending on the
 3    forest management practices employed in a given forest ecosystem (ISA Annex C C.6.3)
 4    Increases in the availability of nitrogen in N-limited forests via atmospheric deposition could
 5    increase forest production over large non-managed areas, but the evidence is mixed, with some
 6    studies showing increased production  and other showing little effect on wood production (ISA
 7    3.3.9). Because leaching of nitrate can promote cation losses, which in some cases create nutrient
 8    imbalances, slower growth and lessened disease and freezing tolerances for forest trees, the net
 9    effect of increased N on forests in the  U.S. is uncertain  (ISA 3.3.9).
10           In managed agricultural ecosystems, nitrogen inputs from atmospheric NOX comprise a
11    small fraction (less than 3 percent) of  total nitrogen inputs, which include commercially applied
12    fertilizers as well as applications of composted manure. And because of the temporal and spatial
13    variability in atmospheric deposition of NOX, it is unlikely  that farmers would alter their
14    fertilization decisions based on expected nitrogen inputs from NOX. And, in some locations,
15    farmers need less nitrogen inputs due  to production of excess nitrogen through livestock. In some
16    locations, nitrogen production through livestock waste exceeds the absorptive capacity of the
17    surrounding land, and as such, excess  nitrogen from deposition of NOX in those locations reduces
18    the capacity of the system to dispose of excess nitrogen, potentially increasing the costs of waste
19    management from livestock operations (Letson and Gollehon, 1996). A USD A Economic
20    Research Service report found that in  1997, 68 counties with high levels of confined livestock
21    production had manure nitrogen levels that exceed  the assimilative capacity of all the county's
22    crop and pasture land (Gollehon et al,  2001). In those locations, additional nitrogen inputs from
23    NOX deposition will result in excess nitrogen, leading to nitrogen leaching and associated effects.

24    4.5.3   Aquatic Nutrient Enrichment
25           The REA case studies focused on coastal estuaries and revealed that while current
26    ambient loadings of atmospheric NOX are contributing to the overall deposit!onal loading of
27    coastal estuaries, other non-atmospheric sources are contributing in far greater amounts in total,
28    although atmospheric contributions are as large as some other individual source types. The
29    ability of current data and models to characterize the incremental adverse impacts of nitrogen
30    deposition is limited, both by the available ecological indicators, and by the inability to attribute
      March 2010                                139                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    specific effects to atmospheric sources of nitrogen. The REA case studies used as the ecological
 2    indicator for aquatic nutrient enrichment, an index of eutrophication known as the Assessment of
 3    Estuarine Trophic Status Eutrophication Index (ASSETS El). This index is a six level index
 4    characterizing overall eutrophication risk in a waterbody. This indictor is not sensitive to
 5    relatively large changes in nitrogen deposition. In addition, this type of indicator does not reflect
 6    the impact of nitrogen deposition in conjunction with other sources of nitrogen.
 7           For example, if NOX deposition is contributing nine tenths of the nitrogen loading
 8    required to move a waterbody from an ASSETS El category of "moderate" to a category of
 9    "poor", zeroing out NOX deposition will  have no impact on the ASSETS El value. However, if
10    an area were to decide to put in place decreases in nitrogen loadings to move that waterbody
11    from "poor" to "moderate," the area would have to reduce the full amount of the loadings
12    through other sources if atmospheric deposition were not considered. Thus, the adverse impact of
13    atmospheric nitrogen is in its contribution to the overall loading, and reductions in NOX will
14    decrease the amount of reductions from other sources of nitrogen loadings that would be required
15    to move from a lower ASSETS El category to a higher category. NOX deposition can also be
16    characterized as reducing the risk of a waterbody moving from a higher ASSETS El category to
17    a lower category, by reducing the vulnerability of that waterbody to increased loadings from
18    non-atmospheric sources.
19           Based on the above considerations, staff preliminarily concludes that the ASSETS El is
20    not an appropriate ecological indicator for estuarine aquatic eutrophication. Staff further
21    concludes that additional analysis is required to develop an appropriate indicator for determining
22    the appropriate levels of protection from N nutrient enrichment effects in estuaries related to
23    deposition of NOX. As a result, staff is unable to make a determination as to the adequacy of the
24    existing secondary NOX standard in protecting public welfare from N nutrient enrichment effects
25    in estuarine aquatic ecosystems.
26           Additionally, nitrogen deposition can alter species  composition and cause eutrophication
27    in freshwater systems. In the Rocky Mountains, for example, deposition loads of 1.5 to 2
28    kg/ha/yr which are well within current ambient levels are known to cause changes in species
29    composition in diatom communities indicating impaired water quality (ISA Section 3.3.5.3). It
30    then seems apparent then that the existing secondary standard for NOX does not protect such
31    ecosystems and their resulting services from impairment.

      March  2010                               140                Draft - Do Not Quote  or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    4.5.4  Terrestrial Nutrient Enrichment
 2          The scientific literature has many examples of the deleterious effects caused by excessive
 3    nitrogen loadings to terrestrial systems. Several studies have set benchmark values for levels of
 4    N deposition at which scientifically adverse effects are known to occur. These benchmarks are
 5    discussed more thoroughly in Chapter 5 of the REA. Large areas of the country appear to be
 6    experiencing deposition above these benchmarks for example, Fenn et al. (2008) found that at
 7    3.1 kg N/ha/yr, the community of lichens begins to change from acidophytic to tolerant species;
 8    at 5.2 kg N/ha/yr, the typical dominance by acidophytic species no longer occurs; and at  10.2 kg
 9    N/ha/yr, acidophytic lichens are totally lost from the community. Additional studies in the
10    Colorado Front Range of the Rocky Mountain National Park support these findings and are
11    summarized in Chapter 6.0 of the Risk and Exposure Assessment. These three values (3.1, 5.2,
12    and 10.2 kg/ha/yr) are one set of ecologically meaningful benchmarks for the mixed conifer
13    forest (MCF) of the pacific coast regions. Nearly all of the known sensitive communities receive
14    total nitrogen deposition levels above the 3.1 N kg/ha/yr ecological benchmark according to
15    the 12 km, 2002 CMAQ/NADP data, with the exception of the easternmost Sierra Nevadas.
16    MCFs in the southern portion of the  Sierra Nevada forests and nearly all MCF communities in
17    the San Bernardino forests receive total nitrogen deposition levels above the 5.2 N kg/ha/yr
18    ecological benchmark.
19          Coastal Sage Scrub communities (CSS) are also known to be sensitive to community
20    shifts caused by excess nitrogen loadings. Wood et al. (2006) investigated the amount of nitrogen
21    utilized by healthy and degraded CSS systems. In healthy stands, the authors estimated that 3.3
22    kg N/ha/yr was used for CSS plant growth (Wood et al., 2006). It is assumed that 3.3 kg N/ha/yr
23    is near the point where nitrogen is no longer limiting in the CSS community. Therefore, this
24    amount can be considered an ecological benchmark for the CSS community. The majority of the
25    known CSS range is currently receiving deposition in excess of this benchmark. Thus, staff
26    concludes that recent conditions where NOX ambient concentrations are at or below the current
27    NOX secondary standards are not adequate to protect against anticipated adverse impacts  from N
28    nutrient enrichment in sensitive ecosystems (systems where N is not limiting).
      March  2010                               141                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1   4.6    TO WHAT EXTENT DO THE CURRENT NOX AND/OR SOX
 2          SECONDARY STANDARDS PROVIDE PROTECTION FROM
 3          OTHER ECOLOGICAL EFFECTS (E.G., MERCURY
 4          METHYLATION) ASSOCIATED WITH THE DEPOSITION OF
 5          ATMOSPHERIC NOX, AND/OR SOX?
 6          It is stated in the ISA (ISA Sections 3.4.1 and 4.5) that mercury is a highly neurotoxic
 7   contaminant that enters the food web as a methylated compound, methylmercury. Mercury is
 8   principally methylated by sulfur-reducing bacteria  and can be taken up by microorganisms,
 9   zooplankton and macroinvertebrates. The contaminant is concentrated in higher trophic levels,
10   including fish eaten by humans. Experimental evidence has established that only inconsequential
11   amounts of methylmercury can be produced in the  absence of sulfate. Once methylmercury is
12   present, other variables influence how much accumulates in fish, but elevated mercury levels in
13   fish can only occur where substantial amounts of methylmercury are present. Current evidence
14   indicates that in watersheds where mercury is present, increased SOX deposition very likely
15   results in additional production of methylmercury which leads to greater accumulation of MeHg
16   concentrations in fish (Munthe et al, 2007; Drevnick et al., 2007).
17          The production of meaningful amounts of methylmercury (MeHg) requires the presence
18   of SO42" and mercury, and where mercury is present, increased availability of SO42" results in
19   increased production of MeHg. There is increasing evidence on the relationship between sulfur
20   deposition and increased methylation of mercury in aquatic environments; this effect occurs only
21   where other factors are present at levels within a range to allow methylation. The production of
22   methylmercury requires the presence of sulfate and mercury, but the amount of methylmercury
23   produced varies with oxygen content, temperature, pH, and supply of labile organic carbon (ISA
24   Section 3.4). In watersheds where changes in sulfate deposition did not produce an effect, one or
25   several of those interacting factors were not in the range required for meaningful methylation to
26   occur (ISA Section 3.4). Watersheds with conditions known to be  conducive to mercury
27   methylation can be found in the northeastern United States and southeastern Canada. The
28   relationship between sulfur and methylmercury production is addressed qualitatively in Chapter
29   6 of the Risk and Exposure Assessment.
     March 2010                              142               Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1          With respect to sulfur deposition and mercury methylation, the final ISA determined: The
 2    evidence is sufficient to infer a causal relationship between sulfur deposition and increased
 3    mercury methylation in wetlands and aquatic environments. However, staff did not conduct a
 4    quantitative assessment of the risks associated with increased mercury methylation under current
 5    conditions. As such, staff are unable to make a determination as to the adequacy of the existing
 6    SC>2 standards in protecting against welfare effects associated with increased mercury
 7    methylation.

 8    4.7    REFERENCES
 9    BJ. Finlayson-Pitts and J.N. Pitts, 2000, Chemistry of the Upper and Lower Troposhere,
10          Academic Press, San Diego, CA

11    Drevnick, P.E., D.E. Canfield, P.R. Gorski, A.L.C.  Shinneman, D.R.  Engstrom, D.C.G. Muir,
12          G.R. Smith, PJ. Garrison, L.B. Cleckner, J.P. Hurley, R.B. Noble, R.R. Otter, and J.T.
13          Oris. 2007. Deposition and cycling of sulfur controls mercury accumulation in Isle
14          Royale fish. Environmental Science and Technology ¥7(21):7266-7272.

15    Fenn, M.E., S. Jovan, F. Yuan, L. Geiser, T. Meixner, and B.S. Gimeno. 2008. Empirical and
16          simulated critical loads for nitrogen deposition in California mixed conifer forests.
17          Environmental Pollution 755(3 ): 492-511.

18    Munthe, 1, R.A. Bodaly, B.A. Branfireun, C.T. Driscoll, C.C. Gilmour, R. Harris, M. Horvat, M.
19          Lucotte, and O. Malm. 2007. Recovery of mercury-contaminated fisheries. Ambio 36:33-
20          44.

21    Scheffe, R.D., P. A. Solomon, R. Husar, T. Hanley, M. Schmidt, M. Koerber, M. Gilroy, J.
22          Hemby, N. Watkins, M. Papp, J. Rice, J. Tikvart andR. Valentinetti, The National
23          Ambient Air Monitoring Strategy: Rethinking the Role of National Networks, JAWMA,
24          ISSN: 1047-3289 J. Air & Waste Manage. Assoc. 2009, 59:579-590 DOI: 10.3155/1047-
25          3289.59.5.579

26    U.S. EPA (Environmental Protection Agency). 1982. Review of the National Ambient Air Quality
27          Standards for Sulfur Oxides: Assessment of Scientific and Technical Information.
      March  2010                               143                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1          OAQPS Staff Paper. EPA-450/5-82-007. U.S. Environmental Protection Agency, Office
 2          of Air Quality Planning and  Standards, Research Triangle Park, NC.

 3   U.S. EPA (Environmental Protection Agency). 1995. Review of the National Ambient Air Quality
 4          Standards for Nitrogen Dioxide: Assessment of Scientific and Technical Information.
 5          OAQPS Staff Paper. EPA-452/R-95-005. U.S. Environmental Protection Agency, Office
 6          of Air Quality Planning and  Standards, Research Triangle Park, NC. September.

 7   U.S. EPA (Environmental Protection Agency). 2008. Integrated Science Assessment (ISA) for
 8          Oxides of Nitrogen and Sulfur-Ecological Criteria (Final Report). EPA/600/R-
 9          08/082F. U.S. Environmental Protection Agency, National Center for Environmental
10          Assessment-RTF Division, Office of Research and Development, Research Triangle
11          Park, NC. Available at http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=201485.

12   Wood, Y., T. Meixner, PJ. Shouse,  and E.B. Allen. 2006. Altered Ecohydrologic response
13          drives native shrub loss under conditions of elevated N-deposition. Journal of
14          Environmental Quality 35:76-92.

15
     March 2010                              144                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1      5.     CONCEPTUAL DESIGN OF AN ECOLOGICALLY RELEVANT
 2                           MULTI-POLLUTANT STANDARD
 3          The objective of this chapter is to describe the conceptual design for a national ambient
 4    air quality standard that links ecological indicators of concern to ambient air indicators of NOX
 5    and SOX. In Chapter 4 of this policy assessment, the limitations of the design of the current
 6    secondary standards are described as they apply to protection of sensitive  ecosystems. The
 7    conceptual design described in this chapter addresses those limitations. The overall concept for
 8    the standards starts  by recognizing that the fundamental welfare effects associated with ambient
 9    NOX and SOX occur through the process of deposition to sensitive ecosystems. As detailed in
10    Chapter 4, previous NOX and SOX NAAQS reviews only considered effects to vegetation via
11    stomatal exposure.  There is now sufficient data to link atmospheric concentrations to adverse
12    effects in ecosystems that are caused by exposure via deposition to soils and surface waters.
13    Deposition is a direct consequence of atmospheric concentration; however it is also modified by
14    factors that vary across the landscape (e.g. elevation and groundcover). Likewise, ecological
15    response to deposition can vary according to ecosystem sensitivity and the ecological indicator
16    of concern. This is the first time a secondary standard for deposition effects related to NOX and
17    SOX has been developed; therefore the conceptual design of a potential standard is described here
18    prior to the specific details on the indicator, level, form and averaging time for such a potential
19    standard that are presented in chapter 6.

20    5.1   COMPONENTS OF THE DESIGN
21          There are four main components to the conceptual design of the standard: atmospheric
22    and ecological indicators, deposition metrics, functions that relate indicators to deposition
23    metrics and factors  that modify the functions. These components of the design are illustrated in
24    Figure 5-1. The squares represent indicators. Ecological indicators are chemical or biological
25    components of the ecosystem that can be linked to N and S deposition based on scientific
26    evidence. Air quality indicators are the chemical species of the criteria air pollutants that best
27    represent the atmospheric pollutants that cause ecological harm in the criteria pollutant
28    categories of oxides of nitrogen and oxides of sulfur. Triangles indicate functions in which two
29    variables are related. The ecological effect function is the relationship between the ecological
      March  2010                               145                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
 1
 2
 3
 4
 5
 6
 9
10
11
12
13
14
15
16
17
18
19
20
21
indicator and deposition over a range of values. The atmospheric deposition transformation
function is the relationship between deposition and the atmospheric concentration of an air
quality indicator. The circles represent factors which will modify the functions. Modifying
factors can vary across the landscape.  The spatial heterogeneity of modifying factors can be
challenging to characterize, and therefore in some cases we present multiple options for how to
incorporate them into the design.
            Ecological
             Indicator
                                                              Variable/Fixed
                                                                Modifying
                                                                 Factors
                      Variable/Fixed
                       Modifying
                        Factors
                                        Deposition
                                          Metric
 Atmospheric
  Deposition
Transformation
   Function
 Ecological
Response to
 Deposition
  Function
Air Quality
Indicator(s)
       Fig 5-1. Schematic diagram of the conceptual design of the standard.
       5.1.1   For which effects is there sufficient information to support setting standards?
       After review of the ISA and REA, CAS AC concluded that aquatic acidification should be
the focus for developing a multi-pollutant standard, based on the quantity and quality of data.
CASAC also recommended that, in addition to aquatic acidification, the EPA should consider
multiple ecological  indicators and made the following statement in their letter to the EPA on
August 28, 2009:
        ".. .the Panel finds the information  in the current REA sufficient to inform setting
       separate standards for terrestrial acidification, eutrophication of western alpine
       lakes and terrestrial nutrient enrichment. However, the Panel believes that setting
       a standard for coastal nutrient enrichment would be difficult because of the
       substantial inputs of non-atmospheric sources of N to these systems."
       The following sections describe the conceptual design for standards based on aquatic
acidification, terrestrial  acidification, eutrophication of high  elevation western lakes and
      March 2010
                                            146
      Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1   terrestrial nutrient enrichment. The focus of the first draft will be on aquatic acidification, but
 2   this general conceptual framework will apply to a broader set of potential endpoints.

 3   5.2   ECOLOGICAL COMPONENTS OF THE STANDARD: AQUATIC
 4          ACIDIFICATION
 5          Details of the conceptual design of the NOX and SOX NAAQS based on aquatic
 6   acidification effects are presented in this section. A summary of our over all approach is given
 7   here to help provide context and support for the more detailed discussions that follow.
 8          At the catchment scale, ambient NOy and SOX add to the total deposition of N and S that
 9   lead to aquatic acidification. NHX is often  another big component of the total N deposition. The
10   load of deposition that causes a desired level of ANC will vary  depending on the characteristics
11   of the ecosystem. The level of ANC is tied to the degree of biological harm to the system from
12   aquatic acidification.
13          The components of the standard are modified for application to aquatic acidification and
14   presented in Fig 5-2. The bidirectional arrows emphasize that the order in which one considers
15   the links between ANC and atmospheric concentrations of NOX and SOX is conceptually
16   important to the standard design. Moreover, different questions may be answered by working
17   through Fig 5-2 from the left to the right versus the right to the  left. For example, working from
18   left to right, when a level of ANC is specified the deposition loadings of N and S that would
19   cause the specified level of ANC can be calculated; in essence this would be a critical load for a
20   specified ANC limit. A comparison between the total amount of deposited N and S to the critical
21   load would determine whether the specified level of ANC is achieved for a catchment. Let's now
22   work through the equation from right to left. If the amount of N and S deposited to a given
23   catchment is known, you could calculate the level of ANC that  would result. The calculated
24   ANC could then be compared to a benchmark value of ANC. In both of these approaches the
25   amount of reduced N would be subtracted from the total N deposition to calculate deposition
26   from NOy.  The  atmospheric concentrations of NOy and SOX would be calculated from the
27   deposition  of NOy and S according to the methods presented in  section 5.4. To determine the
28   appropriate conceptual design from the ecological components  of the standard, the analysis from
29   the REA is evaluated in which critical loads were calculated for a target value of ANC, thereby
30   working from left to right on Fig 5-2.

     March 2010                               147               Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
 1

 2
 3
 4
 5
 6
 9

10

11

12

13

14

15

16

17

18
             Level of
           ANC related to
          biological effects
 Acidification
  Model that
relates ANC to
  deposition
at catchment-
   scale
                                        Deposition
                                        Loading of
                                          N+S
                                          which
                                        represent
                                         national
                                          scale
                                        landscape
                                        categories
<=>/        \<=>|
        Atmospheric
        Deposition
       Transformation
         Function
Concentration
    of the
  Air Quality
  Indicator(s)
            Relationship between the amount of
            deposition and the effect on the selected
            ecological indicator, ANC (described in 5.2)
                                                      Relationship between the amount of
                                                      deposition and the concentration ofNOx
                                                      and SOx (described in 5.4)
       Fig 5-2. Schematic diagram of the conceptual design of the standard based on
       aquatic acidification. From left to right, if a desired level of ANC is known, then
       the concentration of the atmospheric indicators that will cause that level may be
       calculated. From right to left, if the if the concentration of the air quality
       indicators are known than the ANC that will be caused may be calculated.

       The secondary NAAQS would apply to all areas of the country. It is not practical to

evaluate each catchment individually, and that is not the appropriate approach for a national

standard. Here, EPA staff proposes to categorize landscape features nationally, such that within a

category there are generally similar characteristics as far as the relationship of total deposited N

and S to the ANC. Every part of the country would be assigned into one of these bins/ landscape

categories.

       The secondary NAAQS would be based on a judgment as to a specified level of ANC.

For each national acid-sensitivity bin/ landscape category there would be a range of critical loads

for a specified ANC limit from the individual catchments within the total population aggregated

to an acid-sensitivity category. Given that, the EPA would develop a  deposition metric and

associated tradeoff curve that represented the  percentage of the catchments that would achieve

the ANC (DLo/oECo). Therefore a judgment would also need to be made to determine the
      March 2010
                                             148
             Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    percentage of ecosystems that would be targeted to achieve a specified ANC level that applies to
 2    a bin/category.
 3           The following discussions in this section focus on the ecological components of the
 4    standard (ecological indicator, the deposition metric, the ecological response function and its
 5    modifiers). Questions that are relevant to the design of the standard are used to organize these
 6    discussions. The first series of questions (section 5.2.1) considers information presented in the
 7    ISA and REA relevant to the conceptual design, while the second series of questions (section
 8    5.2.2) presents the proposed conceptual design in more detail with an example calculation based
 9    on the Adirondacks case study presented in section 5.5.

10           5.2.1   Conceptual design considerations from the ISA and REA
11           This section presents discussion of the ecological components of the design based on
12    information in the ISA and REA. The information presented here is considered in the
13    development of the design options that are proposed (section 5.2.2).

14           5.2.1.1  Does the available information provide support for the use of ecological
15                   indicators to characterize the responses of aquatic ecosystems to nitrogen and
16                   sulfur deposition ?
17           Ecological indicators of acidification in aquatic ecosystems can be chemical or
18    biological components of the ecosystem that are demonstrated to be altered by the acidifying
19    effects of N and S deposition based on scientific evidence. A desirable ecological indicator for
20    aquatic acidification will be one that is measurable or estimable, linked causally to deposition of
21    N and S, and linked causally to ecological effects known or anticipated to adversely affect public
22    welfare.
23           As summarized in Chapter 2, aquatic acidification is indicated by changes in the surface
24    water chemistry of ecosystems. In turn, the alteration of surface water chemistry has been linked
25    to negative effects on the biotic integrity of freshwater ecosystems. There are a suite of chemical
26    indicators that can be used to assess the effects of acidifying deposition on lake or stream acid-
27    base chemistry. These indicators include acid neutralizing capacity (ANC), surface water pH and
28    concentrations of SC>42", NCV, Al, and Ca2+; the sum of base cations; and the recently developed
29    base cation  surplus. ANC is the most widely used chemical indicator of acid sensitivity and has
30    been found in various studies to be the best single indicator of the biological response and health

      March 2010                                149                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    of aquatic communities in acid-sensitive systems (Lien et al., 1992; Sullivan et al., 2006). The
 2    utility of the ANC criterion lies in the association between ANC and the surface water
 3    constituents that directly  contribute to or ameliorate acidity-related stress, in particular pH, Ca2+,
 4    and Al. ANC is also used because it integrates overall acid status (ISA 3.2.3 and REA 5.2.1) and
 5    the acid-related stress for biota that occupies the water that can be directly related to biological
 6    impairment, specifically the number offish species (ISA 3.2.3).
 7          EPA staff thus concludes that the available information provides support for the use of
 8    ecological indicators to characterize the responses of aquatic ecosystems to nitrogen and sulfur
 9    deposition, and that ANC is the most supportable indicator.

10          5.2.1.2   Does the available information provide support for the development of a
11                  function that relates total nitrogen and sulfur deposition to ecological
12                   indicators?
13          There is evidence to support the link between deposition of N and  S, water chemistry and
14    biota. Atmospheric deposition of NOX and SOX causes aquatic acidification through the input of
15    acid anions (e.g. N(V and SC>42") The anions are deposited either directly to the aquatic
16    ecosystem, or indirectly via terrestrial ecosystems. In other words, when the anions are mobile in
17    the terrestrial soil, they can leach into adjacent waterbodies. Acidification of ecosystems is
18    reflected in a robust relationship between ANC of water and the deposition of NOX and SOX.
19          In the REA, the relationship between deposition and ANC was investigated using models
20    of ecosystem acidification (REA Chapter 4 and REA Appendix 4). These models characterize
21    the relationship between deposition N and S and the ability of an ecosystem to counterbalance or
22    buffer the deposition. The utility of the ecosystem acidification models is  for simulating a variety
23    of water and soil acidification responses at the laboratory, plot, hillslope, and catchment scales.
24    For example, the ANC value  caused by the current amount of deposition could be calculated, or,
25    the level of deposition that causes a specified level of an ecosystem endpoint could be calculated
26    (i.e. a critical load for ANC=50) (ISA appendix A).
27          The models used in the REA were the Steady State Water Chemistry model (SSWC), the
28    First-order Acid Balance model (FAB) and the Model of Acidification of  Groundwater in
29    Catchment (MAGIC). The SSWC and FAB models were used to calculate critical loads for
30    specified ANC levels in the case study areas. MAGIC was used to develop weathering rates that
      March  2010                               150                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
 1
 2
 3
 4
 5
 6
 7
were needed for the Shenandoah critical loads calculation and the F-factor was used for
weathering rates in the in the Adirondacks. MAGIC was also used to show long-term trends
between anthropogenic N and S deposition on ANC dating back to pre-industrial times. It is
important to note that acidification models are data intensive. Water chemistry data from the
TIME and LTM programs, which are part of the Environmental Monitoring and Assessment
Program (EMAP), were input to the acidifcation models. An abbreviated summary of
acidification models and data inputs is given in Table 5.2-1, a complete list is in Appendix A.

Table 5-1. Illustration of how selected models and water chemistry data were used to calculate
critical loads in the REA.

Adirondack
Shenandoah
Weathering rate
as input to CL
model
F-factor
MAGIC
Water chemistry data
input to CL model
EMAP
EMAP
CL calculation:
single value
sswc
sswc
CL calculation:
critical load
function
FAB
FAB
 9          In summary, the EPA staff concludes that the available information supports using the
10    acidification models to characterize the relationship between total nitrogen and sulfur deposition
11    and the ANC ecological indicator.

12          5.2.1.3   Does a quantified relationship exist between the level of a relevant ecological
13                   indicator to an amount of nitrogen and sulfur deposition?
14          A quantified relationship exists between the level of ANC and nitrogen and sulfur
15    deposition. This relationship was analyzed to determine current risk for two case study areas, the
16    Adirondacks and Shenandoahs, in the PvEA using a time series analysis and a critical load
17    approach. The time series analysis was conducted using MAGIC and recent monitoring data. The
18    critical loads analysis used water chemistry data from the Temporally Integrated Monitoring of
19    Ecosystems (TIME) program and Long-term Monitoring (LTM) to calculate critical loads with the
20    SSWC and FAB models.
21          Long-term trends in  surface water nitrate, sulfate and ANC were modeled using MAGIC
22    for the two case study areas. This data was used to compare current  surface water conditions
23    (2006) with preindustrial conditions (i.e. preacidification or 1860). The results showed a
      March  2010
                                           151
Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    dramatic increase in the number of acidified lakes, characterized as a decrease in ANC levels,
 2    since the onset of anthropogenic N and S deposition (REA Appendix 4 Section 5)
 3          More recent trends in ANC, over the time period from 1990 to 2006, were assessed using
 4    monitoring data collected at the two case study areas. In both case study areas, nitrate and sulfate
 5    deposition decreased over this time period. In the Adirondacks, this corresponded to a decreased
 6    concentration of nitrate and sulfate in the surface waters and an increase in ANC (REA 4.2.4.2).
 7    In the Shenandoahs, there was a slight decrease in nitrate and sulfate concentration in surface
 8    waters corresponding to modest increase in ANC from 50 ueq/L in 1990 to 67 ueq/L in 2006
 9    (REA 4.2.4.3 and REA Appendix 4 Section 3.4)
10          A critical load for ANC is the amount (or load per year) of N and S deposition above
11    which a selected level of ANC will be exceeded for individual water bodies. In the REA case
12    study analyses, critical loads and their exceedances were calculated for four values of ANC (i.e.,
13    ANC of 0, 20, 50,  and  100  ueq/L) for 169 lakes in the Adirondacks and 60 streams in the
14    Shenandoahs. Those four ANC values correspond to important points along the ANC response
15    curve that are associated with levels of ecosystem impairment. The case studies used steady-state
16    critical loads models and focus on the combined load of sulfur and nitrogen deposition, below
17    which the ANC  level would still support healthy aquatic ecosystems. For each waterbody, the
18    total deposition in the year  2002 was compared with the estimated critical loads for the four
19    critical limit thresholds to determine which sites exceed their critical limit of deposition and
20    biological protection level.  Estimates of deposition were based on the sum of measured wet
21    deposition values from the year 2002 NADP  network and modeled dry deposition values based
22    on the year 2002 emissions and meteorology  using the Community Multiscale Air Quality
23    (CMAQ) model, respectively (REA 4.2). It is important to note  that a single level of ANC may
24    be caused by a range of deposition values due to heterogeneous  sensitivity among watersheds.
25          In summary, EPA staff concludes that a quantified relationship exists between the level
26    of surface water ANC and an amount of nitrogen and sulfur deposition. This relationship is
27    demonstrated by long-term trends going back to preindustrial conditions in the 1860s, recent
28    trends since the  1990s and critical loads modeling based on 2002 deposition data. Models are the
29    best way to evaluate how multiple environmental factors alter the relationship ANC  and
30    deposition.
      March  2010                               152                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1           5.2.1.4  What are the important variables in the ecological response to deposition
 2                   relationship^)?
 3           There  are numerous variables that modify the ANC to deposition relationship. The effects
 4    of these modifiers are described by models that parameterize ecosystems to simulate the process
 5    of acidification. The steady-state models used for critical loads analysis in the REA required
 6    input data for between 17 and 20 environmental parameters.
 7           The basic principle of the steady-state approach is to determine the maximum acid input
 8    that will balance the system at a biogeochemical safe-limit. Safe-limit is a subjective term that
 9    relates to a particular benchmark (e.g. ANC = 20, 50, 100), representing protection against
10    specific types and magnitudes of aquatic ecosystem response. The steady-state models that were
11    used in the REA relate an aquatic ecosystem's critical load to the weathering rate of its drainage
12    basin expressed in terms of the base cation flux. Weathering rate of geologic parent material is
13    the main source of base cations to an ecosystem. It is considered one of the governing factors to
14    ecosystem critical loads, and  therefore an important variable in the ecological response to
15    deposition relationship. Landscape features that are correlated to ecosystem acid-sensitivity
16    include lithology, elevation, percent forested watershed, and watershed area (Sullivan et al.
17    2007). A more detailed summary of the models and the environmental variables incorporated
18    into the models that were used in the REA is presented in Appendix A.
19           Numerous environmental variables affect the acidification process. Therefore the input
20    data required to run acidification models is rather extensive. For example, MAGIC, a dynamic
21    process based model of acidification, requires atmospheric deposition fluxes for the base cations
22    and strong acid anions as inputs to the model. The volume discharge for the catchment must also
23    be provided to the model. Values for soil and surface water temperature, partial pressure of
24    carbon dioxide and organic acid concentrations must also be provided at the appropriate
25    temporal resolution. The aggregated nature of the model requires that it be calibrated to
26    observational  data from a system before it can be used to examine potential system response. The
27    calibration procedure requires that stream water quality, soil chemical and physical
28    characteristics, and atmospheric deposition data be available for each catchment. The water
29    quality data needed for calibration are the concentrations of the individual base cations (Ca, Mg,
30    Na, and K) and acid anions (Cl, SC>42", and NCV) and the pH. The soil data used in the model
31    include soil depth and bulk density, soil pH, soil cation-exchange capacity, and exchangeable

      March 2010                                153                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    bases in the soil (Ca, Mg, Na, and K). The atmospheric deposition inputs to the model must be
 2    estimates of total deposition (wet and dry).
 3           In summary, the EPA staff concludes there are numerous variables which modify the
 4    ANC to deposition relationship. The relationships between environmental factors are  described
 5    by models that parameterize ecosystems to simulate the process of acidification. Weathering rate
 6    of geologic parent material is the main source of base cations to an ecosystem, and it is therefore
 7    considered one of the governing factors of ecosystem critical loads. Landscape features that are
 8    correlated to ecosystem acid-sensitivity include lithology, elevation, percent forested watershed,
 9    and watershed area. Consideration of the effects of environmental variables on the relationship
10    between environmental variables is  extensive in ecosystem acidification models. The  calibration
11    procedure requires that stream water quality, soil chemical and physical characteristics, and
12    atmospheric deposition data be available for each catchment.

13           5.2.1.5  Are these relationships applicable regionally ?
14           The relationship between ANC and N + S deposition based on catchment- scale modeling
15    is applicable regionally. Response to N and S deposition will vary catchment by catchment.
16    However, modeling every catchment in a region (i.e. a spatial area that includes a large
17    population of individual catchments) is implausible due to the extensive data requirements to
18    inform the simulations. A method to extrapolate watershed-scale analysis to a region was
19    developed in the REA. In that method, the critical loads (combined N+S load) developed for the
20    case study sites were applied over a region using water quality data. Critical load exceedance
21    (i.e., the amount of actual deposition above the critical load, if any) was calculated for each
22    waterbody in the region to quantify  the number of lakes or streams that receive harmful levels of
23    deposition. Lakes and streams with  positive exceedance values, where actual deposition was
24    above its critical load, were not protected at that critical limit (e.g. ANC= 20, 50, 100; REA
25    appendix 4).
26           In the Adirondack case study conducted in the REA, critical load exceedances were
27    extrapolated to lakes defined by the New England EMAP  probability survey. The EMAP
28    probability survey was designed to estimate, with known confidence, the status,  extent, change,
29    and trends in condition of the nation's ecological resources, such as surface water quality. For
30    the Adirondack Case  Study Area, the regional EMAP probability survey of 117  lakes were used
      March  2010                               154                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    to infer the number of lakes and percentage of lakes that receive acidifying deposition above
 2    their critical load of a population of 1,842 lakes. ANC limits of 20, 50, and 100 ueq/L were
 3    examined.
 4          In the Shenandoah case study, critical load exceedances were extrapolated using the
 5    SWAS-VTSSS LTM quarterly monitored sites to the population of brook trout streams that do
 6    not lie on limestone bedrock and/or are not significantly affected by human activity within the
 7    watershed. The total  number of brook trout streams represented by the SWAS-VTSSS LTM
 8    quarterly monitored sites is approximately 310 streams out of 440 mountain headwater streams
 9    known to support reproducing brook trout. ANC limits of 20, 50, and 100 ueq/L were examined.
10    (REA Appendix 4.3.1).
11          In summary, approaches were developed in the REA to extrapolate the ANC-deposition
12    relationship across a region. The data requirements for these approaches include (1) calculation
13    of critical loads of ANC using a catchment-scale model (2) stream chemistry data across the
14    region of concern, and (3) deposition loads across the region. With this information the
15    deposition load that would cause the stream to exceed the critical limit of ANC was calculated as
16    the critical load exceedance.

17          5.2.1.6   Are these relationships applicable nationally ?
18          The relationship between ANC and N + S deposition is applicable nationally.  Areas  that
19    have similar geologic underpinnings and weathering rates should show similar sensitivity to NOX
20    and SOX deposition. The critical load modeling that was used in the REA case studies requires
21    parameterization to each catchment. The spatial scale is small (e.g. catchment level) and the data
22    requirements are great (17+ environmental variables for each catchment) to use this method to
23    determine critical loads across all sensitive regions of the U.S. at this time. It is important to note
24    that acid-sensitivity often varies from catchment to catchment. Even if we did calculate critical
25    loads data for each catchment, aggregation of the catchment-scale data is appropriate for a
26    national standard.
27          The technique developed in the REA for extrapolating catchment-specific results to a
28    regional area determines the number of streams in a given area that show critical load (CL)
29    exceedances based on a selected value of ANC and deposition values for 2002. The approach
30    developed in the case study for extrapolating catchment-specific results to a regional  area is not
      March  2010                               155                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    immediately applicable across the U.S. because data for surface water chemistry and data for
 2    other input parameters is not available at a national scale.
 3           To summarize, the relationship between ANC and N + S deposition is applicable
 4    nationally. However the data required for critical loads analysis and extrapolation that is
 5    available on the regional scale is not available at the national scale. Considering this current data
 6    limitation the utility of the extrapolation approach developed in the REA to the national-scale is
 7    limited. Additional national-scale approaches are discussed in section 5.2.3.

 8           5.2.1.7  Summary
 9           In summary, EPA staff concludes that the available information from the ISA  and REA
10    supports the following characterization of aquatic acidification.  First, there is sufficient support
11    for the use of ecological indicators to characterize the responses of aquatic ecosystems to
12    nitrogen and sulfur deposition, and that ANC is the most supportable indicator. The available
13    information supports  using the acidification models to characterize the ecological response, using
14    ANC as the indicator, to nitrogen and sulfur deposition. Models are the best way to evaluate how
15    multiple environmental factors alter the relationship ANC  and deposition.
16           Heterogeneous sensitivity among watersheds is due in part to  landscape features.
17    Weathering rate of geologic parent material is the main source of base cations to an ecosystem,
18    and is therefore considered one of the governing factors of ecosystem critical loads. Landscape
19    features that are correlated to ecosystem acid-sensitivity include lithology, elevation, percent
20    forested watershed, and watershed area.
21           Modeling every catchment in a region is implausible due to the extensive data
22    requirements. The relationship between ANC and N + S deposition is applicable regionally. A
23    method to extrapolate watershed-scale analysis to a region was developed in the REA. In that
24    method, the critical loads (combined N+S load) developed for the case study sites were applied
25    over a region using water quality data. The data requirements for the regional extrapolation
26    include (1) calculation of critical loads for ANC using a catchment-scale model (2) stream
27    chemistry data across the region of concern, and (3) deposition loads  across the region. The
28    approach developed in the case study areas is not immediately applicable across the U.S. because
29    data for critical loads modeling and surface water chemistry is not available at a national scale.
30    However, it is important to note that the relationship between ANC and N+S deposition is


      March 2010                                156                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    applicable nationally. Areas that have similar geologic underpinnings should show similar
 2    sensitivity to NOX and SOX deposition.

 3           5.2.2   Design options for aquatic acidification
 4           The following design options describe the conceptual approach to integrating the
 5    ecological components of the standard outlined in section 5.1: ecological indicator, modifying
 6    factors, ecological response function and deposition metric. The goal is to illustrate how levels of
 7    NOX and SOX can be set to protect areas of the U.S. from acidic deposition.

 8           5.2.2.1   Is it appropriate to use ANC as the ecological indicator for the conceptual
 9                    design of the NOX and SOX standard based on aquatic acidification ?
10           There is strong  evidence supporting that ANC  is an appropriate ecological indicator for
11    aquatic acidification as discussed in Chapter 2 and Section 5.1.1 (as well as ISA 3.2.3 and REA
12    5.2.1). Options for the level of the indicator are discussed in Chapter 6. The options for the levels
13    are derived from experimental and observed evidence  in the scientific literature showing the
14    biological effects over a range of ANC values.

15           5.2.2.2   What is the appropriate ecosystem acidification model(s) to represent the
16                    ecological response function ?
17           In the REA, critical loads were calculated for specified ANC levels using the SSWC and
18    FAB models, these are referred to as acidification models, acid balance models or critical loads
19    models. The different assumptions of each modeling approach have implications that should be
20    considered in the conceptual design of a deposition-based NOX and SOX  standard. Most notably,
21    biogeochemical pathways of N and S deposition are considered differently in the two  models. In the
22    SSWC model, sulfate is assumed to be a mobile anion (i.e.  S leaching = S deposition), while nitrogen is
23    retained in the catchment by various processes. This assumption that all N is retained by the ecosystem
24    and does not contribute to acidification is incorrect in many instances because nitrate leaching is
25    observed. If nitrogen is leaching out of an ecosystem, obviously it has not been retained. Nitrate leaching
26    is determined from the sum of the measured concentrations of nitrate and ammonia in the runoff. The
27    critical load for sulfur that is calculated by SSWC can be corrected for the amount of nitrogen that
28    contributes to acidification. When an exceedence value for the critical load is calculated, the critical load
29    is subtracted from S deposition plus the amount of nitrate leaching, as it represents the difference between
      March 2010                                157                Draft - Do Not Quote or Cite

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20

21
      Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

N deposition and N retention by the ecosystem. N leaching data used in this calculation is considered
robust.
       In contrast to the SSWC approach, the FAB model includes more explicit modeling of N
processes including soil immobilization, denitrification, and wood removal, in-lake retention of N
and S, as well as lake size. Although N cycling is more detailed in the FAB model, there is
greater uncertainty in the input data needed to characterize the components of the N cycle. The
FAB model yields a deposition load function for a specified level of an endpoint. This function is
characterized by three nodes that are illustrated on Figure  5-3: 1) the maximum of amount of N
deposition when S deposition equals zero (DLmax (N)), 2) the amount of N deposition that will
be captured by the ecosystem before it leaches (DLmin(N)) and 3) the maximum amount of S
sulfur deposition considering the N captured by the ecosystem (DLmax (S)). The function
represents many unique pairs of N and S deposition that will equal the critical load for acidifying
deposition. The slope portion of the function will vary according to attributes of the water body
that is modeled, including lake size and in-lake retention.
                                                                   H exposition
       Figure 5-3. The depositional load function.
       A third modeling approach, which synthesizes components of each model used in the
REA, is suggested by staff for catchment scale modeling in developing the NAAQS. The
foundation of the proposed approach is the SSWC model because there is high confidence in the
input data required. The SSWC model for aquatic acidification is expressed as equation  1.
                                                                                     (1)
22    where,
      March  2010
                                          158
Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    DLANciim(N+S) = depositional load of S and N that does not cause the ecosystems to exceed a
 2    given ANCiim
 3    [BC]0* = the preindustrial concentration of base cations (equ/L)
 4    ANCumit = a "target" ANC level (equ/L)
 5    Q= surface water runoff (m/yr) (this is typically equal to precipitation -evapotranspiration
 6
 7          This model could be further constrained by a quantity of N which would which would be
 8    taken up, immobilized or denitrified by ecosystems and adjust the quantity of deposition required
 9    to meet a specified critical load. This term is represented as DLmin(N) in the FAB model and
10    illustrated in Fig. 5-3. For application in the NAAQS and in the following discussion, the
1 1    parameter is designated with the abbreviation NEco- The acid-base model constrained by NEco is
12    expressed by equation 2.
13                           DLANClimN + S=(BC   -ANC^    + N^                     (2)
14
1 5    where,
16    Neco= nitrogen retention and denitrification by terrestrial catchment and nitrogen retention in the
17    lake
18
19          The term Neco could be derived multiple ways, each yielding different ultimate results.
20    The first is by taking the mean value calculated to represent the long-term amount of N an
21    ecosystem can immobilize and denitrify before leaching (i.e. N saturation) that is derived from
22    the FAB model [denoted as DLmin(N) in the FAB model]. This approach requires the input of
23    multiple ecosystem parameters. Its components are expressed by eq 3.

24                             Neco = fNupt + Nret + (l - r \Nmm + Nden )                       (3)

25    where,
26    Nupt= nitrogen uptake by the catchment
27    Nimm= nitrogen immobilization by the catchment
28    Nden=denitrification of nitrogen in the catchment,
      March  2010                               159                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    N-et = in-lake retention of nitrogen
 2    f =forest cover in the catchment (dimensionless parameter)
 3    r = fraction lake/catchment ratio (dimensionless parameter)
 4
 5          The second approach for estimating Neco is to take the difference between N deposition
 6    and measured N leaching in a catchment as expressed by eq 4.

 7                                      Neco=DL(N)-Nleach                               (4)

 8          It is unclear which approach for calculating NECO should be used in developing the
 9    NAAQS. The two equations can result in quite different values (See section 5.4 for an example
10    calculation).
11          To summarize,  the SSWC model assumes N deposited to the ecosystem is retained by the
12    ecosystem, while also assuming that all S deposition is leached and contributes to aquatic
13    acidification. The critical load is calculated for S deposition, and the N that contributes to
14    acidification is incorporated into the exceedance calculation. The FAB model considers a
15    detailed accounting of the N cycle; however confidence  in the input data to the model is more
16    uncertain. The FAB approach yields a function which may be solved by many unique pairs of N
17    and S deposition. A minimum amount of N deposition that will be captured by the ecosystem
18    before it leaches is included in the calculation of the maximum amount of S deposition. A third
19    approach is suggested by staff as the most appropriate approach for informing the structure of the
20    NOX and SOX secondary standard. This approach constrains the critical load calculated from a
21    SSWC method by a value of NEco [previously defined as DLmin(N)] which accounts for the
22    amount of N deposition that would be taken up by the ecosystem and, therefore, would not
23    contribute to acidification.

24          5.2.2.3  How are results of acidification models aggregated to adequately represent a
25                   larger spatial area and inform a deposition metric?
26          So far in this section, the ecological indicator would be established as ANC. Acidification
27    models are considered  the best way to describe the relationship between ANC and deposition and
28    to describe how this relationship is altered by modifying factors. If deposition is known the
29    model may be run to calculate the resultant ANC. If a target ANC level is desired the model may

      March 2010                               160                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    be run to calculate the corresponding deposition load that should not be exceeded (i.e. the critical
 2    load). The following discussion will focus on the critical load application of the acidification
 3    model. It is important to emphasize that the acidification models are only applied at the spatial
 4    scale of the catchment. Spatial aggregation of critical loads are necessary to inform the
 5    discussion of appropriate design and levels of a national standard.
 6           Acidification models are parameterized for catchments. The critical loads that they
 7    calculate for N and S deposition based on a specified ANC limit vary at the small spatial scale of
 8    the catchment to the degree that acid-balancing properties of the catchments vary. Despite this
 9    variation, the goal of aggregating critical loads from multiple catchments is to develop an
10    appropriately representative deposition value, which is adequately protective of ecosystems and
11    could be  applied over larger spatial areas.
12           Staff proposes evaluating the critical loads for a specified ANC limit of a population of
13    waterbodies to calculate a benchmark deposit!onal load in which a specified percentage of the
14    population does not exceed their critical load. This approach uses the distribution of critical loads
15    from a population to derive a value that is intended to provide protection over a spatial area that
16    is larger than the individual catchment for which a single critical load may be calculated. An
17    example  of this technique is calculated in section 5.5.  The ecological indicator would be a single
18    value of ANC, and the acidification models would calculate the critical loads for the specified
19    ANC level for individual catchments across a spatial area. The deposition metric would be an
20    amount of deposition such that a specified percentage of a population of water bodies does not
21    exceed a critical load for the specified  value of ANC. The deposition metrics could be calculated
22    for populations of catchments that are  categorized according to acid-sensitivity, as described in
23    the next section.

24           5.2.2.4  How are modifying factors of the ecological response to deposition function
25                   considered at the national-scale?
26           As previously noted,  critical loads for ANC vary at a small spatial scale, catchment by
27    catchment. As it is implausible to model the acidification status of every catchment in the U.S,
28    an alternative is to develop a deposition metric for a population of catchments, assuggested  in the
29    previous  section. The following design options focus on relating acid-sensitivity, based on ANC,
30    to a feature(s) of the landscape at a national-scale by creating acid-sensitivity categories. A
      March 2010                                161                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    population of catchments could then be defined to represent these categories and a representative
 2    deposition metric chosen.
 3           Acid sensitivity classes based on bed rock geology
 4           Here an approach is presented in which ecosystem sensitivity to acidification is
 5    categorized into classes based on bedrock geology/ lithology. The approach is supported by
 6    conclusions from the ISA in which geologic bedrock is determined to be the governing factor
 7    that drives ecosystem sensitivity to acidification (ISA 3.2.4.1). Specifically, geologic bedrock
 8    with a low base cation supply leads to ecosystems that are sensitive to acidifying deposition. A
 9    method to develop a deposition metric, based on the distribution of critical loads of a
10    representative population, for each category of acid-sensitivity is presented here.
11           A map was developed to capture the heterogeneity of geologic bedrock that occurs across
12    the eastern U.S. and link it to ecosystem acid-sensitivity (Fig5-4). The method is based on
13    Sullivan et al.(2007) in which 70+ primary lithologies are grouped into 5 categories of acid-
14    sensitivity, using ANC as the ecosystem indicator upon which acid-sensitivity is based. Sullivan
15    et al. (2007) evaluated multiple features of the landscape and found that geology is the landscape
16    parameter that governs ecosystem sensitivity to acidic deposition. The analysis in Sullivan et al.
17    2007 was conducted in the Southern Appalachian Mountains region, which included sites from
18    the states of GA, TN, NC, KT, VA and WV. EPA is conducting additional analyses to further
19    test the concept that lithology correlates to acid sensitivity in case study areas and in the western
20    U.S. EPA staff intends that some of these additional analyses will be available at for review in
21    the second draft of the policy assessment.
22           As previously stated, acidification often varies catchment by catchment. Therefore there
23    will be variation in terms of acid-sensitivity among catchments within each acid-sensitivity  class
24    designated by the map. Despite this variation, lithology is a nationally applicable landscape
25    feature which is known to govern acid-sensitivity. Ultimate detail and rigor would be provided
26    by modeling deposition and consequential acidification of each catchment in the U.S., an
27    approach which would require knowledge of 17+ environmental parameters for each catchment.
28    However classification of the landscape into categories based on geology provides a national -
29    scale landscape feature to extrapolate the results of catchment-scale modeling.
      March 2010                                162                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
                            Acid-Sensitive Areas of the Eastern United States
                                    A Classification based on Bedrock Geology
                                         L'S Environmental Protection Agency
                          Projection Alter*EquwAr
                          F*]t4_e*iMig 0 OOOOCO
                          FUM fMXTiH-g 000003?
                          C*ntr« M
    About the Classification
Tn« nwp !* MM on • 250.000 ttMtoek Q.MW uu
from OM US O»=**jC8i Su-v»y. rl Mt M*n ti*i lifted
Mwd on « mwrnod .n Sunvin *l M ..2006,1. --im vw
                     [   | Slate Boundaries
                     Acid Sensitivity
                     ^^| Cartxinale - Least Sensitive
                        Silaceous - Less Sensilive
                        Argillaceous • Sensitive
                       ~'j Felsic -
                                                                     Water
 1
 2           Fig 5-4. A map of acid sensitive areas of the Eastern U.S. developed from a
 3           lithology-based five-unit geologic classification system after methods in Sullivan
 4           etal. (2007).
 5           Acid sensitivity based on multiple landscape features
 6           Although bedrock geology is a governing factor of acid sensitivity, multiple factors have
 7    been shown to contribute to sensitivity. Topography is a characteristic of the landscape that is
 8    often shown to correlate with acid-sensitivity, specifically low elevations, which generally
 9    receive some cations from higher elevation sites, are less sensitive that higher elevation sites
10    (ISA 3.2.4.1). Could both topography and bedrock geology be included a national map of acid-
11    sensitivity? A map of high elevation could be layered  over the map of bedrock categories. If all
12    high elevation areas were within the sensitive geologic categories, then the additional parameter
13    would further refine the spatial resolution of sensitivity within the bedrock categorization.
14    Moreover, the approach will provide more spatial detail  on the  sensitivity within areas already
15    considered sensitive based on bedrock geology. It's unclear if elevation alone would help
      March  2010
        163
Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    identify new sensitive areas. It's also unclear if greater spatial resolution of sensitivity within
 2    areas already identified as sensitive would be helpful in terms of relating the national-scale
 3    landscape features to critical loads. Should additional multiple features be considered when
 4    categorizing the landscape according to acid-sensitivity? We are providing this design option to
 5    elicit comment; it is presented as a conceptual idea.

 6           5.2.2.5  How is a deposition metric developed so that critical loads for catchments are
 1                   aggregated to adequately represent classes of acid sensitivity based on
 8                   geology?
 9           The values that represent a deposition metric for the acid-sensitivity categories could be
10    derived from the critical loads calculated for the case study analysis in the REA. The case study
11    sites (Adirondack and Shenandoah areas) occur in areas that are predominately composed of the
12    two most acid-sensitive types of bedrock geology. Therefore the case study sites would represent
13    those sensitivity categories. The deposition and atmospheric concentration tradeoff curves for a
14    specified level of ANC for each bedrock geology site would be based  on a deposition metric
15    derived from the distribution of critical loads within the case study areas.  It could be a central
16    value such as the mean or median value or a value representing a percentile of the distribution,
17    such as the  95th percentile. Central estimates, such as the mean, would likely not be projected to
18    achieve the target ANC of the majority of acid-sensitive ecosystems; therefore it may be
19    preferable to calculate the spatially aggregated value for some percentage of catchments to
20    project achieving the ANC for the more sensitive ecosystem types. For example, if projecting
21    85%, 90% or 95% of the aquatic ecosystems achieving the ANC is selected, then the deposition
22    metric that represents the critical load for the 85th, 90th  or 95th percentile of the population would
23    be selected. An example calculation for the Adirondacks is presented in section 5.5.

24           5.2.2.6  How is reduced nitrogen appropriately considered in the deposition metric?
25           Reduced forms of nitrogen deposition are quickly converted to nitrate in the environment
26    and use up the assimilative capacity  of ANC at the same rate as oxidized forms of nitrogen
27    deposition;  therefore, reduced nitrogen deposition must be accounted for in the watershed. There
28    are two basic approaches to accounting for the use of this assimilative capacity.
29           The suggested approach is to subtract the loadings of reduced forms of nitrogen derived
30    for a given  spatial area from the deposition metric that  represents selected percentage of critical

      March 2010                               164               Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    loads for a given population, such that the resultant deposition metric is for sulfur and oxidized
 2    nitrogen only. This approach assumes that the reduced forms of nitrogen deposition are relatively
 3    constant over time. This assumption could lead to over or under protection for an area depending
 4    on whether the actual concentrations of reduced forms of nitrogen increase or decrease over
 5    time. An example for how to subtract  reduced nitrogen from the deposition metric based on
 6    nitrogen and sulfur is given in section 5.5.

 7           5.2.2.7  Summary
 8           In summary, the ecological components of the conceptual design for a standard base on
 9    aquatic acidification include the ecological indicator, ecological response function and its
10    modifiers and the deposition metric. A summary how each component is considered in the
11    conceptual design is given in Table 5-2. Using ANC as the ecological indicator, an approach is
12    suggested for using an acidification model constrained by a parameter for ecosystem N retention
13    to represent the ecological response function. The best way to calculate ecosystem N retention is
14    as of yet unclear. It is proposed that the national landscape is categorized in terms of criteria that
15    denote acid-sensitivity. It is well known that bedrock geology is a governing factor of acid-
16    sensitivity, in other words ecosystem response is modified across the landscape due in part to
17    bedrock geology. It is unclear  if landscape categorization based on geology is the best approach
18    or other criteria/combination of criteria should be used.
19           The distribution of critical loads for a specified target ANC from a population of
20    catchments representing an acid-sensitivity category, based on geology or some combination of
21    factors, can be calculated From this a  deposition metric,  an amount of deposition, could be
22    calculated such that a specified target  percentage of the population of water bodies in the acid-
23    sensitivity category does not exceed a critical load for the specified value of ANC. Moreover, the
24    deposition metric would reflect both the selected level of ANC and the percentage of catchments
25    in the  representative population that do not exceed their critical load. Reduced nitrogen
26    deposition,  average over a determined spatial scale, would be subtracted from the deposition
27    metric yielding a value for allowable deposition from NOy and  SOX. The deposition from NOy
28    and SOX would be converted to atmospheric concentrations of NOy and SOX by the methods
29    described in section 5.4.
      March 2010                                165                Draft - Do Not Quote or Cite

-------
       Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
Table 5-2. Summary of the ecological components of design option 1.
Values given for illustrative purposes only. Levels are discussed in Chapter 6 and ultimately
selected by the administrator.
                      Modifying Factor
                        Geology
                   5 categories of sensitivity
                                                          Variable/Fixed
                                                            Modifying
                                                              Factors
                                         Deposition
                                           Metric
                       Ecological
                     Response to
                    Deposition Function
  Ecological Indicator
                                                             Atmospheric
                                                              Deposition
                                                            Transformation
                                                               Function
                                                           See Section 5.4
                                        Determined by
                                          the % of
                  Acidifcation mode
                                         ecosystems
                                         represented
Proposed levels based
  on biological effects
                               Concentration
                                     of
                                 Air Quality
                                 Indicator(s)
Ecological Indicator
                        Ecological Response
                             Function
  Modifying Factor
  Deposition Metric
ANC; level reflects
degree of
Effects on aquatic
biota in the ecosystem
                       Acidification model
                       constrained by a
                       parameter for N
                       retention
Acid-sensitivity
categories, based on
geologic bed rock or a
combination of factors,
that may be applied at a
national scale
Determined from the
distribution of critical
loads from a
population that can be
related to an acid-
sensitivity category.
Reduced nitrogen
subtracted from the
deposition metric to
yield allowable
deposition from NOX
and SOX.
March 2010
                                                166
                      Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    5.3   ECOLOGICAL COMPONENTS OF THE STANDARD:
 2          TERRESTRIAL ACIDIFICATION, TERRESTRIAL NUTRIENT
 3          ENRICHMENT AND SURFACE WATER NUTRIENT
 4          ENRICHMENT
 5          These effects were not included in the conceptual design for the first draft of the PA,
 6    however a brief summary of our approach for developing standards that are protective of these
 7    ecological effects follows.

 8          5.3.1   Terrestrial Acidification
 9          The deleterious effects of terrestrial acidification on tree species is indicated by base
10    cation to aluminum ratio (Be: Al) of soils. Critical load functions were developed in the REA that
11    relate Bc:Al threshold values (0.6, 1.2 and 10) to values of N+S deposition using the simple mass
12    balance (8MB) model. The exceedance of these critical loads were calculated at the two study
13    sites and then extrapolated over 24 states. Like aquatic acidification, sensitivity of terrestrial
14    ecosystems to acidification is linked to the geologic bedrock. Moreover, areas that are sensitive
15    to aquatic acidification should also be sensitive to terrestrial acidification. Therefore, an
16    approach similar to that described for aquatic acidification could be developed. This would mean
17    that a critical load based on Bc:Al at either 1.2 or 10 would be calculated to protect a percentage
18    of the terrestrial landscape. This value would then be assigned to categories of acid sensitivity
19    based on geology.
20          This could result in two  standards, one for aquatic ecosystems and one for terrestrial
21    ecosystems. This leads to the question, are aquatic or terrestrial ecosystem more sensitive? To
22    answer this question, an analysis was conducted  in which critical loads for the Adirondacks and
23    Shenandoah case study areas were calculated based  on the terrestrial ecosystem indicator, Be: Al,
24    at the level of 1.2 and 10. The terrestrial critical loads were compared to the critical loads  for
25    aquatic ecosystems. A full description of this analysis and results is available in Chapter 7, the
26    results are briefly summarized here. In the Adirondacks case study area, 7 of the  16 watersheds
27    had terrestrial critical acid loads (based on a Bc:Al of 10.0) that were lower and therefore  more
28    sensitive to acidification than all the lakes in the  watershed. However, when the terrestrial critical
29    loads were calculated with a Bc:Al soil  solution ratio of 1.2, only 5 of the 16 watersheds were
30    protected by a terrestrial critical load that was lower than the aquatic critical loads of the lakes. In

      March 2010                               167                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    the Shenandoah case study area terrestrial critical loads offered a higher level of protection than
 2    aquatic critical loads in only one watershed. If two standards were proposed, the one that allows
 3    lower ambient levels of NOX and SOX would be controlling in a given area.

 4           5.3.2  Terrestrial and surface water nutrient enrichment
 5           NOX and NHX are the main contributors to nitrogen deposition. The effects of nitrogen
 6    deposition on terrestrial ecosystems and surface waters are many. Most notable are the effects on
 7    ecosystem biodiversity found across the U.S and affecting multiple taxonomic groups including
 8    vascular plants, algae, mycorrhiza and lichens (ISA 3.3). Unlike terrestrial and aquatic
 9    acidification, there is no one, well-supported chemical or biological indicator of ecosystem
10    effects that occurs across the nation. In order to develop a NAAQS based on nitrogen enrichment
11    effects there needs to be one indicator that can be applied across the nation. It is possible that we
12    could develop an index in which information on different ecological indicators could be input
13    and the output would be an index score that could be consistently applied across the U.S. It is not
14    clear how to develop such an index.
15           Nitrogen critical loads are known for many ecosystem endpoints in the U.S. and are
16    published in the scientific literature.  Additionally, critical loads for ecosystems in Europe, many
17    of which are similar to U.S. ecosystems, have been reported for over a decade, they are
18    continually refined through periodic assessments of the scientific literature, and they are
19    currently supported by a strong weight of peer-reviewed scientific information (ISA  3.3).
20    Additional critical load modeling was not conducted in the REA because of two factors. There
21    are numerous reports in the peer-reviewed scientific literature and there is no model available to
22    conduct such analysis for multiple endpoints and ecosystems. However, based on nitrogen
23    critical loads published in the literature, the REA evaluated the extent of the landscape
24    represented by those critical loads and their exceedances (REA 5.0).
25           A standard that integrates acidification and nutrient effects could conceptually be quite
26    simple. The total nitrogen deposition allowed for a deposition metric based on acidification could
27    be constrained so that it does not exceed a value based on a deposition metric for a nutrient
28    related effect.
      March 2010                                168                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1           5.3.3  Summary
 2           Conceptual design of NOX and SOX NAAQS were not developed for terrestrial
 3    acidification and terrestrial/surface water nitrogen enrichment in the first draft PA, however a
 4    brief summary of a potential structure for these ecological effects is presented. The ecological
 5    indicator for terrestrial acidification would be Bc:Al because it relates to both atmospheric
 6    deposition of N+S and deleterious effects on tree growth. Critical loads would be related to acid-
 7    sensitivity categories and calculated according to similar methods presented for aquatic
 8    acidification effects. This could result in two standards, one for aquatic ecosystems and one for
 9    terrestrial ecosystems. If two standards were proposed, the one that allows lower ambient levels
10    of NOX and SOX would be controlling in a given area. Unlike terrestrial and aquatic acidification,
11    there is no one, well-supported ecological indicator of nitrogen deposition effects that occurs
12    across the nation. In order to develop a NAAQS based on nitrogen enrichment effects there
13    needs to be one indicator that can be applied across the nation.  Although, the specifics of an
14    approach are unclear, it may be possible that we could develop an index in which information on
15    different ecological indicators could be input and the output would be an index score that could
16    be consistently applied across the U.S. A standard that integrates acidification  and  nutrient
17    effects could conceptually be quite simple. The total nitrogen deposition allowed for a deposition
18    metric based on acidification could be constrained  so that it does not exceed a value based on a
19    deposition metric for a nutrient related effect.

20    5.4    LINKING DEPOSITION TO ATMOSPHERIC CONCENTRATION

21           5.4.1  Background
22           Atmospheric pollutants deposit onto land and water surfaces through at least two major
23    mechanisms: direct contact with the surface (dry deposition), and transfer into liquid
24    precipitation (wet deposition). The magnitude of each deposition process is related to the
25    ambient concentration through the time-, location-, process- and species-specific deposition
26    velocity (Seinfeld and Pandis, 1998) and can be conceptualized as:

27                                      DePlD>y =v1Dry-C,Amb                                 (1)

28                                       Dep,Wet=v,wret-CiAmb                                  (2)

      March 2010                               169                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX




       1     Dry    .  Wet     ,    ,      ,     ,     . .      ,   . .    ^  Dry   , ^  Wet    .   .     .
 1    where vt   and v,   are the dry and wet deposition velocities, Dept   and Dept   are the dry and



 2    wet deposition fluxes, Cf   is the ambient concentration, and the /' subscript indicates the



 3    pollutant species under study. The wet deposition velocity term is a conceptualized term and not


 4    a state variable that allows for the grouping of wet and dry deposition. The total deposition of


 5    each pollutant is




 .-                                     r^   Tot   r^    Dry   7-.   Wet                             ,~\
 6                                    Dept   =Depi   +Dept                                (3)





 7    Substituting Equations 1 and 2 into Equation 3 yields




                                   87-.  Tot    Dry  /^< Amb     Wet  ,~t Amb                          / A\
                                   DePl    =vt  v -C,    +v,.   -Ct                             (4)





 9    The total deposition of sulfur or nitrogen would therefore be:




t r\                               r-\    Tot  ^~^ /  Dry     Wet \     /-i Amb                         /c\
10                               Dep^   = ^ (v,.  'y+vi   ) • mi • C,                           (5)

                                             ;




11    where m  is the molar ratio of the atom (sulfur or nitrogen) of interest to the /'th pollutant.



12    Ambient sulfur- and nitrogen-containing pollutants include gases such as sulfur dioxide (SO2),



13    ammonia (NH3), various nitrogen oxides (NO, NO2, HONO, N2O5),  nitric acid (HNO3), and



14    organic nitrates such as peroxyacetyl nitrates (PAN); as well as particulate species such as sulfate


15    (SC>42"), nitrate (N(V), and  ammonium (NH4+). As discussed in chapter 4, the  definitions of NOy


16    and SOX species for the purposes of this review include the sulfur-containing species above and


17    the above oxidized forms of nitrogen (NOy); ammonia and ammonium are not currently included


18    as  listed pollutants (see Chapter 8 for an expanded discussion of the  role of NHX).




19           5.4.2  Aggregation Issues



20           Equation 5 provides a relationship for converting sulfur or nitrogen deposition to


21    "equivalent" ambient concentrations,. A major issue to consider during such conversion is the


22    treatment of spatial, temporal and chemical resolutions of the deposition data and the resulting


23    standards. Since the objective is to set an ambient air quality standard for total oxidized sulfur


24    and nitrogen,  and this is also the  chemical resolution provided by the ecosystem models, it is


25    convenient to use a relationship with the following form:





      March 2010                                170                Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 i                                     r\     Tot   T T-     .-"r   A mb                              / s- \
 1                                     DePs\N   = VSIN • CsiN                                 (6)

 2    where VS/N can be considered an aggregateddeposition to ambient air transformation ratio,
 3    referred to herein as the deposition transformation ratio, that relates total deposition of sulfur or
 4    nitrogen to the total ambient concentration, and represents an average of the species specific v,Tot
 5    ( = v,Dry + v,Wet) values in Equation 5. The sulfur and nitrogen concentrations are the result of
 6    applying the ni; values to the C;^11 values in Equation 5.
 7           Since the deposition critical loads are expressed in terms of annual total deposition, the
 8    most relevant averaging time for equivalent ambient concentrations is the annual average. Data
 9    used to derive annual VS/N values will need to have the same spatial representativeness as the
10    depositonal loads. To be clear, the deposition transformation ratio is not a state variable, but
11    simply is a calculated term that facilitates the linkage between deposition and concentrations
12    which is a necessary step in developing ambient air indicators that are used to assess compliance
13    with a NAAQS. There will  be a tendency that is not scientifically defensible to compare
14    deposition ratios with deposition velocities that are uniquely determined on a species by species
15    basis influenced by numerous factors as discussed earlier.

16           5.4.3   Air Quality Simulation Models
17           Ideally, VS/N values would  be derived for each area of interest from concurrently collected
18    sulfur and nitrogen deposition and concentration measurements. However, no monitoring
19    network currently exists that can provide such information. We therefore propose using output of
20    the CMAQ model for initial calculation of VS/N values.
21           CMAQ provides both concentrations and depositions for a large suite of pollutant species
22    on an hourly basis  for 12 km grids across the continental U.S. Its comprehensive structure is
23    ideal for providing VS/N values that appropriately address the chemical and temporal aggregation
24    issues discussed above, and weighted spatial averages of the gridded data can be used for areas
25    that span multiple grid cells. Potential concerns with using CMAQ-predicted concentrations and
26    depositions for this purpose stem from the various, but unquantifiable uncertainties in model
27    formation and input data, which will be discussed  in the next draft of this PAD.
28           CMAQ does not directly calculate or use VS/N values;  instead the following procedures
29    are used in the code to model deposition:
      March 2010                                171                 Draft - Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1           1) vdry values of gaseous pollutants are calculated in the CMAQ weather module called
 2    the Meteorology-Chemistry Interface Processor (MCIP) through a complex function of
 3    meteorological parameters (e.g. temperature, relative humidity) and properties of the geographic
 4    surface (e.g. leaf area index, surface wetness)
 5           2) vdry values for paniculate pollutants are calculated in the aerosol module of CMAQ,
 6    which, in addition to the parameters needed for the gaseous calculations, also accounts for
 7    properties of the aerosol size distribution
 8           3) vwet values are not explicitly calculated. Wet deposition is derived from the cloud
 9    processing module of CMAQ, which performs simulations of mass transfer into cloud droplets
10    and aqueous chemistry to incorporate pollutants into rainwater, all of which is conceptually
11    contained in the vwet parameter in Equation 2.
12           Due to lack of direct measurements, no performance evaluations of CMAQ's dry
13    deposition calculations can be found; however, the current state of MCIP is the product of
14    research that has been based on peer-reviewed literature from the past two decades (EPA, 1999)
15    and is considered to be EPA's best estimate of dry deposition velocities. Some bias has been
16    found between CMAQ's wet deposition predictions and measured values (Morris et al., 2005);
17    recent analyses suggest that poor simulation of precipitation could be responsible for this (Davis
18    and Swall, 2006), which can potentially be dealt with by recalculating wet deposition using
19    precipitation measurements. Although the model is continually undergoing improvement,
20    CMAQ is EPA's state-of-the-science computational framework for calculating deposition
21    velocities, and was therefore the logical first choice as a source for VS/N values.

22           5.4.4  Oxidized Sulfur and Nitrogen Pollutant Species
23           Ideally, all possible air pollutant species that contribute to ecological adversity would  be
24    considered for VS/N values. The pollutant list is constrained by the source of VS/N values, which is
25    currently CMAQ output. Table 1 lists the oxidized sulfur and nitrogen species currently available
26    in CMAQ whose data will be used for VS/N values.
27           One issue that needs explicit consideration is the contributions of particles larger than
28    PM2.5 to sulfur and nitrogen deposition. A recent review of particle deposition measurements
29    (Grantz, Garner, and Johnson, 2003) showed that coarse particles generally deposit far more
30    sulfate and nitrate in forest ecosystems than fine particles. However, CMAQ does not currently

      March 2010                               172                Draft - Do Not Quote or  Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    provide simulationsof coarse paniculate sulfate and nitrate. This is an issue that needs to be
 2    addressed by developers of either the model or the future SOX/NOX measurement network to set
 3    scientifically sound standards.

 4           5.4.5  Example Calculations
 5           Figure 5-5 shows annual inverse VS/N values16 calculated for each 12 km grid in the
 6    eastern and western domains for a 2002 CMAQ v4.6 simulation, which is the quantity that would
 7    be used for conversion of deposition load tradeoff curves which illustrate (see Section 6) the
 8    combinations of NOy and SOX conventartions that would correspond to an established critical
 9    load. Figure 5-6 shows an example application of these ratios for a lake in the Adirondacks.
10    Deposition load tradeoff curves for this lake (see Section 6for their calculation) are multiplied by
11    the inverse VS/N value from the appropriate grid cell in Figure 1 to convert those depositions to
12    ambient concentrations of sulfur and nitrogen.
13           A CMAQ v4.7 simulation for multiple years (2002-2005) recently became available,
14    which was used to examine the inter-annual variability of inverse VS/N values. The grid-specific
15    coefficients of variation  (CV) are shown in Figure 3. Figure 5-7 shows that CV values are
16    relatively small (< 25%) in the Adirondacks and Shenandoah case study areas. This suggests that
17    a 3-year average of the ratios may be a sufficiently stable representation of deposition velocities
18    for converting the deposition load curves to ambient concentrations in future applications.
      16 Inverse VS/N values represent the multiplier needed to convert deposition levels into atmospheric concentrations of
      NOx and SOx.

      March  2010                                173                Draft - Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
1
2
     Table 5-3. Oxidized sulfur and nitrogen species currently available in CMAQ simulations. Note
     that PNA concentrations are not available in current CMAQ extractions.
:"'.:illlimit C.'IfL-i-,
•inlthr Oxides
Nirr.igsn Oxides








Clinn:i: :L| C.M. u; • S|::::'i-r-^ Hvn.l..)!
SO,
so*-
.\O
NOj
NOJ
-NA
MONO
PAN
PA.NX
NTH
PNA
Hpsris-i Ntn.:"
Hiilliir Dioxide
Hill rare
Nirrogsn Oxide
\irrogen Dioxide
Nitrata
Dinit-mgen peiitoxide
Nirric Acid
Itroxvacetvl nitrate
Hijjher oi'der peroxyflcetyl nitrates
Orfiaiiic Nitrates
BMCkt
>'..,!:-,-
Predoiuiiiautly pftrticulat*


PredomiDantly paniculate






4
5
6
Figure 5-5. VS/N values for each grid cell in the eastern (right) and western (left)
U.S. domains. The top maps are for sulfur and the bottom are for nitrogen.
     March 2010
                                     174
Draft - Do Not Quote or Cite

-------
      Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
                    Deposition
                                                            Concentration
1
2
3
4
                 7     13     20

                  N deposition (kg/hay)
                                                                              -ANC100
                                                                                ANC50
                                                                              - ANC20

                                                                                Current
                                                                              • Conditions
                                                                                (CMAQl
       Figure 5-6. Schematic Diagram illustrating the procedure for converting
       deposition tradeoff curves of sulfur and nitrogen to atmospheric concentrations of
       SOX and NOX.
March  2010
                                                   175
Draft -Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
                                  Coeflicent ol Variation of N Cone Dep ralio, 2002-2005
3
4
           a)
                                  Coellicenl of Variation ol S Cone.Dep ralio. 2002-2005
           b)
                                                                      '
                                                                                CV (%)

                                                                                    rSO.O


                                                                                     37.5


                                                                                     25.0


                                                                                    \ 12.5


                                                                                     0.0
                                                                                CV (%)
                                                                                   — 50.0
                                                                                     37.5


                                                                                     25.0


                                                                                    [12.5


                                                                                     0.0
Figure 5-7. Inter-annual coefficients of variation (CV) of a) nitrogen and b) sulfur
VS/N values, based on a series of 2002-2005 CMAQ v4.7 simulation.
     March 2010
                                        176
Draft -Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    5.5   EXAMPLE CALCULATION FOR THE CONCEPTUAL DESIGN
 2          AND DERIVATION OF AAPI
 3          Section 5.2 describes a proposed conceptual design for a NOX and SOX NAAQS based on
 4    aquatic acidification. To summarize the process of acidification, atmospheric deposition of NOX
 5    and SOX contributes to acidification in aquatic ecosystems through the input of acid anions, such
 6    as NO3" and SO42". The acid-base balance of headwater lakes and streams is controlled by the
 7    level of this acidifying deposition of NOs" and SC>42" and a series of biogeochemical processes
 8    that produce and consume acidity in the watershed. The biotic integrity of freshwater ecosystems
 9    is then a function of the, acid-base balance and the resulting acidity-related stress on the biota
10    that occupy the water. Given some "benchmark level" of ANC [ANClimit]) that appropriately
1 1    protects biological integrity, the depositional load  of acidity DL(N+S) is simply the input flux of
12    acid anions from atmospheric deposition that result in a surface water ANC level equal  to the
13    [ANClimit] when balanced by the sustainable flux of base cations input and the sinks of nitrogen
14    and sulfur in the watershed catchment.

15          5.5.1   Example calculation for the conceptual  design
16          This section summarizes and provides an example calculation of the approach proposed
17    by EPA staff to calculate (1) the acid-base balance of a catchment for a specified ANC level, (2)
18    the N and S deposition tradeoff curves for a deposition metric, which represents  a specified
19    percentage of the total population of water bodies  that do not exceed their critical load at a
20    specified ANC level and (3) the conversion from tradeoff curves for N and S deposition to those
21    for atmospheric concentrations of NOy and SOX. The equations representing deposition  loads and
22    associated tradeoff curves for a specified level of ANC are the basis for deriving the form of the
23    standard discussed above in section (5.5.2).
24          Equation (1) expresses the model that we suggest using to determine the amount of N and
25    S that may be deposited onto a catchment to yield  a specified level of ANC.
26                          DL^ (N + S)= ([BCl - [ANC^ ])Q + Neco                     (1)

27   where,
     March 2010                              177                 Draft-Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    DLANciim(N+S) = depositional load of S and N that does not cause the ecosystems to exceed a
 2    given ANCiim
 3    [BC]0* = the preindustrial concentration of base cations (equ/L)
 4    ANCumit = a "target" ANC level (equ/L)
 5    Q= surface water runoff (m/yr) (this is typically equal to precipitation -evapotranspiration
 6    Neco= nitrogen retention and denitrification by terrestrial catchment and nitrogen retention in the
 7    lake
 8          The term Neco could be derived multiple ways. The first is by taking the mean value
 9    calculated to represent the long-term amount of N an ecosystem can immobilize and denitrify
10    before leaching (i.e. N saturation) that is derived from the FAB model. This approach requires
1 1    the input of multiple ecosystem parameters. Its components are expressed by eq 2.

12                             Neco  = JNupt + Nret + (l - rlNmm + Nden )                       (2)

13    where,
14    Nupt= nitirogen uptake by the catchment
15    Nimm= nitrogen immobilization by the catchment soil
16    Nden=denitrification of nitrogen in the catchment,
17    Nret = in-lake retention of nitrogen
18    f =forest cover in the catchment (dimensionless parameter)
19    r = fraction lake/catchment ratio (dimensionless parameter)
20
21          The second approach for estimating Neco is to take the difference between N deposition
22    and measured N leaching in a catchment as expressed by eq 3.

23                                      Neco=DL(N)-Nleach                                (3)

24    N deposition is composed of NHX deposition (NHxdep) and NOy deposition. It is known that
25    NHxdep contributes to acidification, however the definition of NOX in the CAA does not include
26    NHX, and as such is not defined to provide protection from the acidifying effects of NHX.
27    Therefore, DLANciim(N)  is separated into NHX and NOy.
28                DL^^Ncy + S  = pL^^(N)-DL^^(mx) + DL^^(SOx)          (4)
      March  2010                               178                 Draft -Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1          Equation 1 and 4 will differ catchment by catchment because the acid-base balance of a
 2    catchment is a function of site-specific characteristics. However, for the standard it is desirable to
 3    calculate a deposition load for a specified ANC not for an individual catchment, but a larger
 4    population of catchments.  The site specific values from equation 1 can be used to derive such a
 5    deposition loading, here called the deposition metric, which represents a group or percentage of
 6    water bodies that reach a specified ANC (or higher). For example, if it is desired that all water
 7    bodies reach a specified ANC, the allowable amount of deposition for all water bodies is equal to
 8    the lowest value calculated from equation 1 for the population of water bodies. Because the
 9    deposition metric represents a percentage of individual catchments from a population of water
10    bodies, and not an individual catchment like DLANciim(S+N), the deposition metric is noted by
11    the follow abbreviation DLo/oEC0.
12          As an example of the above approach, we evaluate the population of 169 waterbodies in
13    the Adirondacks used in the REA analysis. For each individual waterbody in the population
14    DLANciim(S+N) at ANCum  = 50 was calculated using the two equations for deriving the Neco
15    term (eq 2 and 3). The distribution of deposition loads for the population was assessed and Table
16    5-5 shows the a few selected values for DLo/oECo. The mean value for DLo/oECo for the 169 water
17    bodies is presented, as well as the values for which 50, 75, 85, 95 and 100% of the water bodies
18    in the population will not exceed their critical load at ANC=50. Note, only 32% of water bodies
19    would not exceed their critical load at ANC=50 for the mean value DLo/oECo because  variability is
20    high in the data set. The deposition and atmospheric concentration tradeoff curves for DLo/oECO
21    equal to 32% and 50% are plotted in the  subsequent figures. The Administrator will  choose
22    which % of water bodies are projected to reach a targeted level of ANC as part of the overall
23    decision on the elements of the standard; this selection may be higher or lower than the examples
24    given here.
      March  2010                               179                Draft-Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
     Table 5-4. Example Calculations for Determining the Percent of Water Bodies Achieving Target
     ANC Levels
     This example is based the population of DLANciimfor and ANC=50 for 169 catchments in the Adirondacks.
     These catchments occur across on three categories of geologic sensitivity. We could separate the DLANciim
     values into sensitivity categories (if info is available) and do the analysis for each category or calculate one
     DLANciim for combined geologic categories. Units are in meq/m2/yr.

Mean
Stdev
Ster
Rank
%tile
50%
75%
85%
95%
100%
NHX
dep
20.40
3.22
0.25






Neco
(eq2)
19.19
3.03
0.23






DLo/oECO(S+N)
using Neco eq 2
162.36
162.92
13.04

99.33
65.62
54.89
45.12
30.22
Neco
(eq3)
63.95
11.15
0.86






DLo/oECO(S+N)
using Neco eq 3
207.55
165.42
13.24

139.22
110.37
95.53
83.99
59.07
% of lakes within
the population that
have ANC > 50
31.7%



50%
75%
85%
95%
100%
4
5
6
       The deposition tradeoff curves for N and S based on DLo/oECO at ANC=50 using the two
approaches for Neco and protective of 32 and 50% of the population of water bodies, are plotted
on Fig 5-8 and 5-9. The values for the maximum deposition values for N and S are given in
Table 5-5.
     Table 5-5. Values for N and S deposition tradeoff curves for ANC = 50, protecting 32 and 50%
     of the population, in Adirondacks case study area as illustrated on Fig 5-8 and Fig 5-9. Units are
     in meq/m2/yr unless noted otherwise.
%
protection
32
50
32
50

Eq2
Eq2
Eq3
Eq3
NHxdep
20.4
20.4
20.4
20.4
Neco
19.19
19.19
63.75
63.75
DLo/oECo
(max N)
162.36
99.33
207.5
139.22
DLo/oECo
(max S)
143.97
80.14
143.6
75.27
DLo/oECo
(max NOY)
141.96
78.9.3
187.15
118.82
     March 2010
                                           180
Draft -Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
1
2

3
                   200
                                                      ANC=50 & 32% lakes protected
                                                      ANC=50 & 50% lakes protected
                                                      Neco
                                                      NHx Deposition
                           Neco=19.19
                        0 NHxdep=20.40  5Q
                                       100           150

                                   N (meq/m2/yr)
          200
Figure 5-8. Tradeoff curve for S and N deposition to protect from aquatic
acidification in the Adirondacks using Neco equation 2.
4

5
6
                200
            -£ 150
             ^•»
            c\i


            |-100
                                                 ANC=50 & 32% lakes protected
                                                 ANC=50 & 50% lakes protected
                                                 Neco
                                                 NHx Deposition
                                        Max(S) =143.6
                                                                               Max(N) =207.5
                   0
                                               100
                                           N (meq/m2/yr)
                                                 150
      200
Figure 5-9. Tradeoff curve for S and N deposition to protect from aquatic
acidification in the Adirondacks using Neco equation 3.
     March 2010
                                     181
Draft -Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
 1
 2
 3
 4
 5
 6
 7
       As previously stated, it is known that NHX deposition (NHxdep) contributes to
acidification. However, the criteria pollutant listed by EPA pursuant to section 108 (a) of the Act
is oxides of nitrogen does not include NHX, and as such is not defined to provide protection from
the acidifying effects of NHX. Therefore, in order to represent the role of NHxdep as a component
of acidification it is subtracted from DLo/oECO(S+N).  The difference is the total allowable
deposition from NOy and SOX to protect a selected % of catchments in the population at a
selected level of ANC [DLo/oECO (S + NOy)] as expressed in equation 5.
                       DL%ECO (NOY +S) = DL%
                                                   %ECO
                                                              - NHX
                                                                    DEP
                        (5)
 9          The NOy and S deposition tradeoff curves for ANC =50, protecting 32 and 50% of the
10   water bodies, are presented in Table 5-6 and plotted on Fig 5-10 and 5-11. If NHX deposition is
11   greater than Neco, then Neco disappears from the tradeoff curve (i.e. Fig 5-11).

     Table 5-6. Values for NOy and S deposition tradeoff curves for ANC = 50, protecting 32 and
     50% of the population in Adirondacks case  study area as illustrated on Fig 5.10 and Fig 5.11.
     Units are in meq/m2/yr unless  noted otherwise.
%
protection
32
50
32
50

Eq2
Eq2
Eq3
Eq3
NHxdep
20.4
20.4
20.4
20.4
Neco (Noy)
Neco < NHxdep
Neco < NHxdep
43.35
43.35
DLmax(S)
141.96
78.93
143.6
75.27
DLmax(Noy)
141.96
78.93
187.15
118.82
12
     March  2010
                                           182
Draft -Do Not Quote or Cite

-------
          Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
1
2
3
               200
            c\i

            o-100
            C/3
                    0
                                                       ANC=50 & 32% lakes protected

                                                       ANC=50 & 50% lakes protected
                            50           100          150
                                    Noy (meq/m2/yr)
      200
       Figure 5-10. Tradeoff curve for S and NOy deposition to protect from aquatic
       acidification in the Adirondacks using Neco equation 2.
4
5
6
                200
                150
                100
                     0
                                                       ANC=50 & 32% lakes protected
                                                       ANC=50 & 50% lakes protected
                                                       Neco
                                  Max(S)=143.6
                             50          100         150
                                    NOy (meq/m2/yr)
                                                                        Max(Noy)=187.15
      200
       Figure 5-11. Tradeoff curve for S and NOy deposition to protect from aquatic
       acidification in the Adirondacks using Neco equation 3.

       The tradeoff curves for the atmospheric concentration of NOy and SOX are presented in

Fig 5-12 and 5-13. Deposition values for NOy and S (from Table 5-6, Fig 5-10 and 5-11) were

multiplied by the ratio of concentrations to depositions (previously referred to as aggregate
    March  2010
                                           183
Draft -Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
1    effective deposition velocitiesl?) for NOX and SOX (VSOX = 0.03824755 ng/m3/meq/m2 and

2    VNOX= 0.04386373 |j,g/m3/meq/m2). This is expressed in equation 5. These velocities were

3    calculated by taking the median value of the concentration of oxidized N to deposition of

4    oxidized N ratio in CMAQ for all grid cells over the Adirondack case study area.
                  [DL%ECO (N o J- Vnoy\+ [DL%ECO (s) • Vsox] = DL%ECO (N+S)- NHX
                                                                      VDEP
                                                                                                (6)
6
7
              CO
               E
               X
               o
               CO
                                                                ANC=50 & 32% lakes protected

                                                                ANC=50 & 50% lakes protected
                    0.00
                                                                  10.00
                                             Noy (ug/m3)
Figure 5-12. Tradeoff curve for atmospheric concentration of SOX and NOy to
protect from aquatic acidification in the Adirondacks using Neco equation 2.
     17 Note to reviewers:  in previous drafts we have referred to the ratios of deposition to concentration for NOy and
     SOx as "aggregate effective velocities." We are revisiting this choice of terms, as it is not as accurate a reflection of
     the parameter as we might prefer.  The concern with continuing to use the term "velocity" in this context is that it
     will be misinterpreted by the scientific community, and in order to avoid confusion, we will likely replace the term
     with "deposition ratio" or some other term that more accurately describes the parameter.
     March 2010
                                      184
Draft -Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
                    10.00
           ANC=50 & 32% lakes protected
           ANC=50 & 50% lakes protected
                 IE
                 ^)
                 X
                 O
                 C/5
                     0.00
                         0.00
                        10.00
                                             Noy (ug/rrr)
 1
 2          Figure 5-13. Tradeoff curve for atmospheric concentration of SOX and NOy to
 3          protect from aquatic acidification in the Adirondacks using Neco equation 3.
 4          5.5.2  Derivation of the Atmospheric Acidification Potential Index (AAPI):
 5          While the conceptual framework above provides a means for calculating tradeoff curves
 6    associated with a specific level of protection (indicated by a target ANC level)  and a specific
 7    percentage of ecosystems protected within an overall sensitive area, it does not provide a clearly
 8    integrated statement that can be expressed as a level such as would be needed for the secondary
 9    standard. The goal of this development of the AAPI is to create an index which can be applied
10    across the nation to convey the potential of an ecosystem to become acidified from atmospheric
11    deposition.
12          The definition of the AAPI form considered here is:
13          Annual Average AAPI: Natural background ANC minus the contribution to
14          acidifying deposition from NHX, minus the acidifying contribution of NOy and
15          SOX. This term is essentially a calculated ANC value that represents a percentage
16          of catchments in a population.
17          In order to derive the AAPI, we start with the basic framework of critical loads discussed
18    in the example above.
19          The approach used to  calculate N and S deposition values for a specified ANC at a
20    catchment-scale is expressed  in Equation 1. The deposition value for a specified ANC will vary
      March  2010
185
Draft -Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    from catchment to catchment based on how the properties that counterbalance the acidifying
 2    deposition vary among catchments. Equation 5 expresses how to calculate a deposition metric for
 3    a specified ANC for a population of waterbodies that could represent a national acid-sensitivity
 4    category. Moreover, the quantity of deposition equal to a specified ANC limit (i.e. critical load)
 5    will vary in eq 1 and 5 depending on the characteristics of the catchment or population of
 6    catchments, respectively. The goal for a secondary NOX and SOX NAAQS is to develop a form
 7    for the standards that allows us to set a single value for the standard across the U.S. To
 8    accomplish this, we rearrange equation (1) to solve for ANC (place ANC on the left hand side of
 9    the equation):
10                          Q. ANC^ = Neco + [BC]0-Q- [Dl(N] + DL(s)]                    (7)

                                                                                           (8)
12          In order to develop a form for the standard in which the level can be expressed as a single
13    national value related to protection against effects that occur at specific values of ANC, a
14    simplified version of equation (8) is:
15                                  ANC]im=g(-)-DL(N + S)                            (9)

16    where,  g()= sustainable flux of base cations from the ecosystem + ecological sinks of N. This
17    term is  equivalent to the pre-industrial ANC level, or the natural background ANC, expressed as:

18                                     g() = ~Neco+[BCl                               (10)

19           Building from equation 9, total nitrogen deposition is split into oxidized and reduced
20    nitrogen because we need to be able to specify the standards in terms of oxides of nitrogen, and
21    so the contribution of reduced nitrogen has to be separated.

22                        ANC^=g()~[DL(NOT) + DL(S)]~-D^NHX)                 (11)
      March  2010                               186                Draft-Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1   where,
 2   DL(Nox)= the deposit!onal load of oxidized nitrogen
 3   DL(NHX)= the deposit onal load of reduced nitrogen, NHX.
 4
 5          In order to judge whether an ecosystem or group of ecosystems meets the ANCimut given
 6   observed NOy and SOX levels, the associated depositional loadings of NOy and S can be
 7   compared directly against calculated deposition tradeoff curves (eq 4), atmospheric
 8   concentrations of NOy and SOX can be compared against the atmospheric concentration tradeoff
 9   curves (eq 5) or, loadings of NOX and SOX can be input into the following equations to obtain the
10   calculated value of ANC, equal to ANC*:

11                           ANC* = g(-) - [L(Noy) + l(SOx)] - L(NHx)                    (11)

12   where,
13   ANC*= the calculated value of ANC given loadings of N and S for comparison against an
14   ANCHmit.
15    L(NOX+S)= the load of NOX+S anions based on observed atmospheric concentrations of NOy and
16   SOX
17   L(NHX) = the load of reduced nitrogen deposition
18   [Note that L(N) = L(NOX+NHX)]
19
20          In equation 11, the ANC* will vary based on the deposition load inputs of Nox, NHX and
21   S at the site of interest. The deposition loads  caused by NOy and S and NHX are inputs, leading to

22                          ANC* = g~[L(Nax) + L(S)]~-DltNHx)                   (12)

23   If ANC* < ANCiim, then the deposition of N  and S exceeds the deposition load to maintain
24   ANCiimit. ANC* is still representative of the calculated ANC based on specific catchment level
25   estimates  of g, Q and NHX.
26          AAPI is equivalent to the equation for calculating ANC* when the catchment specific
27   values for g in equation (9) in Section 5.5.1. are replaced by representative values for acid
28   sensitive areas (based on a percentile of water bodies targeted for an ANC level selected by the

     March 2010                               187                Draft-Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    Administrator), Q and NHX are replaced by average values for aggregate ecosystem areas, and
 2    L(Nox) and L(S) are replaced by terms translating atmospheric NOy and SOX into deposition:
                                                                                        (13)
 4   where NOy and SOX are concentrations of NOy and SOX, respectively, VNOY and VSOX are the
 5   ratios of deposition to concentrations (deposition transformation ratios) for NOy and SOX,
 6   respectively.

 7   5.6   REFERENCES
 8   Davis, J. M., Swall, J. L., 2006. An examination of the CMAQ simulations of the wet deposition
 9          of ammonium from a Bayesian perspective. Atmospheric Environment 40, 4562-4573.

10   EPA, 1999.  Science Algorithms of the EPA Models-3 Community Multiscale Air Quality
1 1          (CMAQ) Modeling System. Tech. Rep. EPA/600/R-99/030, U.S. Environmental
12          Protection Agency, Washington DC.

13   Grantz, D., Garner, J., Johnson, D., 2003. Ecological effects of particulate matter. Environment
14          International 29, 213-239.

15   Lien L; Raddum GG; Fjellheim A. (1992). Critical loads for surface waters: invertebrates and
16          fish. (Acid rain research report no 21). Oslo, Norway: Norwegian Institute for Water
17          Research

18   Morris, R. E., McNally, D. E., Tesche, T. W., Tonnesen, G., Boylan, J. W., Brewer, P.,  2005.
19          Preliminary evaluation of the community multiscale air quality model for 2002 over the
20          Southeastern United States. Journal of the Air and Waste Management Association 55,
21          1694-1708.

22   Seinfeld, J., Pandis, S., 1998. Atmospheric Chemistry and Physics. John Wiley and Sons, Inc.,
23          New York.
     March 2010                              188                Draft-Do Not Quote or Cite

-------
          Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

1    Sullivan TJ; Webb JR; Snyder KU; Herlihy AT; Cosby BJ. (2007). Spatial distribution of acid-
2          sensitive and acid-impacted streams in relation to watershed features in the southern
3          Appalachian mountains. Water Air Soil Pollut, 182, 57-71.

4    Sullivan TJ; Fernandez U; Herlihy AT; Driscoll CT; McDonnell TC; Nowicki NA; Snyder KU;
5          Sutherland JW. (2006). Acid-base characteristics of soils in the Adirondack Mountains,
6          New York. Soil Sci Soc Am J, 70, 141-152.

7
    March  2010                               189                 Draft-Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1              6.     OPTIONS FOR ELEMENTS OF THE STANDARD
 2          The elements of the standard include the ambient air indicator, the form, the level and the
 3    averaging time. The "indicator" of a standard defines the chemical species or mixture of the
 4    criteria air pollutant that is to be measured in determining whether an area attains the standard.
 5    The "form" of a standard defines the air quality statistic that is to be compared to the level of the
 6    standard in determining whether an area attains the standard. The "averaging time" defines the
 7    period of time over which the air quality indicator is averaged, e.g. annual average. The "level"
 8    is the specific quantity to which the air quality statistic will be compared.
 9          EPA has historically established NAAQS so that the locally-monitored ambient
10    concentration of an air pollutant indicator is compared against a specified numerical level of
11    atmospheric concentration, using a specified averaging time and statistical form. For example,
12    the current secondary standard for oxides  of nitrogen uses ambient concentrations of NC>2 as the
13    indicator. Attainment is determined by comparing the annual arithmetic mean of the measured
14    maximum daily 1-hour NO2 concentrations, for a calendar year, against the level  of 0.053 ppm.
15    As discussed in Chapters 4 and 5, a standard using this kind of approach for defining indicator,
16    averaging time, form, and level is not the  most appropriate way to protect sensitive ecosystems
17    from effects associated with ambient concentrations of NOX and SOX. Moreover, the inherently
18    complex and variable linkages between ambient concentrations of NOX and SOX, their deposited
19    forms of nitrogen and sulfur, and the ecological responses that are associated with public welfare
20    effects call for consideration of a more complex and ecologically relevant design of the standard
21    that reflects these linkages.
22          Chapter 5 provided a conceptual framework for a secondary standard that is designed to
23    provide protection of ecosystems against the effects associated with deposition of ambient
24    concentrations of NOX and SOX. This conceptual framework takes into account variable factors,
25    such as atmospheric and ecosystem conditions that modify the amounts of deposited NOX and
26    SOX, and the  associated effects of deposited N and S on ecosystems. Based on the conceptual
27    framework described in Chapter 5, this chapter provides a set of potential options for specifying
28    the elements  of the framework to define a secondary standard for NOX and SOX. Our
29    development of options for the standards recognizes the need for a nationally applicable standard
30    for protection against adverse effects to public welfare, while recognizing the complex and
31    heterogeneous interactions between atmospheric concentrations of NOX and  SOX, deposition, and

      March 2010                               190                 Draft-Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1   ecological response. These options will include elements of the framework related to the air
 2   quality indicator, the averaging time, the form, and the level, which are based on the ecological
 3   indicator, the ecological response to deposition function, the deposition metric, and the
 4   atmospheric  deposition transformation function.
 5          To make the transition from the conceptual framework in Chapter 5, which is developed
 6   largely around the concept of critical loads, to elements of the standard, we propose to focus on
 7   developing a form of the standard that is based on the concepts of critical loads of NOX and SOX
 8   deposition linked to target ANC values, recognizing the limitations in available data and related
 9   uncertainties. Our goal in developing the form of the standard is to create an index, directly
10   expressed in  terms of atmospheric concentrations of NOy and SOX, that can be applied across the
11   nation to convey the potential  of an ecosystem to become acidified from atmospheric deposition.
12          This chapter is structured around questions related to the various elements of a standard.
13   The chapter begins in section 6.1 with a discussion of atmospheric indicators. Section 6.2 then
14   discusses averaging times for the atmospheric indicators. Section 6.3 suggests a possible
15   ecologically  relevant form of the standard. Section 6.4 provides a discussion of issues regarding
16   the spatial area over which a standard might be evaluated, and related issues regarding spatial
17   averaging within areas. Section 6.5 discusses options for specifying target levels for the
18   ecological indicator for aquatic acidification. Section 6.6 addresses issues relating to monitoring
19   of the atmospheric indicators.  Section 6.7 concludes with a discussion of potential ranges of
20   levels for the standard.

21   6.1    WHAT ATMOSPHERIC INDICATORS OF OXIDIZED NITROGEN
22          AND SULFUR ARE APPROPRIATE FOR USE IN A SECONDARY
23          NAAQS THAT PROVIDES PROTECTION FOR PUBLIC WELFARE
24          FROM EXPOSURE RELATED TO DEPOSITION OF N AND S?
25          WHAT AVERAGING TIMES AND STATISTICS FOR SUCH
26          INDICATORS ARE APPROPRIATE TO CONSIDER?
27          Staff concludes that indicators other than NC>2 and SC>2 should be considered as the
28   appropriate pollutant indicators for protection against the acidification effects associated with
29   deposition of NOX and SOX. This conclusion is based on the recognition that all forms of
     March 2010                              191                Draft-Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    oxidized nitrogen and sulfur in the atmosphere contribute to deposition and resulting
 2    acidification, and as such NO2 and 862 are incomplete indicators. Furthermore, staff concludes
 3    that NOy (total oxidized nitrogen) should be considered as an appropriate indicator for oxides of
 4    nitrogen. NOy is defined as NOX (NO and NO2) and all oxidized NOX products: including NO,
 5    NO2, and all other oxidized N-containing compounds transformed from NO and NO2 (Finlayson-
 6    Pitts and Pitts, 2000). As described in Chapter 4, this set of compounds includes NO2 + NO +
 7    HNO3 + PAN +2N2O5  + HONO+ NO3 + organic nitrates + paniculate NO3. Staff concludes that
 8    SOX should be considered as an appropriate indicator for oxides of sulfur. SOX includes sulfur
 9    monoxide (SO), sulfur dioxide,  sulfur trioxide (SO3), and disulfur monoxide (S2O), and
10    particulate-phase S compounds  that result from gas-phase  sulfur oxides interacting with particles.
11          In principle, measured NOy based on catalytic conversion of all oxidized species to NO
12    followed by chemiluminescence NO detection is consistent with this definition. We recognize
13    the caveats associated with instrument conversion efficiency and possible inlet losses which are
14    discussed in Section 5.6. The development of the function that converts atmospheric
15    concentrations of NOy  and SOX  to N and S deposition which incorporates NOy estimates is based
16    on the Community Multi-scale Air Quality (CMAQ) model (EPA, 1999). CMAQ treats the
17    dominant NOy species  as explicit species while the minor contributing non-PAN organic
18    nitrogen compounds are aggregated. Total oxidized sulfur, SOX, requires independent
19    measurements of particle bound sulfate and gaseous sulfur dioxide; methodology and network
20    considerations are discussed in  Section 5.6. The CMAQ treatment of SOX is the simple addition
21    of both species which are treated explicitly in the model formulation. All particle size fractions
22    are included in the CMAQ SOX  estimates. At this time, we consider the contribution of coarse
23    fraction (aerodynamic  diameters between 2.5 and 10 microns) particle bound sulfate to be
24    insignificant from a measurement perspective. Consequently, the routinely measured sulfate
25    from IMPROVE and EPA speciation networks, as well as  CASTNET, are viable candidates for
26    measurement consideration. Consistent with units and the  charge balance relationships applied in
27    ecosystem acidification models, only mass as sulfur or nitrogen is considered requiring
28    conversion of reported particle bound sulfate and nitrate. Precipitation mass is not included
29    explicitly as part of an  atmospheric NAAQS indicator.
      March  2010                               192                 Draft-Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    6.2   WHAT IS THE APPROPRIATE AVERAGING TIME FOR THE AIR
 2          QUALITY INDICATORS NOY AND SOX TO PROVIDE
 3          PROTECTION OF PUBLIC WELFARE FROM ADVERSE EFFECTS
 4          FROM ACIDIFICATION?
 5          Based on the review of the scientific evidence, welfare effects associated with
 6    acidification result from annual cumulative deposition of nitrogen and sulfur, reflected in effects
 7    on the chronic ANC level (measured as annual ANC). It is important to note that chemical
 8    changes can occur over both long- and short-term timescales.  Short-term (i.e., hours or days)
 9    episodic changes in water chemistry can also have significant biological effects. Episodic
10    chemistry refers to conditions during precipitation or snowmelt events when proportionately
11    more drainage water is routed through upper soil horizons that tend to provide less acid
12    neutralizing than was passing through deeper soil horizons. Surface water chemistry has lower
13    pH and acid neutralizing capacity (ANC) during events than during baseflow conditions. One of
14    the most important effects of acidifying deposition on surface water chemistry is the short-term
15    change in chemistry that is termed "episodic acidification." Some streams may have chronic or
16    base flow chemistry that is suitable for aquatic biota, but may be subject to occasional acidic
17    episodes with lethal consequences. Episodic declines in pH and ANC are nearly ubiquitous in
18    drainage waters throughout the eastern United States and are caused partly by acidifying
19    deposition and partly by natural processes. As noted in Chapter 3 of the ISA, while ecosystems
20    are also affected by episodic increases in acidity due to pulses of acidity during high rainfall
21    periods and snowmelts, protection against these episodic acidity events can be achieved by
22    establishing a higher chronic ANC level. Episodic acidification can result from either shorter
23    term deposition episodes, or from longer term deposition on snowpack. Snowmelt can release
24    stored N deposited throughout the winter, leading to episodic acidification in the absence of
25    increased  deposition during the actual episodic acidification event. Protection against a low
26    chronic ANC level is provided by reducing overall annual average deposition levels for nitrogen
27    and sulfur. This supports the conclusion that long term NOX and SOX concentrations are
28    appropriate to provide protection against low chronic ANC levels, which protects against both
29    long term acidification and acute acidic episodes.
     March 2010                              193                Draft-Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1          Long term concentrations are often measured using annual averages. However, given the
 2   multi-year nature of responses to chronic acidification, multi-year averages of concentrations of
 3   NOy and SOX may also be appropriate. In the second draft policy assessment, we will provide an
 4   expanded discussion of the support for different options for the averaging time to best represent
 5   long-term concentrations of NOy and SOX related to chronic acidification.

 6   6.3    WHAT FORM(S) OF  THE  STANDARD ARE MOST APPROPRIATE
 7          TO PROVIDE PROTECTION OF SENSITIVE ECOSYSTEMS
 8          FROM THE EFFECTS OF ACIDIFYING DEPOSITION RELATED
 9          TO AMBIENT NOX AND SOX CONCENTRATIONS?
10          Based on the evidence for joint effects of NOX and SOX through acidifying deposition,
11   staff concludes that it is appropriate to consider changes to the form of the existing NOX and SOX
12   secondary standards to provide protection to ecosystems. Staff notes that in recent reviews of the
13   secondary ozone standards, EPA has considered use of a form of the standard that reflects
14   ecologically relevant exposures, by using a cumulative index which weights exposures at higher
15   concentrations greater than those at lower concentrations based on scientific literature
16   demonstrating the cumulative nature of (Vinduced plant effects and the need to give greater
17   weight to higher concentrations (EPA, 2007). See 75 FR 2938, 2999 (Janaury 19, 2010) In order
18   to recognize the roles that NOX and SOX play in acidification based on their acidifying potentials,
19   and to incorporate the important roles that reduced nitrogen and non-atmospheric variables play
20   in determining the acidifying potentials of NOX and SOX, staff suggests using an Atmospheric
21   Acidification Potential Index (AAPI) that is a more ecologically relevant form relative to the
22   current ambient concentration based forms, based on the derivations in Section 5.5.1. The intent
23   of the AAPI is in effect to weight atmospheric concentrations of NOX and SOX by their
24   propensity to contribute to acidification through deposition, given the fundamental acidifying
25   potential of each pollutant, and the  ecological factors that govern acid sensitivity in different
26   ecosystems. Thus the APPI is more relevant to protecting ecosystems from acidifying deposition
27   compared to simple ambient concentration forms which do not reflect factors that affect
28   acidifying potential.
29          The AAPI is closely tied to  the ecological indicator of acidification, ANC, so that the
30   form of AAPI is intended to identify the atmospheric concentrations of NOX and SOX that will

     March 2010                              194                Draft-Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    result in an equivalent level of a target ANC for a percentage of aquatic ecosystems within a
 2    particular acid sensitive area. Thus, this form is ecologically relevant as it is tied directly to the
 3    ecological indicator that is most directly linked with known ecological effects.
 4           The AAPI incorporates the processes which modify both rates of deposition and
 5    ecological response to deposition caused by NOX and SOX. There is strong evidence in the
 6    scientific literature demonstrating that the amount of deposition caused by NOX and SOX is
 7    modified by atmospheric and landscape factors. Within the ecosystem there are factors, such as
 8    bedrock geology and topography, which modify  the acidifying potential of the nitrogen and
 9    sulfur deposition resulting from  ambient NOX and SOX concentrations.  In addition reduced
10    nitrogen contributes to total nitrogen loading. In  this review, reduced nitrogen is treated as an
11    additional  modifying factor within the ecosystem, which reduces the buffering capacity of the
12    ecosystem, and therefore it increases the impact or sensitivity to additional loading from oxidized
13    forms of nitrogen. In effect this leaves less allowable deposition loading from NOX and SOX
14    before the  ecosystem fails to achieve a target ANC level. Based on this evidence staff concludes
15    that the form should include landscape and atmospheric factors, including reduced nitrogen,
16    which modify the acidifying potential of ambient NOX and SOX concentrations. This form is
17    consistent  with the language of the CAA as discussed in Section 1.5.
18           Selecting a more ecologically-relevant secondary standard form would also be directly
19    responsive to the recommendation of the 2004 National Research Council's report titled Air
20    Quality Management in the United States (NRC, 2004) which encourages the Agency to evaluate
21    its historic practice of setting the secondary NAAQS equal to the primary.
22           In theory, the AAPI could address acidification potential related to both terrestrial
23    acidification and aquatic acidification. For this first draft policy assessment, as discussed in
24    Chapter 5, we define the AAPI for protection against aquatic acidification. In the second draft
25    policy assessment, we will explore the potential to include protection against terrestrial
26    acidification in the AAPI or a related index.
27           The definition of the AAPI form considered here is:
28           Annual Average AAPI: Natural  background ANC minus the contribution to
29           acidifying deposition from NHX, minus the acidifying contribution of deposition
30           from NOy and SOX.
31           Building from the derivation of ANC* provided in Section 5.5.2, the AAPI is equivalent
32    to the equation for calculating ANC* when the catchment specific values for g in equation (9) in

      March 2010                                195                 Draft-Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    Section 5.5.2. are replaced by representative values for acid sensitive areas (based on a percentile
 2    of water bodies targeted for an ANC level selected by the Administrator), Q and NHX are
 3    replaced by average values for aggregate ecosystem areas, and L(N0x) and L(S) are replaced by
 4    terms translating atmospheric NOy and SOX into deposition:
                                                                                           (1)
 6    where NOy and SOX are concentrations of NOy and SOX, respectively, VNOY and VSOX are the
 7    ratios of deposition to concentrations (deposition transformation ratios) for NOy and SOX,
 8    respectively. Deposition transformation ratios are the estimated relationships between
 9    atmospheric concentrations of NOy and SOX and the collocated deposition of Nox and S.  See
10    Chapter 5.4.4 and 5.4.5 for further description of calculation of ratios of deposition to
1 1    concentrations.
12          Note that while equation (1) is used to calculate the value of AAPI for any  observed
13    values of NOy and SOX, the level of the standard for AAPI selected by the administrator should
14    reflect a wide number of factors, including desired level of protection indicated by a target
15    ANCumit, the specified percentile of waterbodies projected to achieve the target ANC, and the
16    various factors and uncertainties involved in specifying all of the other aspects of the standard,
17    such as the classification of landscape areas, the specification of reduced nitrogen deposition, the
18    methodology to determine deposition of NOy and SOX, and the averaging time. As such the
19    administrator may choose an AAPI level higher or lower than the target ANCumit to reflect the
20    combined effect of the all of the components of the standard and their related uncertainty, such
21    that the chosen AAPI, in the context of the overall standard,  reflects her informed judgment as to
22    a standard that is sufficient but not more than necessary to protect against adverse public welfare
23    effects.
24          How are AAPI parameters determined?
25          Other than ambient levels of NOX and SOX, which would be measured values, EPA would
26    determine and specify all of the values for the AAPI parameters, as discussed below.
27          The natural background ANC, g, is a calculated value and is determined by two
28    components: [BC]o* which is closely associated with underlying bedrock which strongly
29    influences the contribution of base cations due to weathering, for which a representative value

      March  2010                                196                Draft -Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    could be determined for a limited set of geologic acid sensitivity classes, and Neco, which
 2    represents the amount of deposited nitrogen that is available for acidification due to uptake,
 3    denirification and immobilization. Neco is estimated using two different approaches: (1) the
 4    individual terms are estimated through available data and modeling or (2) Neco is calculated as
 5    nitrogen deposited minus nitrogen leached, using streamwater measurements of nitrate for
 6    leaching and estimates of nitrogen deposition based on model results and measurements. The
 7    details of these procedures are addressed in chapter 4 and Appendix 4 of the REA.
 8          The runoff parameter Q for each acid sensitive area is determined based on USGS
 9    mapping of runoff values (REFERENCE NEEDED).
10          VNO?, VSOX are calculated from CMAQ by dividing the annual average  NOy, SOX
11    concentration by the total NOy or SOX deposition, respectively, for each grid cell and then
12    aggregating all grid cells in the  acid sensitive area.
13          L(NHX) is calculated using the same procedures applied to CMAQ results for deposited
14    NHX.
15          The VNOY and VSOX are spatially variable, and for the purposes of setting the standard,
16    are determined based on the ratios of total  sulfur and nitrogen depositions to concentrations from
17    CMAQ model outputs  (see Chapter 5 for details of calculation of deposition ratios). V^oy, VSOX
18    are calculated from CMAQ by dividing the annual average NOy or SOX concentration by the total
19    NOy or SOX deposition, respectively, for each grid cell and then computing the  mean or median
20    of all grid cells in the acid sensitive area (the decision on whether the median or mean value
21    should be used is an option for discussion; the mean will give more weight to outlier values
22    relative to the median).
23          NHX is spatially variabe and determined based on monitored and/or CMAQ modeled
24    outputs. The average NHX deposition across grid cells within an acid sensitive region will be
25    used to represent the deposit!onal load of NHX.
26          There will be multiple combinations of concentrations of NOX and SOX that result in a
27    specific value of the AAPI. There will be no single combination of NOX and SOX that solves for a
28    particular value of AAPI in all locations, easured concentrations of annual average NOX and SOX
29    necessary to meet the standards are thus expressed conditionally by the equality in (1), and not
30    by fixed quantities.
      March  2010                               197                 Draft-Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1           In order to provide a set of values for elements of the form, e.g. to develop a specific set
 2    of parameter values for g, VNOY, Vs, and NX, we propose to classify locations in the U.S. into a
 3    set of areas based on sensitivity to acidification. Each area would be assigned a classification for
 4    the g parameter; for example, as described in Section 5.2.2.4, a set of classes of acidification
 5    sensitivity might be able to be developed based on underlying bedrock geology, or bedrock
 6    geology plus other ecosystem variables.  The g parameter (natural background, or preindustrial,
 7    ANC) would then be estimated for each  of those sensitivity classes, based on the critical load
 8    modeling available for each class. Each acid sensitive area would then be assigned a value of g
 9    based on the geology class in which it falls. In the case of VNOY, Vs, and NHX, values for specific
10    areas would be estimated based on the best available monitoring and/or modeling data. Given the
11    limited availability of measured deposition velocities, staff concludes that the calculated
12    deposition ratios based on the CMAQ modeling from 2005 provides the best available source of
13    estimates of VNOY and Vs. Evaluation of the stability of these estimates of deposition ratios over
14    time (see Chapter 5) suggests that in most acid sensiive areas, deposition ratios are quite stable,
15    with a coefficient of variation less than 25 percent across a four year period. While there are a
16    limited number of sites that directly measure deposition of reduced nitrogen, staff concludes that
17    the most widely available and defensible estimates of reduced nitrogen deposition (NHX) are the
18    estimates obtained from the CMAQ modeling from 2005.18
19           It is important to note for this form of the standard that the same AAPI can be obtained
20    with differentcombinations of ambient NOX and SOX concentrations. The implication of the form
21    of the standard expressed in equation (1) is that there will  be a tradeoff curve that reflects the
22    combinations of NOX and SOX that satisfy equation (1) for any specific value of the standard.  The
23    shape of the tradeoff curve will depend on the specific values of G, VNOY, Vs, and NHX for a
24    limited number of specific areas classified based on acid-sensitivity. As discussed in  Chapter 5,
25    all parts of the U.S.  would be classified into areas based on acid-sensitivity.  Within each such
26    area, EPA would specify the parameter values of APPI, leading to a specific tradeoff curve for
27    each area. The levels of NOy and SOX that meet an AAPI standard expressed for a given
28    g() [preindustrial ANC], Q, L(NHX) and VNOy and VSOx:
      18 Note to readers: Maps of CMAQ 2005 estimates of NHx deposition will be included in the second draft policy
      assessment, along with an evaluation of the representativeness of the 2005 NHx deposition for characterizing
      conditions over a multiyear period.

      March 2010                               198                 Draft-Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
12
 1                       VNOy • NOy + V^-ST = g() • Q - L(NHx) - AAPI • Q                 (2)

 2   Note that \^NOy • NOy + VST • ST\ is essentially the critical load of NOy and S, expressed in terms
 3   of atmospheric concentrations. As such, equation (2) can also be expressed in a form similar to a
 4   typical critical load equation as discussed in Chapter 5, e.g.
 5                     VNOy • NOy + VST-ST = (BC*0 - APPIjQ - Neco - L(NHx)               (3)

 6   This expression is based on

 7                              g() = (BC* +)- L(NHx) - AAPI • Q                        (4)
 8   The pairs of NOy and SOX that will meet a given AAPI limit are related through the following
 9   equations

10                                     NO;=Cmm(NOy)                                (5)

11                             SO* =C   (SOx)\/NO  < C  (NO )                       (6)
                                 x    max \    /     y    min \   y /                       \ /
                                 (Cmm(NOy)-CmK(NOy))
•NO*yVNOy>Cmm(NOy)       (7)
13   where,
14   NO*y is the coordinate point for NOy
15   SO*X is the coordinate point for SOX
16   Cmax (SOx) is the concentration of SOX in the atmosphere consistent with DLmax (S)
17   Cmax (NOy) is the concentration of NOy in the atmosphere consistent with DL max (N)
18   Cmm (NO ) is the concentration of NOy in the atmosphere consistent with DL min (N)
19                                  Cmax (SOx) = — DLmax (S)                            (8)
     March 2010                             199                Draft-Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
                         Cmm (NOY ) = -t—DL^(N - NHx)\/NHx < DLmm (N)                 (9)
                                       wo.
                                        =OVNHX>DLmm(N)

                                                                                          (10)
 4    where DLmax(S), DLmax (N), and DLmin(N).are based on the critical load within a sensitive
 5    areas that protects a specified percentile (e.g. 95 °) of water bodies in the area.
 6          Note that Cmin(NOy) is a conditional function determined by the relationship between
 7    total nitrogen buffering capacity in an ecosystem and the amount of reduced nitrogen deposition.
 8    When reduced nitrogen deposition exceeds the buffering capacity of an ecosystem, then all
 9    atmospheric oxidized nitrogen contributes to acidification. When reduced nitrogen deposition is
10    less than the buffering  capacity  of an ecosystem, then some amount of NOy is buffered (i.e. is
1 1    reflected in Cmin(NOy)  but that amount reflects the contribution of NHX to total nitrogen (the
12    amount of buffering capacity used up by reduced nitrogen). In this case, some fraction of the
13    atmospheric oxidized nitrogen may not contribute to acidification.
14          Recall that these three variables are conditional on the chosen level of APPI, and reflect
15    the deposit! onal loadings that are associated with an equivalent level of ANC, e.g. for an APPI  of
16    50, the DLmax(S), DLmax(N), and DLm;n(N) are associated with an ANC of 50. Also recall than
17    DLmax(S) for a given ANC is a function of the "natural" flux of base cations to a watershed,
18    runoff, and the amount of sulfur retention within a waterbody; DLm;n(N) is the minimum amount
19    of deposition of total nitrogen (NHX + NOX) that catchment processes can effectively remove
20    without contributing to the acidic balance; and DLmax(N) for a given ANC is a function of
21    DLm;n(N) and the "natural" flux of base cations to a watershed, runoff, and the amount of
22    nitrogen retention within a waterbody, assuming S is zero. In our framework, DLm;n(N) is
23    calculated from the FAB critical load modeling (equation 5 from Attachment A of the REA) or
24    estimated through measured or modeled values of total nitrogen deposition and nitrate leaching.
25          As discussed in Chapter 5, the specific estimation of G, VNOY, Vs, and NHxin a specific
26    sensitive area will depend on the spatial scale of the sensitive area. Sensitivity can be assessed at
27    the level of individual catchments, however, this presents practical limitations for establishing

      March 2010                               200                Draft -Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    meaningful standards, as there are thousands of catchments within the U.S. Binning classes of
 2    sensitivity within larger spatial areas, e.g. the sensitive ecosystem areas displayed in Figure 6-1
 3    (reproducing Figure 4.2-2 in the REA), can provide a more manageable set of values of G, VNoy,
 4    Vs, and NHX. These parameters can be estimated in several ways for the larger spatial areas.
 5    Mean or median values can be generated across catchments, however, this would lead to
 6    parameter estimates that do not reflect conditions in the more sensitive lakes in the region.
 7    Alternatively, in order to provide a desired level of protection in these larger defined spatial
 8    areas, estimates based on higher percentiles of the distributions of parameters across catchments
 9    can be generated, e.g. the 75th or 95th percentile values of G, VNOY, Vs, and NHxr could be used to
10    provide protection for the more vulnerable aquatic ecosystems, however this would potentially
11    lead to over-protection for less vulnerable ecosystems in the area. The Administrator may
12    consider the balance between protection of particularly sensitive ecosystems and the overall
13    protection for ecosystems in an area as an important element to consider in making decisions
14    about the target level of ANC and the percent of aquatic ecosystems within an area targeted to
15    achieve the specified ANC level. One potentially important modification to this process would
16    be to first remove water bodies that are naturally acidic (e.g. that will  not benefit from reductions
17    in atmospheric NOX and SOX deposition) from the distribution of water bodies in the area prior to
18    determining the mean or 95th percentile. This will increase the likelihood that the estimated g
19    parameter will be representative of ecosystems within an area that are sensitive to NOX and SOX
20    deposition. The second draft policy assessment will explore the implications of alternative
21    combinations of target ANC and percent of aquatic ecosystems protected at the target ANC in
22    areas of different sizes. The second draft policy assessment will also explore methods for
23    determining values of g for areas that are clearly not sensitive to acidification from deposition of
24    NOX and  SOX. These areas may be areas that have very high levels of natural buffering, or may
25    also be areas that are naturally acidified, such that the value of g is less than the target value of
26    ANC. In these naturally acidified areas, reducing deposition from NOX and SOX will not be
27    beneficial, because the areas are adapted to high levels of acidity.
      March 2010                                201                Draft-Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
1
2
3
4
5
6
                                                          /lAppalacfiian
                                                             Plateau
Figure 6-1. Ecosystems sensitive to acidifying deposition in the Eastern U.S.
(Note that Florida represents a special case where high levels of natural
acidification exist unrelated to deposition) This map does not include all sensitive
areas in the U.S. Certain mountainous areas of the Western U.S. are also sensitive
to acidifying deposition.
     March 2010
                                     202
Draft -Do Not Quote or Cite

-------
          Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1   6.4   WHAT ARE THE APPROPRIATE SPATIAL EXTENTS OF THE
 2         BOUNDARIES FOR EVALUATING AAPI? WITHIN THOSE
 3         BOUNDARIES, WHAT ARE THE APPROPRIATE STATISTICS TO
 4         USE IN CALCULATING THE PARAMETERS OF THE AAPI, E.G.
 5         G, VNOY, Vs, AND NHX? WITHIN THOSE BOUNDARIES, WHAT S
 6         THE APPROPRIATE SPATIAL AVERAGING FOR THE AIR
 7         QUALITY INDICATORS NOY AND SOX TO PROVIDE
 8         PROTECTION OF PUBLIC WELFARE FROM ADVERSE EFFECTS
 9         FROM ACIDIFICATION?
10         [Note to reviewers: This section will be added in the second draft policy assessment. In
11   the second draft we plan to provide initial sets ofparameteir values for acid sensitive areas of
12   the U.S., and include an exploration of how the standard might be specified for areas of the U.S.
13   that are not sensitive to deposition o/"NOx and SOX. In addition, we plan to discuss the
14   correlation between the extent of a spatial area and the importance of evaluating alternative
15   percentiles of critical loads to protect a percentage of water bodies in an area, and to discuss
16   how averaging of the VNOY,  FSOX, andNHx should be conducted to best represent the
17   parameters for an area.]

18   6.5   WHAT ARE THE OPTIONS FOR SPECIFYING THE TARGETS
19         FOR THE ECOLOGICAL INDICATOR FOR AQUATIC
20         ACIDIFICATION?
21         Chapter 5 discusses the rationale for use of ANC as the ecological indicator best suited to
22   reflect the sensitivity of aquatic ecosystems to acidification. ANC as an indicator of acidification
23   is causally linked to a number of measures of adversity to ecosystems, including declines in fish
24   populations and diversity of aquatic species. ANC is also causally linked with deposition of
25   nitrogen and sulfur. ANC is thus ideally  suited to serve as the bridge between deposition and
26   ecological effects. As such, staff concludes that ANC is the best available choice as the
27   ecological indicator. CASAC has agreed that ANC represents a suitable ecological indicator for
28   aquatic acidification (EPA-CASAC-09-013). Results from the REA confirm that ANC may be


     March 2010                           203              Draft-Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    used to establish impacts from current depositional loadings (REA 4.2.6). As explained above,
 2    ANC is an indicator of the effects expected to occur given the natural buffering capacity of an
 3    ecosystem and the loadings of nitrogen and sulfur resulting from atmospheric deposition. A
 4    target ANC limit based on a desired level of protection is an important input to the decisions of
 5    the level of AAPI and the percent of ecosystems to be protected.

 6          6.5.1   What levels of impairment are related to alternative levels of ANC?
 7          As discussed in Chapters 2, 3, and 4, specific levels of ANC are associated with differing
 8    levels of ecosystem impairment, with higher levels of ANC resulting in fewer ecosystem
 9    impacts, and lower levels resulting in both higher intensity of impacts and a broader set of
10    impacts. Logistic regression of species presence/absence data against ANC provides a
11    quantitative dose-response function, which indicates the probability of occurrence of an
12    organism for a given value of ANC. For example, the number offish species present in a
13    waterbody has been shown to be positively correlated with the ANC level in the water, with
14    higher values supporting a greater richness and diversity offish species (Figure 6-2). The
15    diversity and distribution of phyto-zooplankton communities are also positively correlated with
16    ANC.
17          The relationship between ANC and ecosystem impacts is non-linear, with a sigmoidal
18    shape. For freshwater systems, ANC levels can be grouped into five major classes: <0, 0-20, 20-
19    50, 50-100, and >100 microequivalents per liter (ueq/L), with each range representing a
20    probability of ecological damage to the community. The five categories of ANC and expected
21    ecological effects are  described Table 2-1 in Chapter 2 and are supported by a large body of
22    research completed throughout the eastern United States (Sullivan et al., 2006).
23          Biota are generally not harmed when ANC values are >100 microequivalents per liter
24    (ueq/L). The number offish species also peaks at ANC values >100 ueq/L. This suggests that at
25    ANC greater than 100, little risk from acidification exists in most aquatic ecosystems. At ANC
26    levels below 100 ueq/L, overall health of an aquatic community can be maintained; however,
27    fish fitness and community diversity begin to decline. At ANC levels between 100 and 50 ueq/L,
28    the fitness of sensitive species (e.g., brook trout, zooplankton) also begins to decline. When ANC
29    concentrations are <50 ueq/L, negative effects on aquatic biota are observed, including large
30    reductions in diversity offish species, and changes in health offish populations, affecting


      March 2010                               204                 Draft-Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    reproductive ability and fitness. ANC levels below 50 are generally associated with death or loss
 2    of fitness of biota that are sensitive to acidification. (ISA 5.2.2.1 and REA 5.2.1.2).
 3          Based on the field data from the Adirondacks and  Shenendoah case study areas, ANC
 4    levels less than 50 are clearly adverse to ecosystem health, and are likely to lead to reductions in
 5    ecosystem services related to recreational fishing. ANC levels between 50 and 100 are
 6    potentially adverse to ecosystem health, and may result in losses in ecosystem services, but the
 7    effects are less severe and greater uncertainty exists as to the magnitude of ecosystem service
 8    impacts. A more comprehensive discussion of uncertainties related to ecological effects at
 9    different ANC levels and related ecosystem services will be included in the second draft policy
10    assessment.
11          The implications of the data from the Adirondacks and Shenendoah case study areas for
12    relating ANC to adverse ecological impacts is transferable to other acid sensitive areas of the
13    U.S. The relationship between species diversity and ANC is quite similar between the two case
14    study areas (see REA Figure 4.2-1), which have different water body types and different
15    geological and topographical features. While the species composition and thereby relative
16    sensitivities of species are likely to vary across the landscape, the rate of impact is likely to be
17    similar. The plot in Figure 6-2 shows a rapid decrease in fish species between an ANC of 100
18    and an ANC of 0. This trend is what would be expected in many systems given similar changes
19    in ANC.
20          Consideration of the appropriate levels of ANC to target in the standard to reduce the
21    likelihood of effects from aquatic acidification can be based upon the above presented categories
22    of aquatic status in Table 2-1. Using this information as well as information provided by both the
23    ISA and REA, the lowest two categories (0 and 0<20) would appear inadequate to protect against
24    catastrophic loss of ecosystem function. While ecological effects occur at ANC levels below 50,
25    the degree and nature of those effects is less significant than at levels below 20. Therefore, three
26    levels of ANC - 20,  50, and 100 - would provide the  Administrator with reasonable range of
27    options in designing an AAPI for protecting public welfare.
28          Given the level of ecosystem impairment occurring at ANC levels below 50, staff suggest
29    that the greatest support is for the Administrator to consider a range for the target ANC between
30    50 and 100 as a basis for the design of the  standard. Selection of target ANC values closer to 50
31    places less weight on the vulnerability of sensitive aquatic ecosystems, while selection of target

      March 2010                               205                Draft-Do Not Quote or Cite

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
      Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

ANC values closer to 100 places more weight on sensitive species within acid sensitive
ecosystems. Staff conclude that while target ANC values between 20 and 50 will not result in
complete impairment of aquatic systems, the level of damages due to ANC as you get lower in
this range are highly likely to result in adverse impacts to public welfare in many locations, due
to the significant reductions in the number offish species in affected waterbodies, and the
reductions in health and reproductive  fitness offish populations and other aquatic organisms.
                              Severe   Elevated   Moderate
                 14
                                              .
                                                .......         .
                    -200   -100
                                 0      100    200    300    400
                                   ANC(ueq/L)
            500
       Figure 6-2. Number offish species per lake or stream versus ANC level and
                           19
       aquatic status category   (colored regions) for lakes in the Adirondack Case Study
       Area (Sullivan et al., 2006).
       The target ANC level specified in designing the standard is only one part in determining
the overall protectiveness of the standard. The degree of protectiveness is based on all elements
of the standard, including the target ANC, the size of the spatial areas over which the standard is
applied, the percent of aquatic ecosystems targeted within a spatial area that is selected by the
Administrator to achieve the selected ANC level, the atmospheric indicator, the method for
calculating g, the calculated values for the deposition transformation ratios (VNOX and VSOX),
     19 The aquatic status categories are based on the literature and are discussed in detail in the REA (REA Appendix 4-
     20)
     March 2010
                                         206
Draft -Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    and the calculated value for reduced nitrogen deposition (NHX). There are widely varying
 2    degrees of uncertainty associated with all of these elements, some being much more certain and
 3    others being much less certain. The specified target ANC level is a crucial part of developing a
 4    standard that is requisite to protect, but it is the overall design and content of the standard that
 5    must be considered in judging the adequacy of protection it provides.
 6          Consideration of the target ANC should also reflect that an adequate level of ANC should
 7    protect against episodic as well as long term effects. Selecting a higher chronic ANC level can
 8    provide greater protection against short term peaks in acidification. In addition,  selection of ANC
 9    values in the range of 20 to 50 provides less protection against these short term episodic effects.
10    Selection of target ANC values in the range from 50 to  100 provides additional protection
11    against episodic peaks in acidification.
12          When considering the appropriate level of a standard to protect against aquatic
13    acidification, it is necessary to take into account both the time period desired for recovery as well
14    as the potential of recovery. Ecosystems become adversely impacted by acidifying deposition
15    over long periods  of time and have variable time frames and abilities to recover from such
16    perturbations. Modeling presented in the REA (REA Section 4.2.4) shows the estimated ANC
17    values for Adirondack lakes and Shenandoah streams under pre-acidification conditions and
18    indicates that for a small percentage of lakes and streams, natural ANC levels would have been
19    below 50. Therefore,  for these waterbodies, no reduction in input is likely to achieve an ANC of
20    50 or greater. Conversely, for some lakes and streams the level of perturbation from long periods
21    of acidifying deposition has resulted in very low ANC values compared to estimated natural
22    conditions. For such waterbodies, the time to recovery would be largely dependent on future
23    inputs of acidifying deposition. These concepts become important in the consideration of the
24    desired level of protection of a standard and will be discussed further in the next draft of this
25    document.
      March 2010                               207                Draft-Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    6.6   WHAT ARE THE APPROPRIATE AMBIENT AIR MONITORING
 2          METHODS TO CONSIDER IN DEVELOPING THE STANDARDS?

 3          6.6.1   What measurements would be used to characterize NOy and SOX ambient air
 4                 concentrations for the purposes of the AAPI based standard?
 5          Ambient NOy, gaseous 862 and particulate sulfate concentrations would be used in
 6    determining compliance with the AAPI. This would require measurements of NOy, sulfate and
 7    sulfur dioxide, all which are conducted as part of current routine monitoring networks (section
 8    3.2). There are issues requiring resolution associated with Federal Reference or Equivalency
 9    Measurement (FRM/FEM) status of measurement techniques, that to date have served as
10    supplemental information, which will require resolution. A FRM for SC>2 exists, but not for NOy
11    or sulfate. Only recently have NOy measurements, which historically were viewed as research
12    venue measurements, been incorporated as "routine" observations, partly as a result of the NCore
13    program. Acquiring FRM status may require better characterization of the conversion
14    efficiencies, mass loss and clear guidance on operating and siting procedures. Particulate sulfate
15    has been measured for several years in the IMPROVE, CASTNET and EPA CSN networks. The
16    nation has over 500 24-hour average, every third day sulfate measurements produced by the
17    PM2.5 speciation networks (IMPROVE and EPA CSN) and nearly 80 CASTNET sites that
18    provide continuous weekly average samples of sulfate with an open inlet accommodating all
19    particle sizes. However, with minor exceptions, the PM2.5 fraction accounts for nearly all sulfate
20    mass. The sample collection period is not an issue for gaseous measurements of NOy and SO2
21    that operate continuously. Some concerns have been raised about the possibility of exclusion of
22    coarse particles from NOy samplers operating at low flow conditions as well as potential
23    difficulties of reducing organically bound and mineralized nitrate. These conversion efficiency
24    and particle size fraction issues are viewed by EPA as relatively minor mass accounting issues
25    that require more clarification but not necessarily technical resolution.

26          6.6.2   What sampling frequency would be required?
27          The averaging time for the standard is likely to be an annual average. Conceptually,
28    extended sampling periods as long as one year would be adequate for the specific purposes of
29    comparing to a standard. However, future assessments that characterize acidification and form


      March 2010                              208                Draft-Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    the scientific basis for subsequent standards reviews and allow for systematic checking of
 2    progress through accountability procedures benefit from more highly resolved data, especially
 3    the evaluation of air quality models that are key components of N/S deposition assessments. In
 4    addition, many of the monitoring approaches that are used throughout the nation sample (or at
 5    least report out) on daily (PM2.5 chemical speciation), weekly (CASTNET) and hourly (all
 6    inorganic gases) periods. There is a tradeoff to consider in sampling period design. For example,
 7    the weekly CASTNET collection scheme covers all time periods throughout a year, but only
 8    provides weekly resolution that misses key temporal and episodic features valuable for
 9    diagnosing model behavior. The every third day, 24-hour sampling scheme used in IMPROVE
10    and EPA speciation monitoring does provide more information for a specific day of interest yet
11    misses 2/3 of all sampling periods. The missing sampling period generally is not a concern when
12    aggregating upward to a longer term average value as the sample number adequately represents
13    an aggregated mean value. Additionally, there is a benefit to leveraging existing networks which
14    should be considered in  sampling frequency recommendations. A possible starting point would
15    be to assume gaseous oxidized species, NOy and SC>2, are run continually all year reporting
16    values every  hour, consistent with current routine network operations. Sulfate  sampling periods
17    should coincide with either the chemical speciation network schedules or CASTNET. There are
18    advantages to coordinating with either network. Ammonia gas and ammonium ion present
19    challenges in that they are not routinely sampled and analyzed for, and the combined quantity,
20    NHX is of interest. Because NHX is of interest, some of the problems of volatile ammonia loss
21    from filters may be mitigated. However, for model diagnostic purposes, delineation of both
22    species at the highest temporal resolution is preferred. While levels of deposited reduced
23    nitrogen would be specified by EPA for purposes of the APPI, monitoring of reduced nitrogen
24    would be important but would not be used in the APPI itself.

25          6.6.3   What are the spatial scale issues associated with monitoring for compliance,
26                 and how should these be addressed?
27          The observation network for NOy, NHX and SOX is very modest and includes a
28    monitoring network infrastructure that is largely population oriented. While there is platform and
29    access infrastructure support provided by CASTNET, NADP and IMPROVE,  those locations by
30    themselves are not likely to provide the needed spatial coverage to address acid sensitive
      March  2010                              209                Draft-Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1   watersheds across the United States. Ambient monitoring at every watershed may not be required
 2   due to the nature of the ambient air quality in acid sensitive areas. An understanding of the
 3   spatial variability of NOy, NHX, sulfate and SO2 will help inform monitoring. Critical load
 4   models are based on annual averages, which effectively serves to dampen much of the spatial
 5   variability.  Furthermore, the development of an area-wide depositional load tradeoff curve
 6   implies focus on region wide characterization. Toward that end, CMAQ concentration fields will
 7   provide insight into the likely spatial representativeness of monitors leading to efficient
 8   application of monitoring resources. For example, the CMAQ based spatial coefficient of
 9   variation (standard deviation/mean) of oxidized nitrogen in the Adirondacks was 1.46%.
10   Improved dry deposition estimates will result from enhancements of ambient monitoring
11   addressing the N/S secondary standards as each additional location could serves a similar role
12   that existing CASTNET sites provide in estimating dry deposition.

13   6.7    TAKING INTO CONSIDERATION INFORMATION ABOUT
14          ECOSYSTEM SERVICES AND OTHER FACTORS RELATED TO
15          CHARACTERIZING ADVERSITY FOR THE ECOLOGICAL
16          EFFECTS BEING ASSESSED IN THIS REVIEW, WHAT IS AN
17          APPROPRIATE RANGE OF ALTERNATIVE STANDARDS FOR
18          THE AGENCY TO CONSIDER?
19          The secondary NAAQS will reflect the public welfare policy judgments of the
20   Administrator, based on the science, as to the level of air quality which is requisite to protect the
21   public welfare from any known or anticipated adverse effects associated with the pollutant in the
22   ambient air. The exposure and risk assessment provide information regarding the effects
23   associated with a number of different welfare endpoints at different levels of air quality,
24   expressed in terms of the joint annual mean concentrations of NOX and SOX determined  such that
25   specific levels of ecosystem protection (for example, ANC greater than 50) are met. Staff also
26   recognizes that in certain naturally acidic ecosystems, even though the ecological benchmarks
27   are exceeded, e.g. ANC may be quite low; NOX and SOX are not contributing to effects because
28   those systems have chronic natural acidity and will not benefit from reductions in atmospheric
29   deposition. The  secondary NAAQS are not intended to provide protection in these types of
     March 2010                              210                Draft-Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    naturally acidic systems. As noted earlier, we will be exploring methods to address the design of
 2    the standard relative to these naturally acidic systems in the second draft policy assessment. The
 3    secondary NAAQS are focused on providing protection in areas where ambient NOX and SOX are
 4    resulting in effects in ecosystems with low natural levels of acidification that are highly sensitive
 5    to additional inputs of acid deposition.
 6           Staff believes that ecosystem effects of NOX and SOX deposition in aquatic ecosystems
 7    are an important public welfare effect of concern. There are several sources of benchmark values
 8    for ANC that can help to inform a determination of adversity. [Additional information on
 9    benchmark values will be provided in the second draft policy assessment] Staff concludes that
10    achieving ANC in the range of 50 to 100 would be likely to provide adequate protection against
11    the effects of acidification on ecosystems.
12           Based on our analyses of risks of impacts on aquatic species diversity and fitness and on
13    the basis of the scientific effects literature, we anticipate that achieving the upper end of this
14    ANC range would substantially decrease the effects of acidification due to NOX and  SOX on
15    aquatic ecosystems.  Additionally, it is anticipated that  achieving the upper end of this range
16    would provide increased protection from NOX and SOX in  areas with higher levels of variability
17    in ecosystem sensitivity due to variability in meteorology, bedrock geology,  topography, land
18    use characteristics, or reduced nitrogen deposition.
19           These ANC levels are estimated to protect sensitive aquatic ecosystems from significant
20    negative effects of NOX and SOX deposition on aquatic biota, including large reductions in
21    diversity offish species, and changes in health offish populations, affecting  reproductive ability
22    and fitness. It is recognized, however, that a standard set within this range would not protect the
23    most sensitive aquatic ecosystems or species within those ecosystems from the effects of NOX
24    and SOX. At ANC levels below 100, while overall health of an aquatic community can be
25    maintained, ANC levels are expected to be such that fish fitness and community  diversity begin
26    to decline. At ANC levels between 100 and 50, ANC levels are expected to be such that the
27    fitness of sensitive species (e.g., brook trout, zooplankton) also begins to decline. Staff notes that
28    at levels of ANC above 100, biota are generally not harmed. As such, achieving an ANC of
29    greater than 100 would be expected to result in little damage from NOX and SOX deposition to
30    aquatic ecosystems.
      March 2010                               211                 Draft-Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1           Specifying an appropriate range of levels for an AAPI standard that is designed and
 2    specified as discussed above involves consideration of the degree to which any specific AAPI
 3    would lead to achieving the desired ANC level, and a judgment as to the degree of protection of
 4    public welfare that is warranted. In general, staff initially conclude that it would be appropriate
 5    for the Administrator to consider an AAPI in the range of 50 to 100. Selection of a range of
 6    AAPI and selection of a specific level of AAPI within that range should incorporate a wide
 7    number of considerations, including the percent of water bodies within acid sensitive areas that
 8    the Administrator determines should be protected at the targeted ANC level.
 9           The Administrator should consider the uncertainties in the ecological effects observed in
10    the literature and the adversity to public welfare associated with those effects. In determining the
11    requisite level of protection for the public welfare from effects on aquatic ecosystems, the
12    Administrator will need to weigh the importance of the predicted risks of these effects in the
13    overall context of public welfare protection, along with a determination as to the appropriate
14    weight to place on the associated uncertainties and limitations of this information.
15           In addition, selection of a specific level of AAPI should consider uncertainties in the
16    design and calculation of the parameters included in the AAPI, including uncertainties in the
17    characterization of natural background ANC  (indicated by g in the AAPI equation), spatial and
18    temporal averaging of aggregate effective deposition velocities (indicated by VNOY and VSOX in
19    the AAPI equation), and spatial and temporal averaging of NHX deposition (indicated by NHX in
20    the AAPI equation).
21
      March 2010                                212                 Draft-Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX


 i         7.     CO-PROTECTION FOR OTHER EFFECTS USING
 2         STANDARDS TO PROTECT AGAINST ACIDIFICATION
 3          To this point, the standard for NOX and SOX centers on ecosystem protection against
 4   aquatic acidification. This chapter focuses on the level of co-protection that this standard would
 5   provide for other ecological effects, including terrestrial acidification, terrestrial nutrient
 6   enrichment, and estuarine eutrophication.

 7   7.1    TO WHAT EXTENT WOULD A STANDARD SPECIFICALLY
 8          DEFINED TO PROTECT AGAINST AQUATIC ACIDIFICATION
 9          LIKELY PROVIDE PROTECTION FROM TERRESTRIAL
10          ACIDIFICATION?
11          In order to understand the  level of protection provided by a NOX/SOX standard based on
12   aquatic acidification to protect against terrestrial acidification effects, an analysis was conducted
13   comparing the critical loads for lakes and streams that would be developed to protect for an
14   aquatic ANC of 50 to the critical loads to protect for either a terrestrial Be: Al ratio of 1.2 or 10
15   averaged across a watershed area.  See Appendix B for full analysis results. The analysis selected
16   16 watersheds with 29 lakes in the Adirondacks case study area, 4 watersheds randomly selected
17   from each of 4 categories of sensitivity reported in the REA: highly sensitive, moderately
18   sensitive, low sensitivity, and not  sensitive. In the  Shenandoah case study area, there were a
19   limited number of watersheds in the low sensitivity and not sensitive range, so 18 of the 20
20   streams in 16 watersheds selected were located in highly and moderately sensitive categories.
21          Results for the Adirondacks showed that critical loads for 29 lakes at an ANC of 50 were
22   lower for 13 lakes than the critical load for the terrestrial watershed areas at a Bc:Al ratio of 10
23   and for 21 lakes at a Bc:Al ratio of 1.2. Perhaps more significant was the result that 13 of the 16
24   lakes in the highly and moderately sensitive areas had a lower critical load than the Bc:Al  10
25   areas and 16 of 16 lakes  in the highly and moderately sensitive areas  had lower critical loads
26   than the Bc:Al 1.2 areas. The Shenandoah region reflected similar results. See table 7.1 below
27   for tabulated results.
28

     March 2010                              213                Draft-Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX
     Table 7-1. Results of comparing aquatic ANC50 critical loads to average terrestrial watershed
     area Bc:Al ratios. Left numbers in each column are the number of lakes or streams that had a
     lower critical load than the terrestrial calculated critical load. Right numbers in each column are
     the number of lakes that had a higher critical load than the watershed calculated terrestrial
     critical loads.

Adirondack Be: Al 10
Adirondack Be Al 1.2
Shenandoh Bc:Al 10
Shenandoh Bc:Al 1.2
Highly Sensitive
7-0
7-0
13-0
13-0
Moderately Sensitive
6-3
9-0
5-0
5-0
Low Sensitivity
0-7
5-2
0-1
0-1
Not Sensitive
0-6
0-6
0-1
0-1
 2          In summary, a comparison of the terrestrial and aquatic critical acid loads for watersheds
 3   in the Adirondacks and Shenandoah Case Study Areas indicated that, in general, the aquatic
 4   critical acid loads offered greater protection to the watersheds than did the terrestrial critical
 5   loads. Generally in situations where the terrestrial critical loads were more protective, the lakes
 6   or streams in the watershed were rated as having "Low Sensitivity" or "Not Sensitive" to
 7   acidifying nitrogen and sulfur deposition. Conversely, when the water bodies were more
 8   sensitive to deposition ("Highly Sensitive" or "Moderately Sensitive"), the aquatic critical acid
 9   loads generally provided a greater level of protection against acidifying nitrogen and sulfur
10   deposition in the watershed. In the next draft of the Policy Assessment Document, we intend to
11   expand this analysis by comparing more levels of ANC to other Bc:Al ratios.

12   7.2    TO WHAT EXTENT WOULD A STANDARD SPECIFICALLY
13          DEFINED TO PROTECT AGAINST AQUATIC ACIDIFICATION
14          LIKELY PROVIDE PROTECTION  FROM TERRESTRIAL
15          NUTRIENT ENRICHMENT?
16          This question will be answered in the next draft of the Policy Assessment Document.
17   Once maximum depositonal loads are calculated for broad areas, we can compare the  derived
18   maximum NOy limits to nutrient enrichment benchmarks found in the REA. Benchmarks for
19   lichens, grasses, mychorrizae, and diatoms will be compared to the aquatic acidification limits
20   for nitrogen.
     March 2010
214
Draft -Do Not Quote or Cite

-------
          Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1   7.3   TO WHAT EXTENT WOULD A STANDARD SPECIFICALLY
 2         DEFINED TO PROTECT AGAINST AQUATIC ACIDIFICATION
 3         LIKELY PROVIDE PROTECTION FROM AQUATIC NUTRIENT
 4         ENRICHMENT?
 5         The REA found that deposition of reactive nitrogen contributed to eutrophication of
 6   estuaries; however, it was also noted that atmospheric deposition of nitrogen is only part of the
 7   total nitrogen load to the estuaries. Due to the complications of separating out the effects of
 8   atmospheric deposition from the effects of other nitrogen loads, CASAC did not recommend that
 9   a secondary NAAQS be set to specifically protect against estuarine eutrophication. In the next
10   draft of the Policy Assessment Document, we will attempt to analyze the benefit to the
11   Chesapeake Bay that attaining an aquatic acidification standard would provide by decreasing
12   nitrogen deposition to the watershed.
13
14
     March  2010                            215                Draft-Do Not Quote or Cite

-------
        Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX


i     8.    CONSIDERATION OF ISSUES REGARDING REDUCED
2               AND OXIDIZED FORMS OF NITROGEN
3         [To be added in the second draft Policy Assessment]
4
5
    March 2010                         216              Draft-Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1                             9.     INITIAL CONCLUSIONS
 2           Staff initial conclusions on the elements of the secondary NOX and SOX standards for the
 3    Administrator's consideration in making decisions on the secondary NOX and SOX standards are
 4    summarized below, together with supporting conclusions from previous chapters. We recognize
 5    that selecting from among alternative policy options will necessarily reflect consideration of
 6    qualitative and quantitative uncertainties inherent in the relevant evidence and in the assumptions
 7    of the quantitative exposure and risk assessments. Any such standard should protect public
 8    welfare from any known or anticipated adverse effects associated with the presence of the
 9    pollutant(s) in the ambient air. In providing these options for consideration, we are mindful that
10    the Act requires standards that, in the judgment of the Administrator, are requisite to protect
11    public welfare. The standards are to be neither more nor less stringent than necessary.
12           To evaluate whether the current secondary NAAQS is adequate or whether consideration
13    of revisions is appropriate, the conclusions and options for the Administrator to consider in this
14    review are based on effects-,  exposure- and risk-based considerations. The exposure and risk
15    assessments reflect the availability of new tools, assessment methods, and a larger and more
16    diverse body of evidence than was available in the last reviews. We have taken a weight of
17    evidence approach that evaluates information across the variety of research areas described in the
18    ISA and in addition includes  assessments of air quality, exposures, and qualitative and
19    quantitative risks associated with alternative air quality  scenarios.
20           Staff notes that since the last review, additional policy-relevant developments have
21    occurred that may also warrant consideration by the Administrator when making decisions about
22    what is requisite to protect public welfare. The NRC report (described in Chapter 6) states:
23    "Whatever the reason that led EPA to use identical primary and secondary NAAQS in the past, it
24    is becoming increasingly evident that a new approach will be needed in the future. There is
25    growing evidence that the current forms of the NAAQS are not providing adequate protection to
26    sensitive ecosystems and crops" (NRC, 2004).
27           The last review raised the following key issues as a rationale for not setting a separate
28    standard for NOX to protect against acidification and nutrient enrichment effects in sensitive
29    ecosystems:
30           1)  Lack of enough consistent information to support a revision of the current secondary
31              standard to protect these aquatic systems.

      March 2010                               217                Draft-Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1           2) Lack of adequate quantitative evidence on the relationship between deposition rates
 2              and environmental impacts
 3           3) Significant uncertainties with regard to the long-term role of nitrogen deposition in
 4              surface water acidity and with regard to the quantification of the magnitude and
 5              timing of the relationship between atmospheric deposition and the appearance of
 6              nitrogen in surface water.
 7           In this current review, staff concludes that important new information has become
 8    available since the last review that supports revising the current NOX and SOX standards.
 9    Specifically, the ISA has concluded that there are causal relationships between NOX and SOX
10    acidifying deposition and effects on aquatic and terrestrial  ecosystems, and the ISA and REA
11    provide substantial quantitative evidence of effects occurring in locations that meet the current
12    NC>2 and SC>2 standards. In addition, substantial new information, based on observational data
13    and rigorous atmospheric modeling, has become available  regarding the role of both nitrogen and
14    sulfur deposition in acidification of sensitive water bodies. This information is sufficient to
15    inform the development of revised secondary standards for NOX and SOX to protect against the
16    effects of acidification20. While there is also new information available on the role of nitrogen
17    deposition on nutrient enrichment effects in terrestrial and  aquatic ecosystems, and the ISA
18    concludes there is a causal relationship between NOX and nutrient enrichment effects,  for this
19    first draft policy assessment, staff have focused on acidification effects due to the substantially
20    greater amount of information available to inform the development of secondary standards.
21           Staff highlights the progress made in considering the joint nature of ecosystem responses
22    to acidifying deposition of NOX and SOX, and notes that the ability to consider revisions to the
23    NOX and SOX secondary standards has been enhanced by our ability to consider a joint standard
24    for NOX and SOX to protect against acidification effects. The development of an appropriate form
25    of the standard linked to a common indicator of aquatic acidification, ANC, is also a significant
26    step forward, as it allows for development of a standard for aquatic acidification designed to
27    provide generally the sme degree of protection across the country, while  still reflecting the
28    underlying variability in ecosystem sensitivity to acidifying NOX and SOX deposition.
29
      20 As we have note earlier in the document, in this draft we have focused on aquatic acidification. However, in the
      second draft policy assessment we plan to more fully explore the possibility of expanding the conceptual model to
      address terrestrial acidification.
      March 2010                               218                Draft-Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    9.1   CONCLUSIONS
 2          As noted throughout this document, because of the complex interactions between NOX
 3    and SOX in the atmosphere and their impacts once deposited in ecosystems, the consideration of
 4    indicators, averaging times, forms, and levels for the two pollutants is being conducted jointly. In
 5    addition, as discussed in Chapters 5  and 6, we are considering structures for the standards that
 6    reflect a more scientifically derived understanding of the relationships between atmospheric
 7    concentrations of NOX and SOX and the primary indicators of ecosystem impacts.
 8          With respect to soil and water effects information, we have evaluated the conclusions
 9    drawn at the end of the last review in light of more recent evidence from studies for a variety of
10    ecological effects endpoints. We place greater weight on U.S. studies due to the species-, site-,
11    and climate-specific nature of ecological responses. With respect to quantitative exposure- and
12    risk-based considerations, we have relied on both monitored and modeled NOX and SOX ambient
13    concentrations and related deposition, as described in Chapter 3 of the REA.
14          Uncertainties associated with the exposure and risk assessments are also discussed,
15    including, where possible, some sense of the direction and/or magnitude of the uncertainties that
16    should be taken into account as one considers these estimates. As with any analysis that relies on
17    complex scientific models, there are a number of unknown and unquantifiable sources of
18    uncertainty. However, each model that has been applied in the risk and exposure assessment
19    represents the best available science and the models have all been subject to substantial levels of
20    peer-review.
21          The following secondary NAAQS conclusions encompass the breadth of policy-relevant
22    considerations described in this policy assessment:
23          (1) Based on the policy-relevant findings from the ISA described in Chapter 2, and while
24             recognizing that important uncertainties and research questions remain, staff conclude
25             that great progress has been made since the last reviews of the secondary standards
26             for NOX and SOX. We generally find support in the available effects-based evidence
27             for consideration of NOX and SOX standards that are at least as protective as the
28              current standard and do not find support for consideration of NOX and SOX standards
29             that are less protective than the current standard. The staff also concludes that
30              consideration of joint standards for NOX and SOX is appropriate given the common
31              atmospheric processes governing the deposition of NOX and SOX to sensitive

      March 2010                               219                 Draft-Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1              ecosystems, and given the combined effects of N and S deposition on acidification of
 2              soil and water.
 3           (2) Staff concludes that ambient NOX is a significant component of atmospheric nitrogen
 4              deposition, even in areas with relatively high rates of deposition of reduced nitrogen.
 5              Staff make this conclusion based on the analysis in Chapter 3 of the REA, which
 6              provides a thorough assessment of the contribution of NOX to nitrogen deposition
 7              throughout the U.S., and the relative contributions of ambient NOX and reduced forms
 8              of nitrogen.
 9           (3) Staff concludes based on the case study results provided in the REA, that current
10              levels of NOX and SOX are associated with deposition that leads to ANC values below
11              benchmark values that cause ecological harm and losses in ecosystem services. Staff
12              concludes that the evidence and risk  assessment support strongly a relationship
13              between atmospheric deposition of NOX and SOX and ANC, and that ANC is an
14              excellent indicator of aquatic acidification. Staff also concludes that at levels of
15              deposition associated with NOX and SOX concentrations at or below the current
16              standards, ANC levels are expected to be below benchmark values that are associated
17              with significant losses in fish species richness, which is associated with reductions in
18              recreational fishing services. While there are many other ecosystem services
19              potentially affected by reductions in ANC, including subsistence fishing, natural
20              habitat provision, and biological control, confidence in the specific translation of
21              ANC values to these additional ecosystem services is much lower.
22           (4) Losses in aquatic resources  associated with ANC levels below 50 are clearly
23              associated with significant losses in economic value. Based on the best available data,
24              just in the northeastern U.S., current  acidification levels  are resulting in $4 million to
25              $300 million in damages annually from lost recreational fishing. This estimate
26              represents only a fraction of the total economic value of ecosystem damages as many
27              impacted resources are not amenable to economic valuation methods. In addition,
28              economic damages are also likely to  occur in other areas affected by acidification,
29              including New England, the Appalachian Mountains (northern Appalachian Plateau
30              and Ridge/Blue Ridge region), and the Upper Midwest. Staff concludes that reducing
31              acidifying deposition of NOX and SOX will result in improvements in public welfare

      March 2010                               220                Draft-Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1              by increasing the quantity and quality of ecosystem services, including recreational
 2              fishing and other services associated with improved water quality.
 3           (5) Staff initially concludes based on the case study results that current levels of ambient
 4              NOX and SOX are associated with deposition that leads to BC:A1 values below
 5              benchmark values that cause ecological harm and losses in ecosystem services. Staff
 6              concludes that the evidence and risk assessment support strongly a relationship
 7              between atmospheric deposition of NOX and  SOX and BC:A1, and that BC:A1 is a
 8              good indicator of terrestrial acidification. Staff also concludes that at levels of
 9              deposition associated with NOX and SOX concentrations at or below the current
10              standards, BC:A1 levels are expected to be below benchmark values that are
11              associated with significant losses in tree health and growth, which are associated with
12              reductions in timber production. While there are many other ecosystem services,
13              including maple syrup production, natural habitat provision, and regulation of water,
14              climate, and erosion, potentially affected by reductions in BC:A1, confidence in the
15              specific translation of BC:A1 values to these  additional ecosystem services is much
16              lower.
17           (6) On the basis of the acidification and nutrient enrichment effects that have been
18              observed to still occur under current ambient conditions and those predicted to occur
19              under the scenario of just meeting the current secondary NAAQS, staff concludes that
20              the current secondary  NAAQS are inadequate to protect the public welfare from
21              known and anticipated adverse welfare effects from aquatic and terrestrial
22              acidification associated with deposition of NOX and SOX.. As discussed above, this
23              conclusion derives from several lines of evidence.
24           (7) Staff has concluded, based on the completeness  of the  available evidence and
25              quantitative risk information, that effects due to  aquatic and terrestrial acidification
26              are most suitable for defining secondary standards for NOX and SOX.  Staff notes that
27              in developing a standard designed to protect  against the effects of acidification due to
28              deposition of NOX and SOX, the resulting standards may not provide protection
29              against known effects associated with nutrient enrichment in aquatic and terrestrial
30              ecosystems.
      March 2010                                221                 Draft-Do Not Quote or Cite

-------
            Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1           (8) It is appropriate to consider using indicators other than NC>2 and 862 as the indicators
 2              for a standard that is intended to address the ecological effects associated with
 3              deposition of NOX and SOX to sensitive ecosystems. Given the reasons discussed in
 4              Chapters 2, 4, and 5 of this policy assessment, staff concludes thatNOx, as defined in
 5              the CAA, is best represented by the atmospheric indicator NOy, defined as NO2 + NO
 6              + HNO3 + PAN +2N2O5 + HONO+ NO3 + organic nitrates + paniculate NO3 is the
 7              more appropriate indicator of oxides of nitrogen,  and that SOX, defined to include
 8              sulfur monoxide (SO), sulfur dioxide, sulfur trioxide (SO3), and disulfur monoxide
 9              (S2O), and particulate-phase S compounds, is the more appropriate indicator of
10              oxides of sulfur.
11           (9) It is appropriate to use the annual average of concentrations of NOy and SOX as the
12              averaging time for the secondary standards, based on the chronic nature of
13              acidification, and the protection against episodic acidification provided by a standard
14              based on annual average concentrations.
15           (10) It is appropriate to consider changing the form of the secondary standards for NOX
16              and SOX as the current form does not take into account the linkages between NOX and
17              SOX in the causation of effects associated with acidification of aquatic ecosystems.
18              Based on the causal linkages between NOX and SOX, deposition of N and S, and the
19              indicator of acidification, ANC, staff concludes that the current forms should be
20              replaced with an atmospheric acidification potential index (AAPI), which reflects the
21              important roles of underlying ecosystem characteristics, determinants of deposition,
22              and reduced nitrogen deposition in determining the potential effects from deposition
23              of NOX and SOX.
24           (11) Staff initial conclusions regarding the elements of the standard, e.g. the target ANC,
25              spatial extent of areas in which the standard will be evaluated, percentiles of aquatic
26              ecosystems within sensitivity classes to be protected for alternative target ANC
27              values, calculated values of deposition transformation ratios,  natural buffering
28              capacity, and reduced nitrogen deposition will be provided in the second draft of the
29              policy  assessment. In addition, staff initial  conclusions regarding consideration of
30              uncertainty and variability in elements of the standard will be developed in the second
31              draft.

      March 2010                                222                 Draft-Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1    9.2   SUMMARY OF KEY UNCERTAINTIES AND RESEARCH
 2          RECOMMENDATIONS RELATED TO SETTING A SECONDARY
 3          STANDARD FOR NOX AND SOX
 4          [This section is still under development. Summary of key uncertainties to be added in
 5    second draft policy assessment. Research and data needs are partial lists that will be more
 6    completely developed in subsequent versions.]

 7          9.2.1   Research Needs to Reduce Uncertainty in the Next Review (focused on
 8                 aquatic acidification)
 9          Based on the information presented in this policy assessment, several information gaps
10    arise that suggest further research is needed in the following areas:
11      •  Developing relationships between aquatic acidity as measured by ANC, and effects on
12         ecological effects and ecosystem services, especially due to incremental changes
13      •  Developing nationwide weathering rates, or weathering rates for aquatic ecosystems
14         sensitive to acidification
15      •  Developing a better understanding of the uncertainty in  critical loads for acidity
16      •  Developing methods for calculating critical loads for surface water acidity when data are
17         absent or of poor quality
18      •  Evaluating ways to combine multiple critical load estimates for surface waters and soils on
19         a national scale
20      •  Estimating ways to determine critical load parameters across different media (e.g., surface
21         waters, soils).

22          9.2.2   Data Needs to Reduce Uncertainty in the Next Review (focused on aquatic
23                 acidification)
24          Improved measurements of reduced nitrogen: Nitrification processes within watershed
25    soil, sediment and vegetation systems effectively convert ammonia gas and ammonium ions to
26    nitrates, which contribute to the overall acidifying loads in ecosystems; consequently, the
27    atmospheric contributions of reduced nitrogen must be accounted for in acidification
28    assessments. We would expect that all or a subset of ambient monitoring platforms supporting


      March 2010                             223                Draft-Do Not Quote or Cite

-------
           Policy Assessment for the Review of the Secondary National Ambient Air Quality Standards for NOX and SOX

 1   the N/S secondary standard will measure both ammonia gas and ammonium ion along with
 2   oxidized sulfur and nitrogen species.
 3          Extended modeling of air quality and deposition to inform monitoring network design: In
 4   addition to providing deposition inputs for watershed models and critical loads analysis, the
 5   spatial and temporal flexibility afforded by air quality modeling can support monitoring network
 6   design and in inform the averaging time period (one or more years) to more appropriately
 7   account for inter-annual variability in NOX and SOX concentrations.
 8          Development of data fusion approaches to combine model results with observational
 9   data: Consideration also will be given to fusing model results with observation fields to improve
10   spatial resolution by taking advantage of the landscape, emissions and meteorological
11   information that affect spatial gradients while relying on observations to reduce the influence of
12   model uncertainties.
     March  2010                               224                Draft-Do Not Quote or Cite

-------
 3                                             APPENDIX A

 4        CHAPTER 5: CONCEPTUAL DESIGN OF THE

 5                                              STANDARD

 6

 7

 8                              First External Review Draft

 9
10
11
12                                                   Prepared by:
13
14                                   U.S. Environmental Protection Agency
15                               Office of Air Quality Planning and Standards
16                                      Research Triangle Park, NC 27711
17
18
19

-------

-------
                                                                                Appendix A
 1                               TABLE OF CONTENTS

 2   Appendix A  Chapter 5:  Conceptual Design of the Standard	1
 3      A. 1  Technical summary of methods used in the REA Aquatic Acidification analysis	1
 4      A.2  Technical summary of critical loads modeling in the REA	2
 5           1.2.1  Preindustrial Base Cation Concentration	5
 6           1.2.2  F-factor	6
 7
 9                                  LIST OF TABLES

10   Table A. 1. Brief summary of objects and methods used in the REA Aquatic Acidification
11                 analysis	1
12   Table A.2 Illustrates SSWC Approach -Environmental Variables	7
13   Table A.3 FAB Approach -Environmental Variables	8
14
15
     March 2010                               i                Draft - Do Not Quote or Cite

-------
                                                                                   Appendix A
1                                This page intentionally left blank.
2
    March 2010                                 ii                 Draft - Do Not Quote or Cite

-------
                                                                             Appendix A
 i                                   APPENDIX A
 2      CHAPTERS: CONCEPTUAL DESIGN OF THE STANDARD
 3         This is supplemental information to support the discussion of the conceptual design of the
 4   standard that is presented in Chapter 5 of the Policy Assessment Document. The aquatic
 5   acidification analyses developed in the REA used a number of different models and calculation
 6   techniques that are important for the development of the standard. The goal of this Appendix is
 7   to summarize information from the REA analysis that is most relevant to the Policy Assessment.
 8   A brief summary of the REA analyses are presented in section 1. In section 2 there is a general
 9   summary and technical discussion of the critical loads modeling approaches that were used in the
10   REA, followed by a brief description of MAGIC model data requirements.

11   A. 1   TECHNICAL SUMMARY OF METHODS USED IN THE REA
12         AQUATIC ACIDIFICATION ANALYSIS
13         The aquatic acidification analysis is presented in Chapter 4 and Appendix 4 of the REA.
14   The analysis uses multiple techniques to show the relationship between ANC and NOX and SOX
15   deposition, as well as determine the current level of risk to water bodies that occur in sensitive
16   areas. A brief summary of the techniques and objectives of the REA analysis is given in Table 1.

17   Table A.I. Brief summary of objects and methods used in the REA Aquatic Acidification
18   analysis.
Technique
Time-series
graphs of
current
conditions
MAGIC


Objectives
1
1
2
3
Data from monitoring networks collected from 1990 to 2006 were
plotted to show trends in concentrations of pollutants, deposition and
acidification for each case study site. The data included surface water
concentration of nitrate, sulfate and ANC; deposition of sulfate and
nitrate; as well as air concentration of SOX, NOX and NH4
Used to estimate the relationship between ANC values and
anthropogenic NOX and SOX emission from the past (preacidification
-I860), present (2002 and 2006) and projected into the future (2020
and 2050). Analysis included 44 lakes from Adirondacks and 60
streams from Shenandoah.
Used to develop input parameters for critical loads modeling (i.e.
weathering rates)
Used for uncertainty analysis
     March 2010
Draft - Do Not Quote or Cite

-------
                                                                                   Appendix A
Technique
Critical Loads
modeling


Regional
Extrapolation

Objectives
1
2
3
1
2
SSWC and FAB models used to calculate critical loads for critical
limits of ANC = 0, 20, 50, 100
Critical loads for ANC critical limits calculated for 169 lakes in the
Adirondacks and 60 streams in the Shenandoah using water quality
data from monitoring sites collected in 2006
Critical loads exceedences calculated by comparing the critical loads
that were calculated by SSWC with deposition data from NADP for
wet deposition and CMAQ for dry deposition, both for the year 2002
117 of the critical loads calculated for the Adirondacks were
extrapolated to lakes defined by the New England EMAP probability
survey, representing 1842 lakes, to infer the # of lakes that exceeded
their critical load
69 of the critical loads calculated for the Shenandoah were
extrapolated to 330 streams based on bed rock geology classification.
 1
 2   A.2   TECHNICAL SUMMARY OF CRITICAL LOADS MODELING IN
 3          THE REA
 4          The critical load of acidity for lakes or streams was derived from present-day water
 5   chemistry using a combination of steady-state models. Both the Steady-State Water Chemistry
 6   (SSWC) model and First-order Acidity Balance model (FAB) is based on the principle that
 7   excess base-cation production within a catchment area should be equal to or greater than the acid
 8   anion input, thereby maintaining the ANC above a preselected level (Reynolds and Norris, 2001;
 9   Posch et al. 1997). These models assume steady-state conditions and assume that all SC>42  in
10   runoff originates from sea salt spray and anthropogenic deposition. Given a critical ANC
11   protection level, the critical load of acidity is simply the input flux of acid anions from
12   atmospheric deposition (i.e., natural and anthropogenic) subtracted from the natural (i.e.,
13   preindustrial) inputs of base cations in the surface water. Final Risk and Exposure Assessment
14   September 2009 Appendix 4, Attachment A - 15 Aquatic Acidification Case Study Atmospheric
15   deposition of NOX and SOX contributes to acidification in aquatic ecosystems through the input of
16   acid anions, such as NO3- and SC>42 The acid balance of headwater lakes and streams is
17   controlled by the level of this acidifying deposition of NO3- and SC>42 and a series of
18   biogeochemical processes that produce and consume acidity in watersheds. The biotic integrity
19   of freshwater ecosystems is then a function of the acid-base balance, and the resulting acidity-
     March 2010
Draft - Do Not Quote or Cite

-------
                                                                                   Appendix A

 1    related stress on the biota that occupy the water. The calculated ANC of the surface waters is a
 2    measure of the acid-base balance:
 3                                     ANC = [BC]* - [AN]*                               (1)
 4    where [BC]* and [AN]* are the sum of base cations and acid anions (NO3- and SC>42),
 5    respectively. Equation (1) forms the basis of the linkage between deposition and surface water
 6    acidic condition and the modeling approach used. Given some "target" ANC concentration
 7    [ANClimit]) that protects biological integrity, the amount of deposition of acid anions (AN) or
 8    depositional load of acidity CL(A) is simply the input flux of acid anions from atmospheric
 9    deposition that result in a surface water ANC concentration equal to the [ANClimit] when
10    balanced by the sustainable flux of base cations input and the sinks of nitrogen and sulfur in the
11    lake and watershed catchment.
12          Critical loads for nitrogen and sulfur (CL(N) + CL(S) ) or critical load of acidity CL(A)
13    were calculated for each waterbody from the principle that the acid load should not exceed the
14    nonmarine, nonanthropogenic base  cation input and sources and sinks in the catchment minus a
15    neutralizing to protect selected biota from being damaged:
16                 CL(N) + CL(S) or CL(A) = BC*dep + BCw - Ecu - AN - ANClimit           (2)
17    Where,
18    BC*dep = (BC*=Ca*+Mg*+K*+Na*), nonanthropogenic deposition flux of base cations BCw =
19    the average weathering flux, producing base cations
20    Ecu (Bc=Ca*+Mg*+K*) = the net long-term average uptake flux of base cations in the biomass
21    (i.e., the annual average removal of base cations due to harvesting)
22    AN = the net long-term average uptake, denitrification, and immobilization of nitrogen anions
23    (e.g. NO3-) and uptake of SO42
24    ANClimit = the lowest ANC-flux that protects the biological communities.
25          Since the average flux of base cations weathered in a catchment and reaching the lake or
26    streams is difficult to measure or compute from available information, the average flux of base
27    cations and the resulting critical load estimation were derived from water quality data (Henriksen
28    and Posch, 2001;  Henriksen et al., 1992; Sverdrup et al.,  1990). Weighted annual mean water
29    chemistry values were used to estimate average base cation fluxes, which were calculated from
30    water chemistry data collected from the Temporally Integrated Monitoring of Ecosystems

      March 2010                                3                 Draft - Do Not Quote or Cite

-------
                                                                                    Appendix A

 1    (TIME)/Long-Term Monitoring (LTM) monitoring networks, that include Adirondack Longterm
 2    Monitoring (ALTM), Virginia Trout Stream Sensitivity Study (VTSSS), and the Shenandoah
 3    Watershed Study (SWAS), and Environmental Monitoring and Assessment Program (EMAP)
 4    (see Section 4.1.2.1 of Chapter 4).
 5          The preacidification nonmarine flux of base cations for each lake or stream, BC*0, is
 6                                 BC*0 = BC*dep + BCw - Ecu                           (3)
 7    Thus, critical load for acidity can be rewritten as
 8            CL(N) + CL(S) = BC*0 - AN - ANClimit = Q.([BC*]0 - [AN] - [ANC]limit),      (4)
 9    where the second identity expresses the critical load for acidity in terms of catchment runoff (Q)
10    m/yr and concentration ([x] = X/Q). The sink of nitrogen in the watershed is equal to the uptake
11    (Nupt), immobilization (Nimm), and denitrification (Nden) of nitrogen in the catchment. Thus,
12    critical load for acidity can be rewritten as
13           CL(N) + CL(S) = (fNupt + (1 - r)(Nimm + Nden)} + (  [BC]0* - [ANClimit])Q     (5)
14    where f and r are dimensionless parameters that define the fraction of forest cover in the
15    catchment and the lake/catchment ratio. The in-lake retention of nitrogen and sulfur was assumed
16    to be negligible. Equation 5 described the FAB model that was applied when sufficient data was
17    available to estimate the uptake, immobilization, and denitrification of nitrogen and the
18    neutralization of acid anions (e.g. NO3-) in the catchment. In the case were data was not
19    available, the contribution of nitrogen anions to acidification was assumed to be equal to the
20    nitrogen leaching rate (Nleach) into the surface water. The flux of acid anions in the surface
21    water is assumed to represent the amount of nitrogen that is not retained by the catchment, which
22    is determined from the sum of measured concentration of NO3- and ammonia in the stream
23    chemistry. This case describes the SSWC model and the critical  load for acidity is
24                               CL(A) = Q.([BC*]0 - [ANC]limit)                         (6)
25    where the contribution of acid anions is considered as part of the exceedances calculation (see
26    Section 1.2.5, below). For the assessment of current condition in both case study areas, the
27    critical load calculation described in Equation 6 was used for most lakes and streams. The lack of
28    sufficient data for quantifying nitrogen denitrification and immobilization prohibited the wide
29    use of the FAB model. In addition, given the uncertainty in quantifying nitrogen denitrification

      March 2010                                 4                Draft - Do Not Quote or Cite

-------
                                                                                     Appendix A

 1    and immobilization, the flux of nitrogen anions in the surface water was assumed to more
 2    accurately reflect the contribution of NO3- to acidification. Several major assumptions are made:
 3    (1) steady-state conditions exist, (2) the effect of nutrient cycling between plants and soil is
 4    negligible, (3) there are no significant nitrogen inputs from sources other than atmospheric
 5    deposition, (4) ammonium leaching is negligible because any inputs are either taken up by biota
 6    or adsorbed onto soils or nitrate compounds, and (5) longterm sinks of sulfate in the catchment
 7    soils are negligible.

 8           1.2.1  Preindustrial Base Cation Concentration
 9           Present-day surface water concentrations of base cations are elevated above their
10    steady state preindustrial concentrations because of base cation leaching through ion exchange in
11    the soil due to anthropogenic inputs of SC>42 to the watershed. For this reason, present-day
12    surface water base cation concentrations are higher than natural or preindustrial levels, which, if
13    not corrected for, would result in critical load values not in steady-state condition. To estimate
14    the preacidification flux of base cations, the present flux of base cations was estimated,
15                      BC*t, given by BC*t = BC*dep + BCw-Ecu +BCexc,                (7)
16    Where BCexc = the release of base cations due to ion-exchange processes. Assuming that
17    deposition, weathering rate, and net uptake have not changed over time, BCexc can be obtained
18    by subtracting Equation 5 from Equation 7:
19                                     BCexc = BC*t-BC*0                               (8)
20    This present-day excess production of base cations in the catchment was related to the long-term
21    changes in inputs of nonmarine acid anions (ASO*2 + ANO3) by the F-factor (see below):
22                                  BCexc = F (ASO*2 + ANO3)                            (9)
23    For the preacidification base cation flux, solving Equation 5 for BC*0 and then substituting
24    Equation 8 for BCexc and explicitly describing the long-term changes in nonmarine acid ion
25    inputs:
26                      BC*0 = BC*t-F(SO*4,t-SO*4,0+NO*3,t-NO*3,0)               (10)
27    The preacidification NO3- concentration, NO*3,0, was assumed to be zero.
      March 2010                                 5                 Draft - Do Not Quote or Cite

-------
                                                                                    Appendix A

 1          1.2.2  F-factor
 2    An F-factor was used to correct the concentrations and estimate preindustrial base concentrations
 3    for lakes in the Adirondack Case Study Area. In the case of streams in the
 4    Shenandoah Case Study Area, the preindustrial base concentrations were derived from the
 5    MAGIC model as the base cation supply in 1860 (hindcast) because the F-factor approach is
 6    untested in this region. An F-factor is a ratio of the change in nonmarine base cation
 7    concentration due to changes in strong acid anion concentrations (Henriksen,  1984; Brakke et al.,
 8    1990):
 9                  F=([BC*]t-[BC*]0)/([SO4*]t-[SO4*]0 + [NO3*]t-[NO3*]0),            (12)
10    where the subscripts t and 0 refer to present and preacidification conditions, respectively. If F=l,
11    all incoming protons are neutralized in the catchment (only soil acidification); at F=0, none of
12    the incoming protons are neutralized in the catchment (only water acidification). The F-factor
13    was estimated empirically to be in the range 0.2 to 0.4, based on the analysis of historical data
14    from Norway, Sweden, the United States, and Canada (Henriksen, 1984). Brakke et al. (1990)
15    later suggested that the F-factor should be a function of the base cation concentration:
16                                    F = sin (Ti/2 Q[BC*]t/[S])                             (13)
17    where
18    Q = the annual runoff (m/yr). [S] = the base cation concentration at which F=l;  and for
19    [BC*]t>[S] F is set to 1. For Norway  [S] has been set to 400 milliequivalents per cubic meter
20    (meq/m3)(circa.8 mg Ca/L) (Brakke et al., 1990). The preacidification SO42- concentration in
21    lakes,  [SO4*]0, is assumed to consist of a constant atmospheric contribution and a geologic
22    contribution proportional to the concentration of base cations (Brakke et al., 1989). The
23    preacidification SO42- concentration in lakes, [SO4*]0 was estimated from the  relationship
24    between [SO42-]o* and [BC]t* based on work completed by Henriksen et al., 2002 as described
25    by the following equation:
26                                 [SO42-]o* = 15 + 0.16 * [BC]t*                           (14)
      March 2010                                 6                 Draft - Do Not Quote or Cite

-------
                                                                                 Appendix A
1
2
Table A.2 Illustrates SSWC Approach - Environmental Variables
                           CL(A) = BC*dep + BCW - Bcu - ANClimit
                              CL(A) = Q ([BC*]0 - [ANC]limit)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Variable
Code
BC dep
BCW
Bcu
ANClimit
Ca*
Mg*
Na*
K*
S04*
CL
S04*
NO3*
Q
[BC*]0
[S04*]0
[N03*]0
F
Description
Sum (Ca*+Mg*+K*+Na*), nonanthropogenic
deposition flux of base cations
Average weathering flux of base cations
Sum (Ca+Mg+K), the net long-term average
uptake flux of base cations in the biomass
Lowest ANC-flux that protects the biological
communities
Sea Salt corrected Surface water concentration
(ueq/L) growing season average. (Ca - (CL x
0.0213))
Sea Salt corrected Surface water concentration
(ueq/L) growing season average. (Mg - (CL x
0.0669))
Sea Salt corrected Surface water concentration
(ueq/L) growing season average. (Na - (CL x
0.557))
Sea Salt corrected Surface water concentration
(ueq/L) growing season average. (K - (CL x
0.0.0206))
Sea Salt corrected Surface water concentration
(ueq/L) growing season average. (SO4 - (CL x
0.14))
Surface water concentration (ueq/L) growing
season average.
Surface water concentration (ueq/L) growing
season average.
Surface water concentration (ueq/L) growing
season average.
The annual runoff (m/yr)
Preindustrial flux of base cations in surface water,
corrected for sea salts
Preindustrial flux of sulfate in surface water,
corrected for sea salts
Preindustrial flux of nitrate, corrected for sea salts
Calculated factor
Source
WetNADPandDry
CASTNET
Calculated (5-17)
USFS-FIA data
Set
Water quality data
Water quality data
Water quality data
Water quality data
Water quality data
Water quality data
Water quality data
Water quality data
USGS
Calculated from water
quality data
Estimated
Equal to 0
Fix values
4
5
    March 2010
                                                            Draft - Do Not Quote or Cite

-------
                                                                                  Appendix A
1   Table A.3 FAB
2   DL(N) + DL(S) =
Approach - Environmental Variables
 {fNupt + (1 - r)(Nimm + Nden) + (Nret + Sret)}
[BC]0* - [ANClimit])Q

1
2
3
4
5
6
7
8
9
10
11
12
13
14
14
15
16
17
18
19
20
Variable
Code
Ndepo
ANClimit
[BC*]0
Ca*
Mg*
Na*
K*
SO4*
CL
S04*
NO3*
Q
f
r
Nret
^ret
Nupt
-L Mmm
Nden
Lake Size
WSH
Description
Total N deposition
Lowest ANC-flux that protects the biological
communities
Preindustrial flux of base cations in surface water,
corrected for sea salt
Sea Salt corrected Surface water concentration
(ueq/L) growing season average. (Ca - (CL x
0.0213))
Sea Salt corrected Surface water concentration
(ueq/L) growing season average. (Mg - (CL x
0.0669))
Sea Salt corrected Surface water concentration
(ueq/L) growing season average. (Na - (CL x 0.557))
Sea Salt corrected Surface water concentration
(ueq/L) growing season average. (K - (CL x
0.0.0206))
Sea Salt corrected Surface water concentration
(ueq/L) growing season average. (SO4 - (CL x 0.14))
Surface water concentration (ueq/L) growing season
average.
Surface water concentration (ueq/L) growing season
average.
Surface water concentration (ueq/L) growing season
average.
The annual runoff (m/yr)
f is a dimensionless parameter that define the fraction
of forest cover in the catchment
r is a dimensionless parameter that define the
lake/catchment ratio
The in-lake retention of nitrogen
The in-lake retention of sulfur
The net long-term average uptake flux of N in the
biomass
Immobilization of N in the soils
Denitrification
Lake size (ha)
Watershed area (ha)
Source
NADP/CMAQ
Set
Calculated from water
quality data
Water quality data
Water quality data
Water quality data
Water quality data
Water quality data
Water quality data
Water quality data
Water quality data
USGS


Estimated
Estimated
USFS-FIA data
Estimated fix value
Estimated fix value
DLMs
Calculated
    March 2010
       Draft - Do Not Quote or Cite

-------
                                                                                     Appendix A

 1          Data requirements for MA GIC
 2          The MAGIC model (Cosby et al., 1985a; 1985b; 1985c) is a mathematical model (a
 3    lumped-parameter model) of soil and surface water acidification in response to atmospheric
 4    deposition based on process-level information about acidification. A process model, such as
 5    MAGIC, characterizes acidification into (l)a section in which the concentrations of major ions
 6    are assumed to be governed by simultaneous reactions involving SC>42" adsorption, cation
 7    exchange,  dissolution-precipitation- speciation of aluminum, and dissolution-speciation of
 8    inorganic carbon; and (2) a mass balance section in which the flux of major ions to and from the
 9    soil is assumed to be controlled by atmospheric inputs, chemical weathering, net uptake and loss
10    in biomass and losses to runoff. At the heart of MAGIC is the size of the pool of exchangeable
11    base cations in the soil. As the fluxes to and from this pool change over time owing to changes in
12    atmospheric deposition, the chemical equilibria between soil and soil solution shift to give
13    changes in surface water chemistry. The degree and rate of change of surface water acidity thus
14    depend both on flux factors and the inherent characteristics of the affected soils.
15          There are numerous input data required to run MAGIC making it rather data intensive.
16    Atmospheric deposition fluxes for the base cations and strong acid anions are required as inputs
17    to the model. These inputs are generally assumed to be uniform over the catchment. The volume
18    discharge for the catchment must also be provided to the model. In general, the model is
19    implemented using average hydrologic conditions and meteorological conditions in annual
20    simulations, i.e., mean annual deposition, precipitation and lake discharge are used to drive the
21    model. Values for soil and surface water temperature, partial pressure of carbon dioxide and
22    organic acid concentrations must also be provided at the appropriate temporal resolution.
23          The aggregated nature of the model requires that it be calibrated to observed data from a
24    system before it can be used to examine potential system response. Calibrations are based on
25    volume weighted mean annual or seasonal fluxes for a given period of observation. The length of
26    the period  of observation used for calibration is not arbitrary. Model output will be more reliable
27    if the annual flux estimates used in calibration are based on a number of years rather than just
28    one year. There is a lot of year-to-year variability in atmospheric deposition and catchment
29    runoff. Averaging over a number of years reduces the likelihood that an "outlier" year (very dry,
30    etc.) is used to specify the primary data on which model forecasts are based. On the other hand,
      March 2010                                 9                 Draft - Do Not Quote or Cite

-------
                                                                                     Appendix A

 1   averaging over too long a period may remove important trends in the data that need to be
 2   simulated by the model.
 3          The calibration procedure requires that stream water quality, soil chemical and physical
 4   characteristics, and atmospheric deposition data be available for each catchment. The water
 5   quality data needed for calibration are the concentrations of the individual base cations (Ca, Mg,
 6   Na, and K) and acid anions (Cl, SC>42", and N(V) and the pH. The soil data used in the model
 7   include soil depth and bulk density,  soil pH, soil cation-exchange capacity, and exchangeable
 8   bases in the soil (Ca, Mg, Na, and K). The atmospheric deposition inputs to the model must be
 9   estimates of total deposition, not just wet deposition. In some instances, direct measurements of
10   either atmospheric deposition or soil properties may not be available for a given site with stream
11   water data. In these cases, the required data can  often be estimated by: (a) assigning  soil
12   properties based on some landscape classification of the catchment; and (b) assigning deposition
13   using model extrapolations from some national or regional atmospheric deposition monitoring
14   network. Soil data for model calibration are usually derived as aerially averaged values of soil
15   parameters within a catchment. If soils data for a given location are vertically stratified, the soils
16   data for the individual soil horizons  at that sampling site can be aggregated based on horizon,
17   depth, and bulk density to obtain single vertically aggregated values for the site, or the stratified
18   data can be used directly in the model.
19
     March 2010                                 10                Draft - Do Not Quote or Cite

-------
 2

 3

 4      Methodologies for National Terrestrial and Aquatic

 5                Acidification Maximum Depositional Load

 e              Approaches: Determining Weathering Rates

 7

 8

 9                                First External Review Draft

10
11
12
13                                                      Prepared for:
14
15                                     U.S. Environmental Protection Agency
16                                 Office of Air Quality Planning and Standards
17                                        Research Triangle Park, NC 27711
18
19
20                                                      Prepared by:
21
22                                                   RTI International
23                                                     P.O. Box 12194
24                                        Research Triangle Park, NC 27709
25
26                                       EPA Contract Number EP-D-06-003
27                                      RTI Project Number 0209897.004.080
28
29
30
31                                                    January 11, 2009
32
33
                                                     HRTI
                                                     INTERNATIONAL

-------

-------
                                                                                     Appendix B

 1                                        Table of Contents
 2
 3    1.  PURPOSE	1
 4    2.  OVERVIEW OF ACIDIFICATION	1
 5         2.1   EPA's Integrated Science Assessment and Risk and Exposure Assessment	1
 6         2.2   Aquatic Acidification and Critical Acid Loads	4
 7          2.1.2  Terrestrial Acidification and Critical Acid Loads	9
 8    3.  AQUATIC BASE CATION WEATHERING METHODOLOGY	14
 9         3.1   Aquatic Base Cation Weathering	14
10         3.2   Methodologies for Determining Base Cation Weathering Values in the United States	16
11          3.2.1  Difficulties in estimating base cation weathering	16
12          3.2.2  Approaches to estimating BCW for Aquatic Acidification	17
13         3.3   Proposed Methodology for Estimating and Mapping Base Cation Weathering for Aquatic
14              Critical Acid Load Calculations	27
15          3.3.1  Potential limitations of proposed methodology	33
16          3.3.2  Uncertainty analyses	34
17    4.  TERRESTRIAL BASE  CATION WEATHERING METHODOLOGY	35
18         4.1   Introduction	35
19         4.2   Terrestrial Base Cation Weathering	36
20         4.3   Methodologies for Determining Base Cation Weathering Values in the United States	39
21          4.3.1  Difficulties in estimating base cation weathering	39
22          4.3.2  Approaches to estimating BCw:	40
23          4.3.3  Proposed methodology for estimating and mapping base cation weathering for
24                terrestrial critical acid load calculations	50
25          4.3.5  Potential limitations of proposed methodology	78
26          4.3.6  "Field Tests" of model and uncertainty analyses	79
27    5.  CONCLUSIONS AND RECOMMENDATIONS	81
28    6.  REFERENCES	82
29    APPENDIX 1 Potentially Applicable National-Scale Geochemical Data	97
30    APPENDIX 2 References for Table 3-2: Applications of the MAGIC Model	102
31
32
     March 2010                                  i                    Draft - Do Not Quote or Cite

-------
                                                                                         Appendix B
                                             List of Figures
 2    2-1. (a) Number offish species per lake or stream versus acidity, expressed as acid neutralizing
 3           capacity for Adirondack Case Study Area lakes (Sullivan et al., 2006). (b) Number offish
 4           species among 13 streams in Shenandoah National Park. Values of acid neutralizing
 5           capacity are means based on quarterly measurements from  1987 to 1994. The regression
 6           analysis shows a highly significant relationship (p < .0001) between mean stream acid
 7           neutralizing capacity and the number offish species	5
 8    2-2. The relationship between the Bc/Al ratio in soil solution and the percentage of tree species
 9           (found growing in North America - native and introduced species) exhibiting a 20%
10           reduction in growth relative to controls (after Sverdrup and Warfvinge, 1993)	10
11    3-1. Process steps for estimating BCW using the MAGIC model with regional extrapolation	28
12    4-1. Areas of continental U.S. that were covered during the last glacial event (Reed and Bush,
13           2005)	39
14    4-2. Process Steps for Estimating BCW Using the PROFILE Regional Model	52
15    4-3. Map Showing the Distribution and Status of SSURGO Data	64
16    4-4. Soil Sampling Locations Included in the USGS Shacklette Dataset	66
17    4-5. Sample Density of USGS National Geochemical Survey	68
18    4-6. NRCS Soil Pedon Sample Pit Locations (30,000 total)	70
19    4-7. NRCS Soil Pedon Pit Sample Locations with Geochemical and Mineralogy Data	71
20
21
22                                           List of Tables
23    2-1. Aquatic Status Categories	6
24    2-2. Summary of Linkages between Acidifying Deposition, Biogeochemical Processes That
25           Affect Ca2+, Physiological Processes That Are Influenced by Ca2+, and Effect on Forest
26           Function	11
27    2-3. The Three Indicator (Bc/Al)cnt Soil Solution Ratios and Corresponding Levels of Protection
28           to Tree Health and Critical Loads	14
29    3-1. Review of Modeling Approaches (and models) to Estimate Base Cation Weathering for
30           Aquatic Critical Acid Load Determinations	23
31    3-2. Locations of Previous MAGIC Applications within the U.S. and Canada1	29
32    3-3. Input Data Requirements of MAGIC Model	31
33    4-1. Review of modeling approaches (and models) to estimate base cation weathering for
34           terrestrial critical acid load determinations	46
35    4-2. The fourteen dominant minerals modeled within PROFILE	50
36    4-3a. Data required to estimate BCW with the regional PROFILE model (version 5.0). The data in
37           this table must be input by the user and are  currently available as a continuous coverage
38           layers for at least a portion of the conterminous United States	53
39    4-3b. Data required to estimate BCW with the regional PROFILE model (version 5.0). The data in
40           this table must be input by the user and are  not currently available as a continuous
41           coverage layers for at least a portion of the  conterminous United  States (will require
42           development of national coverage layer)	53
      March 2010                                    ii                     Draft - Do Not Quote or Cite

-------
                                                                                          Appendix B

 1    4-3c. Data required to estimate BCW with the regional PROFILE model (version 5.0). The data in
 2           this table are used to support calculations within the model and should be reviewed by the
 3           user	54
 4    4-4. Available datasets and databases for the conterminous United States that could be used to
 5           estimate BCW with the regional application of the PROFILE model (version 5.0)	55
 6    4-5. Nitrogen and base cation uptake by forest type (from McNulty et al., 2007)	58
 7    4-6 Datasets with Geochemical and Mineralogy Data for US Soils	61
 8    4-7. Long-Term Ecological Research (LTER) sites that could potentially be suitable as "field
 9           test" sites to validate BCW estimates generated with the regional application of the
10           PROFILE model (version 5.0)	74
11
12
      March 2010                                    iii                    Draft - Do Not Quote or Cite

-------
                                                                                  Appendix B
1
2
    March 2010                                 iv                  Draft - Do Not Quote or Cite

-------
                                                                                     Appendix B

 1    1.     PURPOSE
 2          The purpose of this Work Assignment Task is to develop methodologies for estimating
 3    national terrestrial and aquatic acidification maximum depositional loads. Separate approaches
 4    are developed for terrestrial and aquatic acidification because biogeochemical processes in
 5    aquatic and terrestrial ecosystems for nitrogen and sulfur are not identical. Information about the
 6    key physical, chemical, and biological parameters needed to predict acidification potential in
 7    ecosystems is not always available. For example, weathering rates are key to acidification but are
 8    not available in all parts of the U.S. Knowledge of an ecosystem's weathering characteristics
 9    enables a more accurate assessment of whether acidifying deposition can be neutralized or
10    exceeds an ecosystem's critical load beyond which negative effects in aquatic and terrestrial
11    health may occur.
12          This report presents an introduction to aquatic and terrestrial acidification, followed by
13    reviews of different approaches to estimating base cation weathering and detailed methodologies
14    that could be used to estimate base cation weathering for aquatic and terrestrial critical load
15    calculations.

16    2.     OVERVIEW OF ACIDIFICATION
17    2.1    EPA's Integrated Science Assessment and Risk and Exposure Assessment
18          Deposition of SOX, NOX, and NHX can lead to ecosystem exposure to acidification. The
19    Integrated Science Assessment (ISA) for Oxides of Nitrogen and Sulfur-Ecological Criteria
20    (FinalReport) (ISA) (U.S. EPA, 2008) reports that acidifying deposition has altered major
21    biogeochemical processes in the United States by increasing the sulfur and nitrogen content of
22    soils, accelerating sulfate (SC>42 ) and nitrate (NOs ) leaching from soil to drainage water,
23    depleting soil exchangeable base cations (especially calcium [Ca2+] and magnesium [Mg2+])
24    from  soils, and increasing the mobility  of aluminum (Al) within the soil (U.S. EPA, 2008,
25    Section 3.2.1)
26          The extent of soil acidification is a critical factor that regulates virtually all acidification-
27    related ecosystem effects from sulfur and nitrogen deposition. Soil acidification occurs in
28    response to both natural factors and acidifying deposition (U.S. EPA, 2008, Section 3.2.1).
29    Under conditions of low atmospheric deposition of nitrogen and sulfur, the naturally produced

      March 2010                                   1                    Draft - Do Not Quote or Cite

-------
                                                                                     Appendix B

 1   bicarbonate anion is often the dominant mobile anion, with SC>42" and N(V playing a limited role
 2   with respect to cation leaching. Increased atmospheric deposition of sulfur and nitrogen can
 3   result in marked increases in SC>42" and NCV soil fluxes resulting in the concomitant leaching of
 4   base cations (Ca2+, Mg2+) and toxic cations (Aln+ and H+).
 5          Acidification can impact the health of terrestrial and aquatic ecosystems. One of the
 6   effects of soil acidification is the increased mobility of dissolved inorganic Al, which is toxic to
 7   tree roots, fish, algae, and aquatic invertebrates (U.S. EPA, 2008, Sections 3.2.1.5, 3.2.2.1, and
 8   3.2.3).
 9          The changes in major biogeochemical processes and soil conditions caused by acidifying
10   deposition have significant ramifications for the water chemistry and biological functioning of
11   associated surface waters. Surface water chemistry indicates the negative effects of acidification
12   on the biotic integrity of freshwater ecosystems. Surface water chemistry integrates the sum of
13   terrestrial and aquatic processes that occur upstream within a watershed. Important terrestrial
14   processes include nitrogen saturation, forest decline, and soil acidification (Stoddard et al.,
15   2003). Thus, water chemistry integrates and reflects changes in soil and vegetative properties and
16   biogeochemical processes (U.S. EPA, 2008,  Section 3.2.3.1).
17          Ecological effects occur at four levels of biological organization: (1) the individual; (2)
18   the population, which is composed of a single species of individuals; (3) the biological
19   community, which is composed of many species; and (4) the ecosystem. Low ANC
20   concentrations are linked with negative effects on aquatic systems at all four of these biological
21   levels. For the individual level, impacts are assessed in terms of fitness (i.e., growth,
22   development, and reproduction) or sublethal effects on condition. Surface water with low ANC
23   concentrations can directly influence aquatic organism fitness or mortality by disrupting ion
24   regulation and can mobilize  dissolved inorganic aluminum, which is highly toxic to fish under
25   acidic conditions (i.e., pH <6 and ANC <50 ueq/L). For example, research showed that as the pH
26   of surface waters decreased to <6, many aquatic species, including fish, invertebrates,
27   zooplankton, and diatoms, tended to decline  sharply causing species richness to decline
28   (Schindler,  1988). Van Sickle and colleagues (1996) also found that blacknose dace (Rhinichthy
29   spp.) were highly sensitive to low pH and  could not tolerate inorganic Al concentrations greater
30   than about 3.7 micromolar (uM) for extended periods of time. For example, they found that after
     March 2010                                  2                    Draft - Do Not Quote or Cite

-------
                                                                                    Appendix B

 1   6 days of exposure to high inorganic Al, blacknose dace mortality increased rapidly to nearly
 2   100%.
 3          At the community level, species richness and community structure can be used to
 4   evaluate the effects of acidification. Species composition refers to the mix of species that are
 5   represented in a particular ecosystem, whereas species richness refers to the total number of
 6   species in a stream or lake. Acidification alters species composition and richness in aquatic
 7   ecosystems. There are a number of species common to many oligotrophic waterbodies that are
 8   sensitive to acidification and cannot survive, compete, or reproduce in acidic waters. In response
 9   to small  to moderate changes in acidity,  acid-sensitive species are often replaced by other more
10   acid-tolerant species, resulting in changes in community composition and richness, but with little
11   or no change in total community biomass. The effects of acidification are continuous, with more
12   species being affected at higher degrees  of acidification. At a point, typically a pH <4.5 and an
13   ANC <0 ueq/L, complete to near-complete loss  of many classes of organisms occur, including
14   fish and  aquatic insect populations, whereas others are reduced to only  a few acidophilic forms.
15   These changes in species integrity are because energy cost in maintaining physiological
16   homeostasis, growth, and reproduction is high at low ANC levels (Schreck, 1981, 1982;
17   Wedemeyer et al., 1990).
18          In EPA's Risk and Exposure Assessment for Review of the Secondary National Ambient
19   Air Quality Standards for Oxides of Nitrogen and Sulfur (U.S. EPA, 2009), the negative impacts
20   of acidifying deposition were assessed by conducting case studies of 1) aquatic acidification in
21   Adirondack Mountains lakes and Shenandoah Mountains  streams, and  2) terrestrial acidification
22   in red spruce and sugar maple forests in  the White Mountains of New Hampshire and in
23   Pennsylvania, respectively. The results of these  case studies revealed the significance of base
24   cation weathering in predicting aquatic and terrestrial acidification impacts. The results further
25   highlighted the need to select weathering methodologies that can be applied across geologically
26   diverse ecosystems in the United States.  This report uses the information from the Risk and
27   Exposure Assessment as a starting point to identify and evaluate approaches to predicting
28   weathering at other locations and larger  scales in the United States. In this report, RTI
29   recommends methodologies (including computer models) for application in the United States,
30   assesses the availability of input data for those methodologies, identifies potential remedies to
     March 2010                                  3                    Draft - Do Not Quote or Cite

-------
                                                                                    Appendix B

 1   limited data availability, and describes uncertainties with the methodologies in predicting
 2   acidification impacts.
 3   2.2    Aquatic Acidification and Critical Acid Loads
 4          Surface water chemistry is a primary indicator of acidification and the resulting negative
 5   effects on the biotic integrity of freshwater ecosystems. Chemical parameters can be used to
 6   assess effects of acidifying deposition on lake or stream acid-base chemistry. These receptors
 7   include surface water pH and concentrations of SC>42", NCV, Al, and Ca2+; the sum of base
 8   cations; and the recently developed base cation surplus. Another widely used water chemistry
 9   indicator for both atmospheric deposition sensitivity and effects is acid neutralizing capacity
10   (ANC). The utility of the ANC criterion lies in the association between ANC and the surface
11   water constituents that directly contribute to or ameliorate acidity-related stress, in particular pH,
12   Ca2+, and Al. ANC is also used because it integrates overall acid status and because surface
13   water acidification models do a better job projecting ANC than they do for projecting pH and
14   dissolved inorganic Al concentrations.
15          For the purpose of this study,  ANC of surface waters is simply measured as the total
16   amount of strong base ions minus the total amount of strong acid anions:

17                   ANC = (Ca2++ Mg2++ K++ Na++ NH4) - (SO42" + NCV+CO           (2-1)

18          The unit of ANC is usually microequivalents per liter (ueq/L). If the sum of the
19   equivalent concentrations of the base cations exceeds those of the strong acid anions, then the
20   ANC of a waterbody will be positive. To the extent that the base cation sum exceeds the strong
21   acid anion sum, the ANC will be higher. Higher ANC is generally associated with high pH and
22   Ca2+ concentrations; lower ANC is generally associated with low pH and A13+ concentrations and
23   a greater likelihood of toxicity to biota.
24          Low ANC coincides with effects on aquatic systems (e.g., individual species fitness loss
25   or death, reduced species richness, altered community structure). At the community level,
26   species richness is positively correlated with pH and ANC (Kretser et al., 1989; Rago and
27   Wiener, 1986) because energy cost in maintaining physiological homeostasis, growth, and
28   reproduction is high at low ANC levels (Schreck, 1981, 1982; Wedemeyer et al.,  1990). For
29   example, Sullivan and colleagues (2006) found a logistic relationship between fish species
30   richness and ANC class for Adirondack Case Study Area lakes (Figure 2-la) that indicates the

     March 2010                                  4                   Draft - Do Not Quote or Cite

-------
                                                                                     Appendix B
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
probability of occurrence of an organism for a given value of ANC. In the Shenandoah Case
Study Area, a statistically robust relationship between acid-base status of streams and fish
species richness was also documented (Figure 2-lb). In fact, ANC has been found in various
studies to be the best single indicator of the biological response and health of aquatic
communities in acid-sensitive systems (Lien et al., 1992; Sullivan et al., 2006).
       Biota are generally not harmed when ANC values are >100 microequivalents per liter
(ueq/L). The number offish species also peaks at ANC values >100 ueq/L (Bulger et al., 1999;
Driscoll et al., 2001; Kretser et al.,  1989; Sullivan et al., 2006). Below 100 ueq/L,  ANC fish
fitness and  community diversity begin to decline (Figure 2-1). At ANC levels between 100 and
50 ueq/L, the fitness of sensitive species (e.g., brook trout, zooplankton) also begins to decline.
When ANC concentrations are <50 ueq/L, they are generally associated with death or loss of
fitness of biota that are sensitive to acidification (Kretser et al., 1989; Dennis and Bulger, 1995).
l-,\
Number of Fish Specicsj
k PO Q1 N) i^ £Ti £U ft pyj
1 J 1 1 1 1 L 1
1
Jp0
--..— J i
•^wfi^^
"
f 1 — — i 	 1 	 1 	 1 	 1 —
-200 -100 0 100 200 300 400 5<
ANC(ueq/U
(b)


3
3 . .
M 5
Jl
& 1 -
u.
1
2
i 2
z 2
1 -
in n •



! ~^_
!-"' \
\.'\ i

£
\x
U* 1
^1
t
|
\ M
|1



flf! \ !!<
t i

: | §
1 i
t i
•25 0 25 50 TZ ICO 115 150 17S 200 225 250 27
Average AUC (jieq/L]
       Figure 2-1. (a) Number offish species per lake or stream versus acidity, expressed as
       acid neutralizing capacity for Adirondack Case Study Area lakes (Sullivan et al., 2006).
       (b) Number of fish species among 13 streams in Shenandoah National Park. Values of
       acid neutralizing capacity are means based on quarterly measurements from 1987 to
       1994. The regression analysis shows a highly significant relationship (p < .0001) between
       mean stream acid neutralizing capacity and the number of fish species.
       When ANC levels drop to <20 ueq/L, all biota exhibit some level of negative effects.
Fish and plankton diversity and the structure of the communities continue to decline sharply to
levels where acid-tolerant species begin to outnumber all other species (Matuszek and Beggs,
1988; Driscoll et al., 2001). Stoddard and colleagues (2003) showed that to protect biota from
episodic acidification in the springtime, base flow ANC levels need to have an ANC of at least
     March 2010
                                                                 Draft - Do Not Quote or Cite

-------
                                                                                        Appendix B


 1    30-40 ueq/L. Complete loss offish populations and extremely low diversity of planktonic

 2    communities occur when ANC levels stay <0 ueq/L. Only acidophilic species are present, but

 3    their population numbers are sharply reduced (Sullivan et al., 2006).

 4           The critical load approach can be used to connect current deposition of nitrogen and

 5    sulfur to the acid-base condition and biological risk to biota of lakes and streams in the study

 6    through the defined ANC thresholds. Calculating critical load exceedances (i.e., the amount of

 7    deposition above the critical load) allows the determination of whether current deposition poses a

 8    risk of acidification to a given group of waterbodies. Low critical load values (i.e., less than 50

 9    meq/m2 yr) mean that the watershed has a limited ability to neutralize the addition of acidic

10    anions, and hence, it is susceptible to acidification. The greater the critical load value, the greater

11    the ability of the watershed to neutralize the additional acidic anions and protect aquatic life.

12    This approach also allows for the comparison of different levels of ANC thresholds (e.g., 0

13    ueq/L (acidic), 20  ueq/L (minimal protection), 50  ueq/L (moderate protection), and 100 ueq/L

14    (full protection)) and their associated risk to the biological community. Table 2-1 provides a

15    summary of the biological effects experienced at each of these limits.


      Table 2-1. Aquatic Status Categories
                         Category Label ANC Levels* Expected Ecological Effects
       Acute
       Concern
<0 micro
equivalent
per Liter
(ueq/L)
Near complete loss offish populations is expected. Planktonic
communities have extremely low diversity and are dominated by
acidophilic forms. The number of individuals in plankton species that
are present is greatly reduced.
       Severe
       Concern
0-20 ueq/L
Highly sensitive to episodic acidification. During episodes of high
acidifying deposition, brook trout populations may experience lethal
effects. Diversity and distribution of zooplankton communities decline
sharply.
       Elevated
       Concern
20-50 ueq/L
Fish species richness is greatly reduced (i.e., more than half of
expected species can be missing). On average, brook trout populations
experience sublethal effects, including loss of health, reproduction
capacity, and fitness. Diversity and distribution of zooplankton
communities decline.
       Moderate
       Concern
50-100
ueq/L
Fish species richness begins to decline (i.e., sensitive species are lost
from lakes). Brook trout populations are sensitive and variable, with
possible sublethal effects. Diversity and distribution of zooplankton
communities also begin to decline as species that are sensitive to
acidifying deposition are affected.
      March 2010
                                                    Draft - Do Not Quote or Cite

-------
                                                                                  Appendix B
Category Label ANC Levels* Expected Ecological Effects
Low Concern
>100 |jeq/L
Fish species richness may be unaffected. Reproducing brook trout
populations are expected where habitat is suitable. Zooplankton
communities are unaffected and exhibit expected diversity and
distribution.
 1          There are numerous methods and models that can be used to calculate critical loads for
 2   acidity. Drawing on the peer-reviewed scientific literature (Dupont et al., 2005), this study will
 3   use a steady-state critical load model that uses surface water chemistry as the base for calculating
 4   the critical load. A combination of the Steady-State Surface Water Chemistry (SSWC) and First-
 5   Order Acidity Balance (FAB) models were used to calculate the critical load. Both the SSWC
 6   model  and FAB are based on the principle that excess base-cation production within a catchment
 7   area should be equal to or greater than the acid anion input, thereby maintaining the ANC above
 8   a preselected level (Reynolds and Norris, 2001; Posch et al., 1997). These models assume
 9   steady-state conditions and assume that all SO42 in runoff originates from sea salt spray and
10   anthropogenic deposition. Given a critical ANC protection level, the critical load of acidity is
11   simply the input flux of acid anions from atmospheric deposition (i.e., natural and
12   anthropogenic) subtracted from the natural (i.e., preindustrial) inputs of base cations in the
13   surface water.
14          Critical loads for nitrogen and sulfur (CL(N) + CL(S)) or critical load of acidity CL(A)
15   are calculated for each waterbody from the principle that the acid load should not exceed the
16   nonmarine, nonanthropogenic base cation input and sources and sinks in the catchment minus a
17   neutralizing to protect selected biota from being damaged:

18                  CL(N) + CL(S) or CL(A) = BC*dep + BCW - Bcu - AN - ANCiimit          (2-2)

19   where
20           BC dep =  nonanthropogenic deposition flux of base cations
21                     (BC*=Ca*+Mg*+K*+Na*)
22             BCW =  the average weathering flux, producing base cations
23             Bcu =  the net long-term average uptake flux of base cations (Bc=Ca*+Mg*+K*) in
24                     the biomass (i.e., the annual average removal of base cations due to
25                     harvesting)

     March 2010                                 7                   Draft - Do Not Quote or Cite

-------
                                                                                    Appendix B
 1              AN  =  the net long-term average uptake, denitrification, and immobilization of
 2                      nitrogen anions (e.g. N(V) and uptake of S(V
 3          ANCumit =  the lowest ANC-flux that protects the biological communities.
 4          In order to estimate a critical load from water quality data alone, a relation to the
 5   preacidification nonmarine flux of base cations for each lake or stream, BC „, is used.

 6                                   BC*0 = BC*dep + BCW - Bcu                           (2-3)

 7          Thus, the critical load for acidity can be rewritten as

 8               CL(N) + CL(S) = BC*0 - AN - ANC,,mt = Q.([ BC*]0 - [AN] - [ANC],,mt)       (2-4)

 9   where the second identity expresses the critical load for acidity in terms of catchment runoff (Q)
10   m/yr and concentration ([x] = X/Q). In cases where data are available, the FAB model is applied
11   to quantify the [AN] term of the critical load calculation (derivation provided in Appendix 4,
12   Attachment A of U.S. EPA, 2009). Where data are not available the contribution of nitrogen
13   anions to acidification was assumed to be equal to the nitrogen leaching rate into the surface
14   water. The flux of acid anions in the surface water is assumed to represent the amount of
15   nitrogen that is not retained by the catchment, which is determined from the sum of measured
16   concentration of N(V and ammonia in the stream chemistry. This case describes the SSWC
17   model and the critical load for acidity is

18                                 CL(A) = Q.([BO]o-[ANC]iimit)                         (2-5)

19   where the contribution of acid anions is considered  as part of the exceedances calculation. With
20   this approach several major assumptions are made:  (1) steady-state conditions exist, (2) the effect
21   of nutrient cycling between plants and soil is negligible, (3) there are  no significant nitrogen
22   inputs from sources other than atmospheric deposition, (4) ammonium leaching is negligible
23   because any inputs are either taken up by biota or adsorbed onto soils or nitrate compounds, and
24   (5) long-term sinks of sulfate in the catchment soils are negligible.
25          To determine a value for BC*0 with the SSWC method, estimates of BCdep are available
26   from previous works including the recent REA (U.S. EPA, 2009).  Assumptions or estimates for
27   BCU and AN can be made based on attributes of the area of study, including vegetation

     March 2010                                 8                    Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B

 1   characteristics. But the average flux of base cations weathered in a catchment and reaching the
 2   lake or streams (BCW) is difficult to measure or compute from available information (Henriksen
 3   and Posch, 2001; Henriksen et al., 2002; Langan et al., 2001). In the previous work for the Risk
 4   and Exposure Assessment case studies (U.S. EPA, 2009) the average flux of base cations and the
 5   resulting critical load estimation were derived from water quality data (Henriksen and Posch,
 6   2001; Henriksen et al., 1992; Sverdrup et al., 1990). Weighted annual mean water chemistry
 7   values were used to estimate average base cation fluxes, which were calculated from water
 8   chemistry data collected from several national  and regional monitoring programs. For a national
 9   assessment, however, new methods must be developed to estimate the BCW flux, which is critical
10   to the critical load  calculation, through consistent, nationally-applicable means.
11   2.1.2  Terrestrial Acidification and Critical Acid Loads
12          Due to the impact of acidifying nitrogen and sulfur deposition on soil solution base  cation
13   (Be) and aluminum concentrations, the Bc/Al ratio in the soil solution is often used as the
14   chemical or critical indicator of terrestrial  acidification. It was recently used as an indicator in the
15   U.S. EPA's Risk and Exposure Assessment for oxides of nitrogen and oxides of sulfur (U.S.
16   EPA, 2009). This Bc/Al ratio links acidifying deposition to biological responses or  end points,
17   such as reduced plant or tree growth, within an ecosystem. In a meta-analysis of studies that
18   explored the relationship between Bc/Al ratio in soil solution and tree growth, Sverdrup and
19   Warfvinge (1993a) reported the Bc/Al ratios at which growth was reduced by 20%  relative to
20   control trees. This  20% reduction in tree growth was selected as the critical value because it was
21   thought to represent a significant reduction in growth (H. Sverdrup personal communication,
22   2009b) and approximates the Bc/Al value  that  would result in a 10% reduction in normal tree
23   growth under field conditions (Sverdrup and Warfvinge, 1993a). Figure 2-2 presents the
24   findings of Sverdrup and Warfvinge (1993 a) based on 46 of the tree species (native and
25   introduced) that grow in North America. This summary indicates that there is a 50% chance of
26   negative tree response (i.e., >20% reduced growth) at a soil solution Bc/Al ratio of  1.2 and  a
27   75% chance at a Bc/Al ratio of 0.6. These  findings clearly demonstrate a relationship between
28   Bc/Al ratio and tree health; as the Bc/Al is reduced, there is a greater likelihood of a negative
29   impact on tree health.
     March 2010                                  9                   Draft - Do Not Quote or Cite

-------
                                                                                     Appendix B

10
a
o
"o
2 1 -
o
E/3
a
0
ta
o
PQ
.01 '
(


^\
V, ,,,,,.. (Bc/AlJrit = 1.2
%-^_^_^ 	
^^* ***** *~*^^ i UP/ Ah = n h
'v,. \J— '^' ^^^irit W.LJ
^^-

^^^\
\
V


) 25 50 75 1(
Cumulative Percentage of Species Exhibiting Reduced Growth Response












DO
 2          Figure 2-2. The relationship between the Bc/AI ratio in soil solution and the percentage of
 3          tree species (found growing in North America - native and introduced species) exhibiting
 4          a 20% reduction in growth relative to controls (after Sverdrup and Warfvinge, 1993).
 5          The tree species most commonly studied in North America to assess the impacts of
 6   acidification due to total nitrogen and sulfur deposition are red spruce (i.e., Picea Rubens) and
 7   sugar maple (i.e., Acer saccharum). Both species are found in the eastern United States, and soil
 8   acidification is widespread throughout this area (Warby et al., 2009). Based on the results from a
 9   compilation of laboratory studies, red spruce growth can be reduced by 20% at a Bc/AI soil
10   solution ratio of approximately 1.2, and a similar reduction in growth may be experienced by
11   sugar maple at a Bc/AI ratio of 0.6 (Sverdrup and Warfvinge 1993a).
12          Red spruce is found scattered throughout high-elevation sites in the Appalachian
13   Mountains, including the southern peaks. Noticeable fractions of the canopy red spruce died
14   within the Adirondack, Green, and White mountains in the 1970s and 1980s. Although a variety
15   of conditions, such as changes in climate and exposure to ozone, may  impact the growth of red
16   spruce (Fincher et al., 1989; Johnson et al.,  1988), acidifying deposition has been implicated as
17   one of the main factors causing this decline. Based on the research conducted to date, acidifying
18   deposition can cause a depletion of base cations in upper soil horizons, Al toxicity to tree roots,
19   and accelerated leaching of base cations from foliage (U.S. EPA, 2008, Section 3.2.2.3). Such
20   nutrient imbalances and deficiencies can reduce the ability of trees to respond to stresses, such as
21   insect defoliation, drought, and cold weather damage (DeHayes et al., 1999; Driscoll et al.,
     March 2010
10
Draft - Do Not Quote or Cite

-------
                                                                                     Appendix B
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
2001), thereby decreasing tree health and increasing mortality. Additional linkages between
acidifying deposition and red spruce physiological responses are indicated in Table 4.3-1.
Within the southeastern United States, periods of red spruce decline slowed after the 1980s,
when a corresponding decrease in SC>2 emissions, and therefore acidic deposition, was recorded
(Webster et al., 2004).
       Sugar maple is found throughout the northeastern United States and the central
Appalachian Mountain region. This species has been declining in the eastern United States since
the 1950s. Studies on sugar maple have found that one source of this decline in growth is related
to both acidifying deposition and base-poor soils on geologies dominated by sandstone or other
base-poor substrates (Bailey et al., 2004; Horsley et al., 2000). These site conditions are
representative of the conditions expected to be most susceptible to impacts of acidifying
deposition because of probable low initial base cation pools and high base cation leaching losses
(U.S. EPA, 2008, Section 3.2.2.3).  The probability of a decrease in crown vigor or an increase in
tree mortality has been noted to increase at sites  with low Ca2+ and Mg2+ as a result of leaching
caused by acidifying deposition (Drohan and Sharpe,  1997). Low levels of Ca2+ in leaves and
soils have been shown to be related to lower rates of photosynthesis and higher antioxidant
enzyme activity in sugar maple stands in Pennsylvania (St. Clair et al., 2005). In addition, plots
of sugar maples in decline were found to have Ca2+/Al ratios less than 1, as well as lower base
cation concentrations and pH values compared with plots of healthy sugar maples (Drohan et al.,
2002). Sugar maple regeneration has  also been noted to be restricted under conditions of low soil
Ca2+ levels (Juice et al., 2006). These indicators  have  all been shown to be related to the
deposition of atmospheric nitrogen and sulfur. Additional  linkages between acidifying deposition
and sugar maple physiological responses are indicated in Table 2-2.
Table 2-2. Summary of Linkages between Acidifying Deposition, Biogeochemical Processes That Affect
Ca2+, Physiological Processes That Are Influenced by Ca2+, and Effect on Forest Function
Biogeochemical Response to
Acidifying deposition
Leach Ca2+ from leaf membrane
Reduce the ratio of Ca2+/AI in
soil and soil solutions
Physiological Response
Decrease the cold tolerance of
needles in red spruce
Dysfunction in fine roots of red
spruce blocks uptake of Ca2+
Effect on Forest Function
Loss of current-year needles in
red spruce
Decreased growth and
increased susceptibility to stress
in red spruce
     March 2010
                                            11
Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B
Biogeochemical Response to
Acidifying deposition
Reduce the ratio of Ca2+/AI in
soil and soil solutions
Reduce the availability of
nutrient cations in marginal soils
Physiological Response
More energy is used to acquire
Ca2+ in soils with low Ca2+/AI
ratios
Sugar maples on drought-prone
or nutrient-poor soils are less
able to withstand stresses
Effect on Forest Function
Decreased growth and
increased photosynthetic
allocation to red spruce roots
Episodic dieback and growth
impairment in sugar maple
      Source: Fenn and colleagues, 2006.
 1          Although the main focus of the Terrestrial Acidification Case Study outlined in the Risk
 2   and Exposure Assessment for Review of the Secondary National Ambient Air Quality Standards
 3   for Oxides of Nitrogen and Sulfur (U.S. EPA, 2009) was an evaluation of the negative impacts of
 4   nitrogen and  sulfur deposition on soil acidification and tree health, it should be recognized that
 5   under certain conditions, nitrogen and sulfur deposition can have a positive impact on tree health.
 6   Nitrogen limits the growth of many forests (Chapin et al., 1993; Killam, 1994; Miller, 1988), and
 7   therefore, in such forests, nitrogen deposition may act as a fertilizer  and stimulate growth.
 8   Forests where critical acid loads are not exceeded by nitrogen and sulfur deposition could
 9   potentially be included within this  group of forests that respond positively to  deposition. These
10   potential positive growth impacts of nitrogen and sulfur deposition are discussed further, and the
11   results of case study analyses are presented in Attachment A of Appendix 5 of the Risk and
12   Exposure Assessment (U.S. EPA, 2009).
13          In summary, among potential influencing factors, including elevated ozone levels and
14   changes in climate, the acidification of soils is one of the factors that can negatively impact the
15   health  of red  spruce and sugar maple. Mortality and susceptibility to disease and injury can be
16   increased and growth decreased with acidifying deposition. Therefore, the health of sugar maple
17   and red spruce was used as the endpoints (ecological responses) to evaluate acidification in
18   terrestrial systems. "Health" in the context of the Risk and Exposure Assessment terrestrial
19   acidification case study was defined as the physiological condition of a tree that impacts growth
20   and/or mortality.
21          The Simple Mass Balance (8MB) model was used to estimate critical loads of acidity in
22   the Risk and Exposure Assessment case study (Equation 2-1). The full derivation of this equation
23   is detailed in  the TCP Mapping and Modeling Manual (UNECE, 2004).
     March 2010
12
Draft - Do Not Quote or Cite

-------
                                                                                     Appendix B

 1                 CL(S + N) = BCdep - Cldep + BCW -Bcu + N, + Nu + Nde - ANCle,cnt          (2-1)

 2   where
 3            CL(S+N) = forest soil critical load for combined nitrogen and sulfur acidifying
 4                      deposition ((N+S)comb)
 5               BCdep = base cation (Ca2+ + K+ + Mg2+ + Na+) deposition1
 6                Cldep = chloride deposition
 7                BCW = base cation (Ca2+ + K+ + Mg2+ + Na+) weathering
 8                 Bcu = uptake of base cations (Ca2+ + K+ + Mg2+) by trees
 9                   N; = nitrogen immobilization
10                  Nu = uptake of nitrogen by trees
11                 Nde = denitrification
12            ANCie,crit = forest soil acid neutralizing capacity of critical load leaching

13   Some of these parameters had defined or selected input values (BCdep, Cldep, N;, Nu and Nde),
14   while four of these parameters, including BCW, Bcu, Nu and ANCie,Crit, required calculation.
15          For the Risk and Exposure Assessment's terrestrial acidification case study, three values
16   of the indicator of critical load, expressed as (Bc/Al)crit soil solution ratio, were selected to
17   represent different levels of tree protection associated with total nitrogen and sulfur deposition:
18   0.6, 1.2, and 10 (Table 2-3). The (Bc/Al)crit ratio of 0.6 represents the highest level of impact
19   (lowest level of protection) to tree health and growth and was selected because 75% of species
20   found growing in North America experience reduced growth at this Bc/Al ratio. In addition, a
21   soil solution Bc/Al ratio of 0.6 has been linked to a 20% and 35% reduction in sugar maple and
22   red spruce growth, respectively. The (Bc/Al)crit ratio of 1.2 is considered to represent a moderate
23   level of impact, as the growth of 50% of tree species (found growing in North America) was
24   negatively impacted at this soil solution ratio. The (Bc/Al)crit ratio of 10.0 represents the lowest
25   level of impact (greatest level of protection) to tree growth; it is the most conservative value used
26   in studies that have calculated critical loads in the United States and Canada (Canada (McNulty
27   et al., 2007; NEG/ECP, 2001; Watmough et al., 2004).
      1 The ICP Mapping and Modeling Manual (UNECE, 2004) recommends that wet deposition be corrected for sea salt
      on sites within 70 km of the coast. Both the HBEF and KEF case study areas are greater than 70 km from the coast,
      so this correction was not used.

      March 2010                                  13                   Draft - Do Not Quote or Cite

-------
                                                                                     Appendix B
     Table 2-3. The Three Indicator (Bc/AI)crit Soil Solution Ratios and Corresponding Levels of Protection to
     Tree Health and Critical Loads
Indicator (Bc/AI)crit Soil
Solution Ratio
0.6
1.2
10.0
Level of Protection to Tree
Health
Low
Intermediate
High
Critical Load
High
Intermediate
Low
 1          The prediction of tree protection achieved using each of these three indicator ratios of
 2   0.6, 1.2, and  10.0 includes an important estimation of base cation weathering as shown in
 3   Equation 2-1, above. The purpose of this report is to describe the methodologies, data
 4   requirements, data availability, and uncertainties associated with estimating base cation
 5   weathering.

 6   3.     AQUATIC BASE CATION WEATHERING METHODOLOGY
 7          The ISA (US EPA, 2008) reports that the principal factor governing the sensitivity of
 8   terrestrial and aquatic ecosystems to acidification from sulfur and nitrogen deposition is geology
 9   (particularly  surficial geology). Geologic formations having low base cation supply generally
10   underlie the watersheds of acid-sensitive lakes and streams. Other factors that contribute to the
11   sensitivity of soils and surface waters to acidifying deposition include topography, soil
12   chemistry, land use, and hydrologic flowpath. Surface waters in the same setting can  have
13   different sensitivities to acidification, depending on the relative contributions of near-surface
14   drainage water and deeper groundwater (Chen et al., 1984; Driscoll et al.,  1991; Eilers et al.,
15   1983). Lakes and streams in the United States that are sensitive to episodic and chronic
16   acidification  in response to SOX,  and to a lesser extent NOX, deposition tend to occur at relatively
17   high elevation in areas that have  base-poor bedrock, high relief,  and shallow soils (U.S. EPA,
18   2008, Section 3.2.4.1).
19   3.1    Aquatic Base Cation Weathering
20          Base  cation weathering for aquatic acidification critical loads must be representative of
21   the catchment around the waterbody of interest. This aspect of quantification of the weathering
22   rate provides the difference when calculating weathering rates for aquatic versus terrestrial
23   analysis purposes. The process of weathering itself provides the only natural in-soil source of
24   alkalinity that is available to neutralize acidity inputs to the system over the long term. Chemical
25   weathering of the mineral matrix within soils supplies base cations that are removed from soil
     March 2010
14
Draft - Do Not Quote or Cite

-------
                                                                                    Appendix B

 1   due to acid inputs. Therefore, the rate of weathering of the soils within a catchment is dependent
 2   on the chemical and physical properties of the soil (Sverdrup et al., 1992; Whitfield et al., 2006).
 3   As indicated in Section 2.2, the average flux of base cations weathered in a catchment and
 4   reaching the lake or streams (BCW) is difficult to measure or compute from available information
 5   (Henriksen and Posch, 2001; Henriksen et al., 2002; Langan et al., 2001). Approaches also differ
 6   based on whether the weathering rate needs to account for only in-soil processes (profile
 7   measurements and models) or whether it needs to account for the flux of base cations to surface
 8   water (spatially integrated catchment data and models) (Langan et al., 2001).
 9          In the Aquatic Acidification case study in the REA Report (U.S. EPA, 2009), BCW rates
10   were not directly calculated. Instead, the F-factor approach was used to calculate the pre-
11   acidification, non-marine flux of base cations (BC*0) for each lake or stream. An F-factor
12   (explained in Section 3.2.2) is a ratio of the change in non-marine base cation concentration due
13   to changes in strong acid anion concentrations (Henriksen, 1984; Brakke et al.,  1990), as shown
14   in the following equations:

15                         BC*0 = BC*t - F (SO*4,t - SO*4,o + NO*3,t - NO*3)0)                 (3-1)

16                     F = ([BC*]t - [BC*]0)/([S04*]t -  [S04*]o + [NO3*]t - [NO3*]o)             (3-2)

17   where the subscripts t and 0 refer to present and pre-acidification conditions, respectively. The
18   pre-acidification N(V concentration, NO*3j0, was assumed to be zero. Several attempts have
19   been made to create empirical relations for the F-factor and the pre-acidification SO/
20   concentration. Although the Aquatic Acidification case study  relied on two of these relations, it
21   must be noted that they were developed for areas outside of the U.S. and, therefore, cannot be
22   applied to the conditions found within U.S. soils and climates without introducing a source of
23   uncertainty (Henriksen and Posch, 2001; Henriksen et al.,  2002; Brakke et al., 1989; Posch et al.,
24   1997). Notwithstanding the lack of U.S.-based empirical relations, the F-factor  can be used to
25   derive BCW  estimates. Assuming that all atmospheric deposition of base cations that falls within
26   a catchment passes through to the surface water and that one can accurately estimate the uptake
27   of base cations within the catchment, the BCW could ultimately be backed out of these
28   relationships. However, both of these assumptions are likely to introduce an additional amount of
29   uncertainty into the BCW estimates.

     March 2010                                  15                   Draft - Do Not Quote or Cite

-------
                                                                                     Appendix B

 1          For a national aquatic acidification assessment, different methods must be employed to
 2   estimate BCW rates. In some studies, simple assumptions for the BCW are utilized. For instance, in
 3   a study by Dupont and colleagues (2005) using the SSWC, the authors assumed that weathering
 4   rates were time-independent and did not affect critical load estimates. In more advanced process
 5   modeling applications, such as ones using the Model of Acidification of Groundwater in
 6   Catchments (MAGIC), weathering rates can be adjusted during calibration and allowed to vary
 7   over ranges like 0 and 5 times the observed watershed base cation export for base cation
 8   weathering (Sullivan et al., 2004). There are several different approaches to estimating the
 9   weathering rate of a soil or a catchment, ranging from empirical relations to mass balance
10   methods to calibrated process models. According to Whitfield and colleagues (2006) "to date no
11   method has proven to be superior in application to different soil types and differing levels of soil
12   acidification." The remainder of this section is intended to examine  the BCW estimation methods
13   that would be applicable to a national aquatic acidification critical loads analysis giving
14   consideration to the limitations of the method and the possible data  and processing requirements
15   for the analysis.
16   3.2    Methodologies for Determining Base Cation Weathering Values in the United States
17   3.2.1  Difficulties in estimating base cation weathering
18          Consideration must be given to several factors in the estimation of base cation weathering
19   fluxes for aquatic acidification (Sverdrup et al.,  1992; Whitfield et al., 2006; Rapp and Bishop,
20   2009; Henriksen and Posch, 2001; Henriksen et al., 2002):
21          1. The weathering contribution of the entire catchment must be  understood and not
22             simply the weathering contribution of certain soil profiles within the catchment.
23             Additionally, the various types of land use (e.g. agriculture or forest) within a
24             catchment may all affect weathering rates differently.
25          2. When utilizing soil profile weathering methods, the  characteristics of the entire soil
26             profile must be considered and weighted according to catchment composition as
27             opposed to only the rooting zone in individual profiles as used in determining
28             weathering for terrestrial acidification purposes.
29          3. Based on the critical load method chosen, it is often necessary to assume that the BCW
30             remains constant over the length of the analysis. While this simplifies the estimation
31             of BCW, it introduces uncertainty into any analysis. The length of the analysis

     March 2010                                  16                   Draft - Do Not Quote or Cite

-------
                                                                                     Appendix B

 1              scenario must be sufficiently long and have supporting data in order to provide a
 2              long-term average, which is not subjected to short-term variations.
 3          4.  The data requirements for a national assessment necessitate using similar data sources
 4              for all applications so that assumptions and methods can remain constant across the
 5              nation.
 6          5.  The application of any empirical relations for calculation of BCW or intermediate
 7              component of BCW (e.g., the F-factor) must be validated against the geographic region
 8              in which they will be applied. Given that most empirical relations developed to date
 9              were based on data from European nations, these relations need to be recalibrated to
10              data from the U.S.
11          Given all of these factors, estimation of BCW for a national application poses a significant
12    challenge. The methods detailed in the following section seek to balance the limitations and
13    benefits of each approach to estimation of BCW.
14    3.2.2  Approaches to estimating BCW for Aquatic Acidification
15          Work presented in the scientific literature over the last two decades provides several
16    different approaches researchers have taken to estimate the BCW rates for aquatic effects. These
17    approaches do not always differentiate between the actual weathering processes in-soil and the
18    other ion  exchange processes taking place (Langan et al., 2001). Approaches to estimating BCW
19    also vary  between terrestrial and aquatic studies. Aquatic studies of acidification must capture
20    the weathering rates of all soil horizons which contribute base cations and not solely the rooting
21    zone as specified in terrestrial  acidification studies (Whitfield et al., 2006).
22          Four  general categories of approaches are outlined for determining BCW for aquatic
23    acidification critical loads calculations using the SSWC.
24          1.  Budgets studies of catchments or watersheds;
25          2.  Historical weathering rate determinations;
26          3.  Empirical relations; and
27          4.  Process-based  models.
28          In the case of empirical data relations and process-based models, specific methods are
29    provided. The strengths and weaknesses of either the general category or specific approach in
30    terms of both utilization in aquatic acidification critical loads calculations and estimation of BCW
31    are examined in the following paragraphs and Table 3-1.

      March 2010                                 17                   Draft - Do Not Quote or Cite

-------
                                                                                    Appendix B

 1          Budget Studies - Budget studies are simple means of determining fluxes within a system
 2   by balancing the masses coming into and going out of a system. In determining the BCW, a mass
 3   balance would be performed around the base cations fluxes within a watershed, where
 4   atmospheric deposition constitutes the main source input and streamflow the main output. Within
 5   the balance, base cation retention is also accounted for through uptake by biomass and
 6   immobilization in the soil. The BCW developed from budget studies represent integrated values
 7   for the whole watershed as desired for aquatic acidification estimates as opposed to only
 8   weathering from the rooting zone as desired for terrestrial acidification (Sverdrup and Warfvinge
 9   1988; Miller, 2001). Depending on how the balance is set up, the balance can be a single
10   equation around the total base cation flux or a series  of equations for each individual cation. The
11   setup of the equations leads to the primary  limitation of the method in that while it is a relatively
12   simple concept, the individual fluxes within the balance are not easily measured or known
13   (Bricker et al., 2003).
14          Most mass balance calculations require an assumption of steady-state behavior. This
15   assumption is easily justifiable over long periods of record. Additional limitations of the method
16   evolve from the number of unknown fluxes (e.g. weathering rate of individual minerals) within
17   the equations defining the balance. Researchers have utilized a variety of techniques to overcome
18   this limitation, including  applying simplifying assumptions or adding additional equations.
19   However, with each assumption or additional equation, a greater amount of uncertainty that must
20   be quantified is added into the analysis. Data sources for a mass balance can also be variable
21   depending on the complexity  of the relationships defined within the balance. While databases
22   and studies may exist for major elements at a variety of sites, comparable data for trace or more
23   complex elements  may be lacking (Velbel and Price, 2007).
24          Historical Rate Determinations - This approach is detailed in Section 4 for terrestrial
25   acidification approaches. Because the BCW flux required for aquatic  acidification approaches
26   requires characterization  of the whole soil profile averaged across a catchment or watershed, this
27   approach can become computationally intensive for aquatic purposes. While it is possible to
28   conduct such an approach on  a small scale  for an aquatic acidification assessment, it is more
29   likely suited to terrestrial applications and so explanation is provided in those sections of the
30   document.
     March 2010                                  18                   Draft - Do Not Quote or Cite

-------
                                                                                         Appendix B


 1           Empirical Relations - A number of empirical relationships have been developed to

 2    calculate BCW, or related factors, from water quality data alone. Empirical relationships are only

 3    as strong as the data on which they are based and are only applicable to the geographic regions

 4    from which the calibration data is obtained.

 5           F-Factor: The F-factor is defined as the ratio of change in non-marine base cation concentrations
 6           due to changes in strong acid anion concentrations (Henriksen,  1984; Brakke et al., 1990). (See
 7           Section 3.1.) A situation where F = 1 indicates that only soil acidification occurs within the
 8           catchment, i.e. all incoming protons are neutralized in the catchment. When F = 0, then only
 9           water acidification is occurring and none of the incoming protons are neutralized in the
10           catchment. Using historical data from Norway, Sweden, U.S.A. and Canada, the F-factor was
11           estimated empirically to be in the range 0.2-0.4 (Henriksen, 1984). Several empirical
12           relationships have been developed in order to calculate the F-factor based on current base cation
13           concentrations using data from Norway (Brakke et al., 1990) or on pre-acidification base cation
14           concentration using data from Finland (Posch et al.,  1993).

15           There are several limitations to using the F-factor. While it is simple to apply anywhere the data
16           is available to satisfy the empirical  relations, these relations are really only valid in Norway,
17           Finland, or wherever the specific relation was derived. In several instances, researchers have
18           applied the Norway- or Finland-based relations to Canadian (Watmough et al., 2005) and U.S.
19           study locations (Henriksen et al., 2002; Dupont et al., 2005;  U.S. EPA, 2009) with the assumption
20           that the empirical equations provide adequate characterization of the relationship between base
21           cation concentrations and the F-factor.
22           A second major limitation in utilizing the F-factor is that this derived factor does not specifically
23           quantify the BCW flux.  Instead it provides calculation of the base cations leached from the soil,
24           which includes BCw and base cations derived through deposition inputs to the system, or
25           removed by harvesting (Henriksen  et al., 2002; Rapp and Bishop, 2009). Although the SSWC is
26           most often used with the F-factor, in a national application where we seek to specifically quantify
27           the BCW, an alternative method should be used.
28           Indicator element in conjunction with weathering ratios: Chen et al., 2004: "Weathering rates at
29           Arbutus watershed could also be obtained using sodium as indicator element, as  described by
30           Gbondo-Tugbawa et al. (2001). The weathering inputs of the indicator element (sodium) could be
31           derived using a mass balance approach, and the derived sodium weathering rate was used in
32           conjunction with base cation weathering ratios reported by Johnson and Lindberg (1992) for the
33           HF to derive weathering rates of other base cations. Using this method, the weathering rates of
34           sodium and calcium derived for Arbutus watershed are very similar to values derived through
35           calibration, whereas rates of magnesium and potassium derived using these two methods showed
36           some discrepancies (Table II)."
37           Weathering rates vs. Stream chemistry or landscape variables: This approach begins with a set of
38           BCW for a  specific set of water bodies. The values of BCW are then regressed against the stream
39           chemistry parameters, such as ANC, in order to find a correlation relationship. These regression
40           relationships are then applied to stream chemistry of other water bodies within a defined region of
41           interest to find the BCW for the water bodies. In areas where  stream chemistry is not available,
42           landscape  variables can be used in place to find correlations with the BCW.
43           The limitation with this approach is that a statistically significant number of BCW values must be
44           available from which to create a regression relationship. Also, the region in which the
45           extrapolation is valid must be defined. In work by Sullivan and colleagues (2004), extrapolation


      March 2010                                    19                    Draft - Do Not Quote or Cite

-------
                                                                                        Appendix B


 1           of modeled BCW values was completed using groupings of physiographic region and ANC class.
 2           The value of this approach is that it can specifically be applied to BCW and can be based on
 3           modeled, monitored, or estimated BCW values as long as there are a sufficient number of values
 4           for extrapolation. Other works have extrapolated ANC values based on chemistry and landscape
 5           variables in a similar manner with a high level of success (Sullivan et al., 2007b, Nanus et al.,
 6           2009).

 7           Process-based Models - Mineral weathering terms within modeling simulations can be a

 8    large source of uncertainty as the weathering term utilized in most process models, in attempts to

 9    represent reality, impacts the loss of base cations to surface waters. Therefore, when little is

10    known about the true weathering rate or the constraints on its values, models must utilize

11    calibration procedures against in-stream water chemistry data to arrive at a  likely weathering rate

12    (Chen et al., 2004; Sullivan et al., 2004).

13           Process-based models vary greatly  in their range of processes represented, complexity of

14    representations, time step, and required data inputs. Overall, there is no perfect model but the

15    best candidate for a task can be chosen provided the available data, the area of concern, and the

16    goals of the analysis. In this case, we  would seek to use all available data resources in order to

17    derive a range of spatially-explicit BCW values across the nation.

18           Descriptions of the four candidate process models available for use  across the country for

19    determining BCW are provided below. In order to provide as concise a description as possible,

20    these model summaries are taken directly from the scientific literature. Summation of the

21    strengths and weaknesses of each model is provided after the model description.

22           DavCent-Chem: "DayCent-Chem links together two widely accepted and  tested models—(1) a
23           daily time-step nutrient cycling and soil hydrology model, version 5 of the DayCent model
24           [Parton et al., 1998], and (2) PHPvEEQC, an aqueous geochemical equilibrium model [Parkhurst
25           and Appelo, 1999]—to form a model that simulates N, P, S, and carbon (C) ecosystem dynamics
26           and soil and stream water acid-base chemistry (fig. 1.2). DayCent-Chem computes atmospheric
27           deposition, soil water fluxes, snowpack and stream dynamics, plant production and uptake, soil
28           organic matter decomposition, mineralization, nitrification, and denitrification (left side of fig.
29           1.2) while utilizing PHREEQC's low-temperature aqueous geochemical equilibrium calculations,
30           including CO2 dissolution, mineral denudation, and cation exchange, to compute soil water and
31           stream chemistry (right side of figure). DayCent-Chem's daily soil solution and stream water
32           chemistry calculations make it possible to use the model to investigate the potential for episodic
33           acidification" (Hartman et al., 2009).

34           DayCent-Chem was recently applied to eight different mountain watersheds from the west to the
35           east with success in certain capacities, therefore, making it a suitable candidate for a national
36           analysis. These applications did highlight difficulties in determining realistic weathering rates in
37           certain areas. With DayCent-Chem, a user must specify an initial value for weathering, which
38           may be adjusted during calibration. In several instances, this value was first set to measured or
39           estimated values for the area of interest and then modified largely during calibration (Hartman et
40           al., 2009). While the daily time step and biotic processes represented by the model provide a

      March 2010                                   20                    Draft - Do Not Quote or Cite

-------
                                                                                           Appendix B


 1           more complex view of the environment, they also add a complexity to the model that appears to
 2           greatly impact the estimation of the key parameter for this analysis.
 3           MAGIC: "MAGIC is a lumped-parameter model of intermediate complexity, developed to
 4           predict the long-term effects of acidic deposition on surface water chemistry [Cosby et al., 1985a,
 5           1985b]. The model simulates soil solution chemistry and surface water chemistry to predict the
 6           monthly and annual average concentrations of the major ions in these waters. MAGIC consists of
 7           (1) a section in which the concentrations of major ions are assumed to be governed by
 8           simultaneous reactions involving SO42" adsorption, cation exchange, dissolution-precipitation-
 9           speciation of Al and dissolution-speciation of inorganic C and (2) a mass balance section in
10           which the flux of major ions to and from the soil is assumed to be controlled by atmospheric
11           inputs, chemical weathering, net uptake and loss in biomass  and losses to runoff. At the heart of
12           MAGIC is the size of the pool of exchangeable base cations  in the soil. As the fluxes to and from
13           this pool change over time owing to changes in atmospheric deposition, the chemical equilibria
14           between soil and soil solution shift to give changes in surface water chemistry. The degree and
15           rate of change of surface water acidity thus  depend both on flux factors and the inherent
16           characteristics of the affected soils" (Sullivan  et al., 2004).
17           The  strengths of MAGIC lie in its simplicity and ability to be applied for a large number of
18           lakes/streams in batch processes. MAGIC has been in use since the 1980s, has been widely
19           applied within the eastern portions of the country with more  limited applications in the West. (See
20           Section 3.3 for further discussion.) The simplicity of MAGIC's mass balances approach also
21           counts as one of its limitations because it may not account for all of the biotic processes that
22           affect the weathering rate. MAGIC determines the BCW through calibration to  water chemistry
23           data. The "fuzzy optimization" procedures now built into MAGIC allow for an optimized value
24           of the BCW to be determined from a series of calibrations at each modeling location (Sullivan et
25           al., 2004).

26           PnET-BGC: "PnET-BGC is an integrated forest-soil-water model that has been used to assess the
27           effects of air pollution and land disturbances on  forest and aquatic ecosystems [Gbondo-Tugbawa
28           et al., 2001]. The model was developed by linking two submodels: PnET-CN (PnET-carbon and
29           nitrogen) [Aber et al., 1997] and BGC [Gbondo-Tugbawa et al., 2001]. The main processes in the
30           model include tree photosynthesis, growth and productivity, litter production and decay,
31           mineralization of organic matter, immobilization of nitrogen, nitrification [Aber et al., 1997],
32           vegetation and organic matter interactions of major elements, abiotic soil processes, solution
33           speciation, and surface water processes [Gbondo-Tugbawa et al., 2001].... For lake  simulations,
34           it is assumed that the water  column is completely mixed. The model predicts monthly
35           concentrations and fluxes of major solutes in lake water, monthly concentrations and pools of
36           exchangeable cations and adsorbed sulfate in soil, and monthly fluxes of major solutes from soil
37           and forest vegetation" (Zhai et al., 2008).
38           Chen and others (2004) nicely summarize the  tradeoffs associated with utilizing the PnET-BGC
39           model: "A strength of PnET-BGC over other acidification models is its ability to simulate
40           [vegetation and microbial processes]. However,  this representation can also be a limitation. The
41           model depicts large element pools in soil and large fluxes through biotic processes. Any change
42           in these pools and fluxes will greatly influence the element budgets. If these simulated fluxes are
43           not accurate, then model predictions will misrepresent element dynamics." Additionally, almost
44           all of the PnET-BGC applications to date have been completed within the eastern portions of the
45           country mostly focusing in the Adirondacks and Hubbard Brook Experimental Forest (Gbondo-
46           Tugbawa et al., 2001; Chen et al., 2004; Zhai et  al., 2008). Expanding this model to western or
47           southern areas would necessitate large amounts of data gathering and processing as well as testing
48           of the representation of the biotic processes found in these differing ecosystems.
      March 2010                                   21                     Draft - Do Not Quote or Cite

-------
                                                                                           Appendix B


 1           PROFILE: "PROFILE [WarfVinge and Sverdrup, 1992] is a steady-state soil chemical model
 2           with a weathering rate sub-model that calculates weathering rates (for each base cation) explicitly
 3           using independent soil properties. Mineral dissolution reactions governing the rate of weathering
 4           involve many components in the liquid phase including H2O, FT, OH", CO2 and organic acids.
 5           These serve as the principle method for cataloguing the contribution of chemical reactions
 6           between soil solution and silicate minerals to base cation release. Inhibition of the reactions
 7           through increased concentrations of the products is accounted for by rate reduction factors.
 8           Precipitation of secondary minerals is subtracted from the total base cation release rate. Climate
 9           data, soil properties and detailed soil mineralogy are used as inputs to the model [WarfVinge and
10           Sverdrup, 1992]" (Whitfield et al., 2006).
11           The PROFILE model is more fully explained in Section 4 for the terrestrial BCW approaches.
12           While the PROFILE model provides a highly deterministic, process-based representation of
13           mineral weathering, trying to utilize this model to determine the base cations weathering and
14           reaching surface water bodies requires the representation of all soil horizons that may contribute
15           to weathering and the summarization of BCW calculations by catchments surrounding each water
16           body of interest. These two qualifications on top of the basic PROFILE application introduce a
17           large amount of complexity into the modeling analysis.

18
19
      March 2010                                   22                    Draft - Do Not Quote or Cite

-------
                                                                                                                      Appendix B
Table 3-1. Review of Modeling Approaches (and models) to Estimate Base Cation Weathering for Aquatic Critical Acid Load Determinations



Model
Approach
Budget Studies






Historical Rate
Determinations





Empirical Data
Relations





F-Factor









Description of
Method
mass balance of
inputs and outputs of
base cations within
catchment or
watershed


loss of base cations
in soil profile relative
to stable element
(Zr, Ti, quartz or
rutile)


modeled
relationships
between surface
water characteristics
and site conditions
or atmospheric
deposition measures
a factor that
combines the effects
of deposition and
weathering






Data
Require-
ments
medium






low






low- high






low-
medium









Model
Complexity
Medium






Low






low- high






Low






Suitability for
Estimating BCW for
Aquatic Critical Acid
Load Determinations in
The United States
low; BCW estimate is
often an integrated value
for whole catchment or
watershed



medium; restricted to
sites with young soils of
known age (eg., soils
that have formed since
the most recent glacial
event, -20,000 years
ago)
low-medium






low; most accurately
applied to sites similar to
those where the model
was derived; if new
derivations can be
completed for the U.S.
the suitability of this
method would increase

Suitability tor Mapping
BCW Over Large
Regions in The United
States
low - medium (based on
data availability); may
require Sr isotope ratio of
stream chemistry to
separate exchangeable
versus weathered base
cation sources
low; restricted to sites
with young soils and
sites where historical rate
determinations have
been conducted


low-medium






low; most accurately
applied to sites similar to
those where the model
was derived; if new
derivations can be
completed for the U.S.
the suitability of this
method would increase




References
Brickeret al., 1993;
Velbel and Price, 2007





Sverdrup et al., 1998;
Sverdrup et al., 1990












Brakke et al., 1990;
Henriksen and Posch,
2001; Henriksen et al.,
2002; Rapp and
Bishop, 2009



March 2010
23
Draft - Do Not Quote or Cite

-------
                                                                                                                 Appendix B



Model
Approach
Indicator
element in
conjunction with
weathering
ratios





Weathering
rates vs. Stream
chemistry or
landscape
variables



Process-Based
Models








Description of
Method
determine
weathering rate
through mass
balance methods for
element such as
sodium (Na) then
apply defined ratios
to determine
weathering rates of
additional elements
utilize weathering
rates determined by
other methods and
extrapolate to
additional areas
based on site
characteristics

Steady-state and
dynamic models that
rely on mathematical
relationships
representing soil and
surface water
processes


Data
Require-
ments
low-medium









low-
medium






medium -
high








Model
Complexity
low









low






medium - high





Suitability for
Estimating BCW for
Aquatic Critical Acid
Load Determinations in
The United States
low; most accurately
applied to sites similar to
those where the model
was derived






medium; suitability will
depend on the ease at
which derived weathering
rates can be obtained
and how strong the
regressions between
BCW and site
characteristics are







Suitability tor Mapping
BCW Over Large
Regions in The United
States
low; most accurately
applied to sites similar to
those where the model
was derived






medium; relatively good
success has been had at
extrapolating BCwto
additional sites based on
stream chemistry;
suitability will depend on
data availability











References
Gbondo-Tugbawa et
al., 2001,2002; Chen
etal.,2004







Sullivan et al., 2004,
2007a, 2007b; Webb
et al., 1994; Nanus et
al., 2009










March 2010
24
Draft - Do Not Quote or Cite

-------
                                                                                                                 Appendix B



Model
Approach
DayCent-Chem









MAGIC









PnET-BGC










Description of
Method
Mineral weathering
rates are set and
then calibrated
within the process
model; Rates are
specified by mineral
and not necessarily
base cations alone


BCW determined
through calibration
to fulfill the
requirements of a
catchment mass
balance by
optimizing simulated
soil and surface
water chemistry to
monitored values
BCW determined
through calibration
and held constant
throughout dynamic
modeling
simulations






Data
Require-
ments
high









medium -
high








high










Model
Complexity
high









medium- high









medium - high







Suitability for
Estimating BCW for
Aquatic Critical Acid
Load Determinations in
The United States
medium - high; provides
daily time step results
which can be used to
estimate time to recovery
or time to damage;
uncertainty on how well
model can simulate BCW
in some areas will impact
confidence of results in
these areas
medium - high;
numerous applications in
the east with some, but
fewer in number,
applications in the west;
little coverage in the
Midwest but these areas
are less of a concern for
aquatic acidification
effects
low-medium; model
applications mostly
completed only within
northeastern U.S.
vegetation and other
biotic processes
represented by the
model would need
validation to other
regions of the country

Suitability tor Mapping
BCW Over Large
Regions in The United
States
medium; DayCent-Chem
has had trouble in
estimating mineral
weathering rates in some
areas of the country





medium- high; will be
restricted in areas where
soils data are lacking
(some western areas);
otherwise, highly
applicable in any areas
where MAGIC
applications have been
completed

low-medium; because
BCwis found through
calibration alone for this
model, the other model
processes and input data
must be validated for any
application before
calibration can be used
for BCW





References
Hartman et al., 2007;
Hartman et al., 2009








Cosby et al., 1985a,
1985b, 1989a; Sullivan
et al., 2004; Sullivan et
al., 2008






Gbondo-Tugbawa et
al., 2001; Chen et al.,
2004; Zhaietal., 2008







March 2010
25
Draft - Do Not Quote or Cite

-------
                                                                                                                 Appendix B



Model
Approach
PROFILE











Description of
Method
BCW determined as
a function of
weathering of
individual soil
minerals and field-
based soil and biotic
conditions




Data
Require-
ments
high











Model
Complexity
high








Suitability for
Estimating BCW for
Aquatic Critical Acid
Load Determinations in
The United States
medium - high; may have
restrictions in desert
regions and areas that
are lacking necessary
data; also must be able
to characterize
catchment summary
values and not solely
individual profiles

Suitability tor Mapping
BCW Over Large
Regions in The United
States
medium - high; may have
restrictions in desert
regions and areas that
are lacking necessary
data








References
Warfvinge and
Sverdrup, 1992 and
1995; Sverdrup, 1990






March 2010
26
Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B

 1   3.3    Proposed Methodology for Estimating and Mapping Base Cation Weathering for
 2          Aquatic Critical Acid Load Calculations
 3          In determining the proposed methodology for a national assessment, the identified
 4   strengths and weaknesses of each approach in the previous section had to be weighed against one
 5   another. Because every method required a large environmental data component, the largest
 6   deciding factor in the proposed approach became the number and spatial representation of
 7   previous applications of an approach within the United States. This decision factor immediately
 8   ruled out applying any of the empirical relationships (e.g. F-factor, relation of BCW to stream
 9   chemistry) derived primarily with data from other countries, although it did not rule out deriving
10   new relationships using the same methods. Ultimately, the F-factor approach was not chosen
11   because it did not directly provide a BCW rate. Additionally, application of an empirical relation
12   alone provided little information on the long-term versus current state of the ecosystem.
13   Therefore, a combination of a process-based model determination of BCW rates with regional
14   expansion of these rates through empirical relations is proposed at the methodology for a
15   national assessment.
16          Utilizing a process-based model, which can calibrate BCW rates to stream or lake
17   chemistry across any number of years, provides a credible long-term estimate of the BCW rate
18   that can be input into the SSWC in order to obtain the system critical load. The process-based
19   model most widely applied throughout the U.S. to date is the MAGIC model. The intermediate
20   complexity of this model provides a balance between data inputs required to run the model and
21   the processes involving base cations, nitrogen, and  sulfur within a watershed, which is
22   considered a requirement of providing a national assessment. Finally, because of its wide
23   application, MAGIC has been extensively tested against independent databases providing the
24   opportunity for iterative model testing and refinement (Sullivan, 2000).
25          The following steps outline the main processes of the method:
26          Step 1.  Definition of MAGIC study sites and the regions to which each grouping of
27                  study sites may be extrapolated.
28          Step 2.  Data gathering and processing for population of the MAGIC model  for
29                  each study site with additional regional data gathering of available
30                  stream/lake chemistry and landscape  parameters.
31          Step 3.  MAGIC modeling application on selected stream/lake study sites where
32                  BCW is arrived at through calibration against water chemistry data.
     March 2010                                27                  Draft - Do Not Quote or Cite

-------
                                                                                          Appendix B
1
2
3
Step 4.  Extrapolation of BCW for modeled streams/lakes to other waterbodies
         within the region through correlation analysis using stream chemistry data,
         where available, and landscape parameters in its absence.

Figure 3-1 provides a flow chart of these steps and their components.
                              Identification MAGIC study sites and applicable regions for
                                                extrapolation
                             Identification of input data for the MAGIC model and of stream
                                      chemistry and landscape parameters
                                       Input  Data Classes
                                                                       Atmospheric
                                                                        Deposition
                                                                     (including annual
                                                                       precipitation)
                                            Construct Model Input
                                                  Tables
                                      MAGIC Application at Study Sites with
                                     Calibration to Water Chemistry for BCw
                                   Correlation Analysis between Site BCw Values
                                  and Stream Chemistry or Landscape Parameters
                                 Application of Regional Regression Relationships
                                 Developed by Site Grouping to Un-Modeled Sites
5
6
7
CMapping of BCw Values Across \
                                Regions of the Nation     J

Figure 3-1. Process steps for estimating BCW using the MAGIC model with regional
extrapolation
     March 2010
                                       28
Draft - Do Not Quote or Cite

-------
                                                                                     Appendix B
 1          Step 1.  Definition of MAGIC study sites and the regions to which each grouping of
 2                   study sites may be extrapolated.
 3          MAGIC has been used to assess acidification impacts in a large number of areas across
 4   North America (Table 3-2). These previous applications should be utilized where possible to
 5   provide a starting point for the national analysis. Sites within the eastern United States likely
 6   provide a wide range of coverage from which initial extrapolations can begin. Within the mid-
 7   west and western areas of the country, additional sites will need to be investigated. Authors of
 8   these studies should be contacted to obtain data sources and model results. Previous model
 9   applications should be compared for the years and objectives of the analysis and input data to
10   determine if the results already created could be utilized in an extrapolation analysis without
11   rerunning the model.
     Table 3-2. Locations of Previous MAGIC Applications within the U.S. and Canada1
Location(s)
25 lakes in south-central Ontario, Canada
2 catchments located in Nova Scotia, Canada

Maryland
36 lake catchments in the Adirondack
Mountains of New York
40 to 50 sites within each of three
physiographic provinces in the eight-state
southern Appalachian Mountains region
33 representative watersheds in the
Adirondacks
Shenandoah National Park
60 Southern Appalachian streams
Joyce Kilmer And Shining Rock Wilderness
Areas (North Carolina/Tennessee)
Monongahela National Forest, West Virginia
Shasta Lake, Idaho
Libby Lake, Montana
Popo Agie Wilderness, WY, and Weminuche
Wilderness, CO
Rocky Mountain, Grand Teton, Sequoia, and
Mount Rainier National Parks
The Loch, a subalpine lake in Rocky Mountain
National Park in Colorado
2 locations in the Sierra Nevadas
Reference
Aherne, J, P.J. Dillon, and B.J. Cosby. 2003.
Dennis, I.F., T.A. Clair, and B.J. Cosby. 2005

Ellis, H., and M. Bowman. 1994.
Church, M.R. and J. Van Sickle. 1999.
Sullivan, T.J., B.J. Cosby, AT. Herlihy, J.R. Webb,
A.J. Bulger, K.U. Snyder, P.P. Brewer, E.H. Gilbert,
and D.L. Moore. 2004.
Sinha, R., M.J. Small, P.P. Ryan, T.J. Sullivan, and
B.J. Cosby. 1998.
Bulger, A. J; Dolloff, C. A.; Cosby, B. J.; Eshleman,
K. N.; Webb, J. R., and Galloway, J. N. 1995
Bulger AJ, Cosby BJ, Webb JR. 2000.
Sullivan, T.J. and B.J. Cosby. 2002
Sullivan, T.J. and B.J. Cosby. 2004
Eilers J.M., B.J. Cosby, J.A. Bernet, T.A. Sullivan,
1998.
Bernett, J.A., Eilers J.M., B.J. Cosby. 1997.
Sullivan, T.J., Cosby, B.J., Bernert, J.A., and Eilers,
J.M. 1998.
Cosby and Sullivan. 2001
Sullivan, T.J., B.J. Cosby, K.A. Tonnessen, and D.W.
Clow. 2005.
Sullivan and Eilers, 1996
      References from this table are presented in Appendix 2.
     March 2010
29
Draft - Do Not Quote or Cite

-------
                                                                                    Appendix B

 1          When selecting sites for MAGIC analyses that will later be used in an extrapolation
 2   analysis, Sullivan and colleagues (2004) outlined two key considerations:
 3          1.   Do not select too many watersheds for modeling that occurred in the same general
 4              area in order to avoid skewing the results too heavily to one portion of the region for
 5              the following extrapolation step
 6          2.   Screen sites to remove those in which the water chemistry data were not internally
 7              consistent or for which available data suggested the possibility of significant
 8              influence from road salt, geological sulfur, land use, or insect defoliation.
 9          Step 2.   Data gathering and processing for population of the MAGIC model for
10                   each study site with additional regional data gathering of available
11                   stream/lake chemistry and landscape parameters.
12          The data requirements of the MAGIC model are summarized in Table 3-3. The table
13   includes both data inputs derived from monitoring data and constant parameters that the user
14   must set based on available data and methods suggested by previous MAGIC applications.
15   Additional information on data inputs can be found in: Cosby and colleagues, 1985a; Cosby et
16   al., 1985b; Sullivan and Cosby,  2004; Sullivan et al., 2007c.
17          Due to the wide range of water quality monitoring assessments conducted within the
18   United States, a large amount of water quality data is typically available to work from.  Similarly,
19   in recent times advances and expansions of atmospheric modeling have been conducted
20   providing a large amount of deposition estimates from which to pull model input data. The area
21   of data most lacking, especially in the western United States, is the composition of soils. Sources
22   of soil data are discussed in Section 4.3.3. Given that there may be areas in which soils data are
23   not available, work by Sullivan  and others used a tiered assessment of MAGIC applications to
24   overcome this obstacle. The tiers consisted of: (1)  chemistry data were available from within the
25   watershed to be modeled with multiple soil sampling sites in an individual watershed aggregated
26   on an area-weighted basis; (2) soils data within the catchment were missing but were available
27   from a nearby watershed underlain by  similar geology; and (3) soils data were neither available
28   from within the watershed nor from nearby watersheds on similar geology. In order to populate
29   soil characteristics for tier 2 and 3 watersheds, a surrogate approach was used meaning that these
30   watersheds were paired with a watershed for which all input data were available. In order to be
31   paired, watersheds had to have similar streamwater characteristics (ANC, sulfate, and base cation

     March 2010                                 30                   Draft - Do Not Quote or Cite

-------
                                                                                     Appendix B
1   concentrations), physical characterization (location, elevation), and bedrock geology data
2   (Sullivan et al., 2004).

           Table 3-3. Input Data Requirements of MAGIC Model
Data Class
Catchment
Stream Chemistry
Aqueous Phase - Equilibrium
Constants
Solid Phase - Weathering
and Exchange Constants
Soil Composition
Atmospheric Deposition
Data Element
Area
Relative area of lake/stream
PH
ANC
Ca'+
Mg^+
K+
Na+
SO/"
N03-
cr
Aluminum solubility constant
Slope of pH-pAl relationship
Organic acid
Organic aluminum
Inorganic aluminum speciation
Inorganic carbon speciation and
dissociation of water
Cation exchange selectivity
coefficients
Weathering rates (Ca"+, Mg"+, K+,
NH4, S042", Cr, N03", F)
Thickness
Total cation exchange capacity
Exchangeable bases (Ca^+, Mg^+,
K+, and Na+)
Bulk Density
Porosity
PH
Sulfate adsorption half saturation
Aluminum solubility constant
Slope of pH-pAl relationship
Annual precipitation
Ca^+
Mg'+
K+
Na+
SO/"
NH4
NO3"
cr
Measure

fraction
unitless
eq/L
eq/L
eq/L
eq/L
eq/L
eq/L
eq/L
eq/L
logio
unitless
logio
logio
eq/m^/yr
Depth (m)
eq/kg
mg/kg
kg/mj
fraction
unitless
eq/nr3
logio
unitless
Volume (m/yr)
Total annual
deposition
(eq/ha/yr)
    March 2010
31
Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B

 1          Step 3.   MAGIC modeling application on selected stream/lake study sites where
 2                   BCW is arrived at through calibration against water chemistry data.
 3          As was completed with the REA (U.S. EPA, 2009), batch processing of MAGIC models
 4   at a range of sites can be completed. Calibration of those sites with available data (streamwater
 5   chemistry, soil chemical and physical characteristics, and atmospheric deposition) is completed
 6   by setting values of the "fixed parameters" within the system and comparing the output of the
 7   model run to the observed values of such characteristics as stream ANC. There are eight
 8   parameters optimized through this method including the BCW rate. The eight observations used to
 9   drive the calibration procedure include the current soil exchangeable pool size and current output
10   flux of each of the four base cations. The model is iteratively run adjusting the "fixed
11   parameters" from a specified range of values (representing uncertainty in knowledge of these
12   parameters), so that the outputs match the observed parameters within an acceptable margin of
13   error. The set of "fixed parameters" that are obtained that allow the model to meet this
14   acceptable of margin of error become the range of calibrated parameters from which the median
15   is chosen to represent the parameter value for the watershed. "The use of median values assures
16   that the simulated responses approximate the most likely behavior of each watershed, given the
17   assumptions inherent in the model and the data used to constrain and calibrate the  model"
18   (Sullivan et al., 2004). This "fuzzy optimization" procedure has been developed for use with
19   MAGIC modeling to help quantify the uncertainties within the modeled parameters (Sullivan et
20   al., 2004). Using these calibration procedures of each site MAGIC run will provide not only an
21   estimate of BCW but an expected range of values in which BCW falls, thereby providing bounds
22   and certainty limits for the following extrapolation step.
23          Step 4.   Extrapolation of BCW for modeled streams/lakes to other waterbodies
24                   within the region through correlation analysis using stream chemistry data,
25                   where available, and landscape parameters in its absence.
26          Regionalization of MAGIC modeling results can be completed through either "binning"
27   sites based on characteristics like physiographic region and ANC concentration (Sullivan et al.,
28   2004) or creating regional regressions to relate site characteristics (chemistry or landscape) to a
29   parameter of interest (e.g., ANC; Sullivan et al., 2007a). In order to provide some measure  of
30   "goodness of fit" to the extrapolations, we have chosen to proceed with creating regression
31   relationships between the BCW determined through calibration of the MAGIC model and either

     March 2010                                 32                  Draft - Do Not Quote  or Cite

-------
                                                                                     Appendix B

 1   water chemistry or landscape parameters. In previous studies, the landscape variables considered
 2   for regression relationships with ANC have included elevation, watershed area, ecoregion,
 3   lithology, forest type and geological sensitivity class (Sullivan et al., 2007b). We expect to
 4   follow similar methods to create the relations with BCW (i.e., the response variable). Within each
 5   region of extrapolation landscape variables appropriate to the region will be selected. For
 6   example, the types of forest selected for inclusion may vary between an extrapolation in the
 7   Southern Appalachians as opposed to the Rocky Mountains. In all instances, there must be
 8   adequate representation of the variable within all modeled and non-modeled watersheds or it will
 9   be eliminated from the pool of candidate variables available for regression analysis.
10          Sullivan and others (2007b) relied on the corrected Aikake's Information Criteria (AIC)
11   to evaluate all possible correlation relations. The corrected version of the  evaluation criteria was
12   used because of the relatively  small sample sizes available from which to build the regressions.
13   Additional evaluation criteria can easily be applied for choosing the best-fitting and most
14   meaningful regressions for extrapolation from a set of individual modeled sites to a larger set of
15   regional sites. Potential criteria for evaluating individual variables within  correlation models
16   include partial F tests, t-values, and variance inflation factors. To evaluate the model as a whole
17   statistics such as PRESS, coefficient of determination, adjusted coefficient of determination,
18   Mallow's Cp, and root mean square error of the model can all be utilized  (Helsel and Hirsch,
19   1992). If a commercial statistical package, such as SAS, is chosen to complete this portion of the
20   analysis the predefined routines and groupings of evaluation statistics can be employed with
21   relative ease.
22   3.3.1  Potential limitations of proposed methodology
23          The limitations with the proposed methodology can be divided into five distinct
24   categories:
25          1.  MAGIC is  an intermediate level model that does not take into  account biotic
26              processes which may affect the calculation of BCW rates within a watershed.
27          2.  While MAGIC is the most widely applied acidification model  within the U.S. it still
28              faces the challenge of having limited applications in the Midwest and western states.
29          3.  As an extension of the bias in eastern applications, processing and organization of the
30              data required for input into the MAGIC model in the East far exceeds that of the
31              West. Additionally, there is an indication that soils data are more incomplete or hard

     March 2010                                  33                    Draft - Do Not Quote or Cite

-------
                                                                                     Appendix B

 1              to obtain in the West. Note that the terrestrial acidification national assessment faces
 2              even greater demands in terms of soil composition data needs. As such, there can be a
 3              combined effort in obtaining new data that will benefit both assessments.
 4          4.  The population and calibration of specific site applications of MAGIC across the
 5              country constitutes a major modeling effort. However, it may be possible to leverage
 6              previous applications.
 7          5.  The proposed approach calls for the creation of several different regional
 8              extrapolations of BCW rates based on sets of individual MAGIC applications. The
 9              success of these extrapolations remains to be seen and will depend upon the
10              limitations mentioned above in even applying the MAGIC model at a multitude of
11              locations and upon the availability of a statistically significant number of model
12              outcomes on which to base the regressions for each regional analysis.
13          While these limitations may seem extensive, there are many possibilities for overcoming
14   the limitations. For example, criteria on model application years can be relaxed to include more
15   of the previously completed MAGIC applications in lieu of updating and rerunning models at the
16   same sites. And, joint data collection between the aquatic and terrestrial acidification
17   assessments can allow the most efficient use of resources and demands on other agencies.
18   3.3.2  Uncertainty analyses
19          As typical with any process based model, the major uncertainties in MAGIC include
20   input data variability, model calibration uncertainty, and the ability of the mathematical model
21   processes to represent reality. Within this national analysis, there will also be uncertainty
22   associated with regional extrapolation of modeling results from individual watersheds to the
23   region. However, Sullivan and colleagues (2004) state that these "errors and uncertainties are not
24   additive, but rather would be expected to some extent to cancel each other out."
25          Several research projects have undertaken attempts to quantify the relative magnitude of
26   the effects of sources of uncertainty for regional, long-term MAGIC simulations using Monte
27   Carlo methods (Cosby et al.,  1989b, 1989c, 1990; Hornberger et al., 1989, 1990). While the
28   results of these studies indicated that the different sources of uncertainty can have varying levels
29   of impacts on the outputs of the MAGIC model, the development of the "fuzzy optimization"
30   technique for calibration was designed to reduce these impacts of uncertainty. With "fuzzy
31   optimization" there is an explicit accounting within different uncertainty categories and a

     March 2010                                  34                   Draft - Do Not Quote or Cite

-------
                                                                                      Appendix B

 1    resulting time-variable measure of overall simulation uncertainty for each state variable. One
 2    way the optimization procedure reduces uncertainty is in its selection of parameter and variable
 3    values for the "fixed parameters" from distributions of possible values rather than having a user
 4    select a single value during a single calibration (Sullivan et al., 2004).
 5           Outside of the operation and calibration of the MAGIC model, the parameterization of
 6    the input data provides another source of uncertainty. As identified in previous sections, the soils
 7    composition data are expected to be the greatest source of uncertainty. If a tiered approach to
 8    populating soils data for watersheds lacking in data is used, uncertainty with the method can be
 9    examined by calibrating selected tier  1 watersheds twice, once using the appropriate site-specific
10    soils data, and a second time using borrowed soils data from an alternate site, using either tier 2
11    or tier 3 protocols. A comparison between the results from each of the scenarios can then be
12    made to determine the magnitude of difference in output parameters. If this analysis can be done
13    at multiple sites, than a sensitivity analysis can be performed over the results to determine if
14    there is a consistent bias in results from modeling analyses utilizing tier 2 or 3 procedures
15    (Sullivan et al., 2004).

16    4.      TERRESTRIAL BASE CATION WEATHERING METHODOLOGY
17    4.1     Introduction
18           Geology is one of the most important factors in determining the potential sensitivity of an
19    area to terrestrial acidification (U.S. EPA, 2008, Section 3.2.4). In particular, the characteristics
20    of the soils and the upper portion of the bedrock can impact the acid-neutralizing ability of the
21    soils in a particular area. Acid-sensitive soils are those which contain low levels of exchangeable
22    base cations and low base saturation (U.S. EPA, 2008, Section 3.2.4). Bedrock composition and
23    soil pH are two characteristics that are directly related to the ability of a system to neutralize
24    acid. Soils overlying bedrock, such as calcium carbonate (e.g., limestone), which is reactive with
25    acid, are more likely to successfully neutralize acidifying deposition than soils overlying
26    nonreactive bedrock. In addition, soils with higher pH (i.e., more alkaline) have a greater
27    capacity to neutralize acidifying deposition.
28           This section reviews the effect of acidification known as base cation weathering,
29    describes its significance in estimating critical acid loads, and identifies methodologies for
30    estimating base cation weathering. Further, this report recommends a methodology for potential

      March 2010                                  35                   Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B

 1   use in the review of the NOX and SOX secondary National Ambient Air Quality Standards and
 2   describes the steps and information resources needed to apply that methodology across the
 3   United States.
 4   4.2    Terrestrial Base Cation Weathering
 5          In the calculation of terrestrial critical acid loads using the simple mass balance (8MB)
 6   methodology, base cation weathering (BCW)2 is defined as "the release of base cations from
 7   minerals in the soil matrix due to chemical dissolution" (UNECE, 2004), and this weathering
 8   occurs in the rooting zone  of the soil profile and consists of the release of calcium (Ca2+),
 9   magnesium (Mg2+), potassium (K+) and sodium (Na+). It does not include the removal of base
10   cations from  soil ion exchange complexes (cation exchange sites) or the degradation of soil
11   organic matter. Base cations from these sources have already been released through the
12   weathering process. Base cation weathering is often a dominant source of base cations in soils,
13   replacing Ca2+, Mg2+, K+ and Na+ that are lost through leaching and uptake by plant (Langan et
14   al., 1995; Langan et al.,  1996; Ouimet, 2008). Therefore, BCW plays an important role in
15   determining the sensitivity of a site to acidifying nitrogen and sulfur deposition (Hodson and
16   Langan 1999a). The BCW term is also one of the most influential parameters in the 8MB
17   calculations of terrestrial critical acid loads. Li and McNulty (2007) determined that 49% of the
18   variability in critical load estimates was due to this term. Sverdrup and colleagues (1995)
19   (reference in Langan et al., 1996) determined that BCW can account for 90% of the variation in
20   critical loads.
21          For the Terrestrial Acidification case study  in the Risk and Exposure Assessment (U.S.
22   EPA, 2009), BCW rates were calculated using the clay-substrate method (McNulty et al., 2007).
23   This method was selected for the Risk and Exposure Assessment because it is one of the most
24   commonly used methods to estimate BCW for critical load analyses in North America (Ouimet et
25   al., 2006; Watmough et al., 2006; McNulty et al., 2007; Pardo and Duarte, 2007), and has been
26   used to map critical loads across the United States (McNulty et al., 2007). However, the
27   applicability of the clay-substrate method is most likely limited because it is an empirical model
28   that appears to be based on a modification of the soil type - texture approximation method that
29   was developed on a restricted number sites in northern Europe that were glaciated during the last
     2 Within the 8MB equation, Bcw refers to the weathering of Ca2+, Mg2+ and K+.
     March 2010                                 36                   Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B

 1   glacial advance (CLAD, 2009; H. Sverdrup personal communication, 2009a, UNECE, 2004). It
 2   relies on a classification of the acidity of soil parent material and soil clay content and consists of
 3   three equations (equations 4-1 - 4-34).

 4                     Acid Substrate: BCe = (56.7 x %clay)- (p.32 x (%clay)2)               (4-1)

 5               Intermediate Substrate: BCe = 500 + (53.6 x %clay)- (o. 18 x (%clay)2)        (4-2)

 6                           Basic Substrate: BCe = 500 + (59.2 x %clay)                    (4-3)

 7   where
 8             BCe = empirical soil base cation (Ca2+ + K+ + Mg2+ + Na+) weathering rate
 9                     (eq/ha/yr)
10           % clay = the percentage of clay (determined by particle size) within the rooting zone
11                     of soil profile.

12   Critical load experts from both the United States and Canada have commented that the clay-
13   substrate model, in general, appears to perform well in young soils that have formed since the
14   last glaciations (approximately 20,000 years before present). However, the model may not be
15   suitable or provide accurate estimates on older, more weathered soils that were not impacted by
16   the last glaciation (P. Arppersonal communication, 2009). These soils have undergone
17   weathering for a longer period of time and the relationships between clay particle size and base
18   cation release may not be as strong as in younger soils (H. Sverdrup personal communication,
19   2009a). To our knowledge, however, there have been no published studies that have tested this
20   hypothesis and compared BCW estimates generated with the clay-substrate model and other
21   methods on sites underlain by old, more weathered and recently glaciated soils. At least one
22   study has compared the clay-substrate BCW method with estimates from other models on
23   glaciated soils in Canada and found that the rate estimates were similar within the area of
24   assessment (Whitfield et al., 2006).
25          Results from the Risk and Exposure Assessment (U.S. EPA, 2009) appear to support the
26   distinction between the suitability of applying the clay-substrate model to glaciated versus older,
27   non-glaciated soil environments. As outlined in Appendix 5 of the Risk and Exposure

     March 2010                                 37                   Draft - Do Not Quote or Cite

-------
                                                                                     Appendix B

 1   Assessment, the regression analysis assessing the relationship between the growth of sugar
 2   maple (Acer saccharum) and critical acid load exceedance was not significant (p=0.38) when all
 3   plots were included in the analysis. However, when the analyses were restricted to sites located
 4   on younger, glaciated soils, which resulted in the removal of 25% of the data from the analyses,
 5   the linear regression relationship was significant at the p=0.10 level. Improvements in the
 6   significance of the relationship may, in part, have been due to the greater accuracy of the BCW
 7   estimates in the critical load calculations for the plots north of the glaciations line.
 8          The majority of the conterminous United States was not directly impacted by the most
 9   recent glacial advance (Figure 4-1) and some of the soils in these areas have not been influenced
10   by glaciations in at least 700,000 years (Sverdrup et al., 1992). Only ten states had their full land
11   area impacted by glaciers during the glacial advance 20,000 years ago. Therefore, if the concerns
12   and supportive results regarding the suitability of the clay-substrate model for the estimation of
13   BCW on older, non-recently-glaciated soils are correct, the model may not be an appropriate
14   method to estimate BCW for a large portion of the United States.  Given that the BCW parameter is
15   one of the most influential variables within the 8MB calculations to estimate critical acid loads,
16   it is particularly important to use a method that provides accurate and  defendable estimates of
17   BCW. Therefore, any and all future work focused on  estimating and mapping terrestrial critical
18   acid loads in the United States, should acknowledge the potential limitations of the clay-substrate
19   model and consider the adoption of a BCW modeling approach that is transferable and can be
20   applied to multiple locations and different soil conditions and soil ages.
     March 2010                                  38                   Draft - Do Not Quote or Cite

-------
                                                                                     Appendix B
                                                           I Approximate area affected by most recent glaciation
 1
 2          Figure 4-1. Areas of continental U.S. that were covered during the last glacial event
 3          (Reed and Bush, 2005).
 4   4.3    Methodologies for Determining Base Cation Weathering Values in the United States
 5   4.3.1  Difficulties in estimating base cation weathering
 6          Base cation weathering is one of the most difficult parameters to estimate (Sverdrup et
 7   al., 1990; Ouimet and Duchesne, 2005; Langan et al., 1996), as it is a function of a time, soil
 8   mineralogy, and a variety of other environmental biotic and abiotic factors. Weathering occurs
 9   over centuries and millenia and results in the  chemical and physical alterations of parent material
10   and minerals. Minerals that are present in the soil may no longer resemble the original bedrock
11   parent material (C. Smith personal communication, 2009). In addition, the soil may be derived
12   from parent material that was transported to its current location and does not resemble the
13   underlying bedrock. Abiotic factors including temperature and moisture and location on the
     March 2010
39
Draft - Do Not Quote or Cite

-------
                                                                                    Appendix B

 1   landscape and biotic factors including vegetation and soil microbes can also impact base cation
 2   weathering through removal of base cations and chemical weathering of minerals (Brady and
 3   Weil, 2002). Combined, these factors pose many challenges to determining BCW in the soil
 4   profile for terrestrial critical acid load estimations. As a result, a variety of BCW methods and
 5   approaches have been developed (Sverdrup et al., 1990).
 6   4.3.2  Approaches to estimating BCw:
 7          Methods and models that have been developed to estimate BCW for critical acid load
 8   determinations differ significantly in the approaches used to generate weathering estimates
 9   (Langan et al., 1995; Sverdrup et al., 1990; UNECE, 2004). For the purposes of this work
10   assignment, BCW methods and models are grouped into three main approaches:
11          1. budget studies of catchments or watersheds;
12          2. historical weathering rate determinations; and
13          3. empirical and mathematical models.
14   Each of these approaches vary in complexity, data intensity and scalability, thereby offering
15   different strengths and weaknesses to estimating BCW. In addition, these approaches differ in
16   their abilities to map BCW over regional and larger land areas. Table 4-1 provides a summary of
17   the approaches to BCW, critical load  models that use the approaches, the strengths and
18   weaknesses of the different models,  and the suitability of the approaches and models to map BCW
19   and therefore critical loads over large areas. For a model to be suitable for large-scale mapping, it
20   must be quick to apply, supported  by existing databases, not require extensive and costly
21   analyses, and be transferable to sites with varying conditions and geological histories (Sverdrup
22   etal., 1990).
23          Budget Studies - The budget  study approach, also referred to as input-output balances
24   (Kolka et al., 1996; Langan et al.,  1996; Starr et al., 1998), estimates BCW as a component of the
25   mass balance input and output of cations within a catchment or watershed (Langan et al., 1996;
26   Sverdrup et al., 1990; Sverdrup et  al., 1998). In most catchments, the main source of input is
27   atmospheric deposition and output is streamflow, and base cation retention is accounted for
28   through uptake by biomass and immobilization in the soil. Base cation weathering is therefore
29   determined through mass balance differences between these different input, output and storage
30   pools. The main strengths of this method are that it only requires a moderate amount of input
31   data and relies on data collected from the catchment or watershed. In addition, it offers the

     March 2010                                  40                   Draft - Do Not Quote or Cite

-------
                                                                                    Appendix B

 1   potential for mapping multiple catchments, if the necessary input data is available. However,
 2   several of the drawbacks to this method include an assumption that the catchment is in a steady -
 3   state condition and the cation exchange capacity does not change over time (Langan et al., 1996;
 4   Miller, 2001; Sverdrup et al., 1998). In addition, it is often difficult to determine BCW within the
 5   rooting zone of individual soil profiles because the BCW estimates from budget studies represent
 6   integrated values for the whole watershed, the full soil profile, bedrock weathering and all soil
 7   processes (Sverdrup and Warfvinge 1988; Miller, 2001). It is also difficult to separate
 8   contributions of base cations from exchange sites versus mineral weathering and  chemical
 9   dissolution (Sverdrup et al.,  1990; Miller, 2001). Therefore, it is challenging, and potentially
10   erroneous, to use budget  studies of catchments to estimate BCW for terrestrial critical acid loads.
11   It may be possible to modify the budget study approach and evaluate base cation  input and
12   output in the rooting zone of individual soil profiles (Kolka et al., 1996), and to separate base
13   cations from exchanges soil  pool versus weathering sources using techniques such as  the analysis
14   of strontium (Sr)  isotope ratios (Miller et al., 1993). However, the soil profile approach would
15   require lysimeter measurements of soil solution chemistry at each site and the soil solution would
16   also need to be analyzed for Sr isotope ratios (87Sr/86Sr).  Both of these analyses would be very
17   time intensive and would not be practical over large areas.
18          Historical Rate Determinations - The historical weathering rate approach, also
19   sometimes referred to as  element depletion (Langan et al., 1996; Miller, 2001)  or pedological
20   mass balance (PMB) (Ouimet and Duchesne, 2005; Ouimet, 2008), estimates BCW by
21   determining the relative depletion of base cations to the depletion of a stable element  as a
22   function of the age of the soil profile (Langan et al., 1996). Zirconium (Zr), titanium (Ti), rutile
23   and sometimes quartz are typically selected as the stable soil elements for this method (Langan et
24   al., 1996;  Sverdrup et al., 1998) because they are very resistant to weathering (Starr et al.,  1998).
25   This technique is commonly applied to soils that were formed since the last glaciation, and
26   characterizes the  ratio of base cations to the stable element in the upper weathered soil horizons
27   and the unweathered C horizon. It assumes that  post glaciation, the mineral matrix of the soil
28   consisted  of freshly ground material that was not previously exposed to weathering (Sverdrup et
29   al., 1998), the lowermost soil is representative of the parent material and the stable element is
30   found in a constant proportion throughout the soil profile (Langan et al., 1996;  Starr et al., 1998).
31   Over time, base cations are weathered and lost from the profile through uptake and leaching, but

     March 2010                                  41                   Draft - Do  Not Quote or Cite

-------
                                                                                    Appendix B

 1   the concentration of the stable element remains constant due to resistance to weathering (Starr et
 2   al., 1998). Main strengths of this approach are that it is a good technique to estimate weathering
 3   in young soils, does not require a large amount of data and BCW is relatively easy to calculate.
 4   However, this approach also presents some major weaknesses. It estimates the historic BCW rate
 5   which may be differ from the current weathering rate. Historic weathering rates may
 6   underestimate current BCW because the historical weathering occurred under more neutral
 7   conditions with less acidifying deposition (Sverdrup et al., 1990). Conversely, historic
 8   weathering rates may be higher than current BCW if the original post glaciations soil contained a
 9   significant proportion of easily weathered material that have since been depleted (Miller,  2001).
10   Studies have indicated that the initial phase of weathering lasts a few hundred to several
11   thousand years and can deplete  a maximum  of 25% of the mass during this period (Sverdrup et
12   al., 1998). In addition, the historic rate approach is not suitable for older, more weathered soils.
13   In such soils, it is often difficult to determine the amount of time since the last glacial or mass
14   disturbance  event that caused the formation  of newly ground material (H. Sverdrup personal
15   communication,  2009b). Therefore, this method cannot be applied to all locations and it is often
16   difficult or impossible to extrapolate results  to larger geographical areas. Since a large proportion
17   of the soils in the United States were not influenced by the most recent glaciation, the historic
18   rate approach to estimate BCW for terrestrial critical acid load estimates could only be applied to
19   a fraction of the land area.
20          Empirical and Mathematical Models - Empirical  and mathematical models estimate BCW
21   based on laboratory- and field-based relationships between soil, abiotic and biotic factors. Over
22   the past several decades, a large number of BCW models have been developed for terrestrial
23   critical acid load determinations. Initial models were developed from a limited number of sites
24   and data. More recent models incorporate a larger number of factors and are more complex and
25   data intensive. One of the first BCW models was the Skokloster Assignment which is a semi-
26   empirical method that was devised during the Critical Load Workshop in Skokloster, Sweden in
27   1988 (UNECE, 2004). It divides minerals into 5-6 mineral classes based on the dominant
28   weatherable soil minerals and assigns a range of critical acid loads to each. This method was
29   later expanded to include a larger range of minerals and to estimate BCW based on the relative
30   abundance of fast versus slow weathering minerals. The Skokloster Assignment was originally
31   based on soils with density, moisture content, clay content and pH conditions similar to the soils

     March 2010                                  42                   Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B

 1   in the three Gardsjon catchments in Sweden, and was validated against a preliminary version of
 2   the PROFILE model, described further below (Hodson and Langan 1999b; H. Sverdruppersonal
 3   communication, 2009b).
 4          A second model, the Soil Type - Texture Approximation assigns weathering rate classes
 5   to soils based on soil texture and parent material acidity classes. It was developed for European
 6   forest soils (UNECE, 2004). As  described earlier, it is believed that the clay-substrate model that
 7   is used extensively throughout North America was derived from the Soil Type - Texture
 8   Approximation.
 9          A third model, the Total Base Cation Content Correlation was developed using Zr(SiO4)
10   and historical rate approach applied to eleven sites in Sweden (UNECE, 2004). Correlation
11   between historical BCW rates and the total content of base cations in the undisturbed bottom soil,
12   corrected for temperature, were used to develop equations to estimate the weathering of Ca2+, K+
13   and Mg+2 (Olsson et al., 1993). For a more complete review and description of these three
14   empirical models see Hodson  and Langan (1999) and UNECE (2004). In general, the main
15   benefits of these models are the minimal data requirements, transferability, and the potential to
16   be applied to multiple sites. Therefore, such models offer good options for mapping of BCW for
17   critical acid load determinations. However, these models also have several key weaknesses
18   which limit their utility for estimating BCW in many locations, including a large proportion of the
19   United States. All of the models were determined using data from a limited number of sites
20   within Sweden and other regions of Europe and are based on average or generalized
21   relationships. Therefore, similar to the clay-substrate model, these models may do a reasonable
22   job of estimating BCW on sites that were recently glaciated and/or have similar conditions to the
23   Swedish sites. However, they  should not be applied everywhere, as they may poorly estimate
24   BCW on sites with older, more weathered soils. As stated for Total Base Cation Content
25   Correlation, the method is only applicable to granitic soils (Hodson and Langan, 1999) and
26   should be used with caution because the relationships are based on Nordic geological history
27   (UNECE, 2004).
28          A fourth model that supported the creation of some of the aforementioned empirical BCW
29   models, and is currently in its  5th version is PROFILE (Warfvinge and Sverdrup 1992 and 1995;
30   Sverdrup, 1990). PROFILE (version 5.0) is a mechanistic, mathematical, steady-state, kinetics
31   model that calculates the weathering of Ca2+, Mg2+, and K+ in each horizon of a soil profile

     March 2010                                 43                   Draft - Do Not Quote or Cite

-------
                                                                                     Appendix B

 1   (Akselsson et al., 2005). It is unique and differs from other empirical models in that it calculates
 2   BCW rates for soil from independently measured geochemistry and soil conditions (Jonsson et al.,
 3   1995; Ouimet and Duschesne 2005). It combines laboratory-based evaluations of mineral-
 4   specific chemical dissolution with field-based conditions and other soil measurements to
 5   estimate individual weathering rates of Ca2+, Mg2+, and K+ (Langan et al., 1996). The model
 6   includes 14 of the most common primary and secondary soil minerals (Table 4-2)3, and their
 7   release of base cations in five separate reactions (Sverdrup et al., 1990; Hodson et al., 1997;
 8   Sverdrup etal., 1998):
 9          i)     Reaction with hydrogen ion (H+) and dissolved aluminum (Al)
10          ii)     Reaction with water and dissolved Al
11          iii)    Reaction with hydroxyl ion (OH") and dissolved Al
12          iv)    Reaction with carbon dioxide (CO?)
13          v)     Reaction with strongly complexing polydentate organic acids
14   The reaction rates are calculated using constants contained within the model and  data input by
15   the user, and the total base cation release rate by chemical weathering is calculated as the sum of
16   all parallel simultaneous process rates minus the rate of precipitation of secondary solid phases
17   (Sverdrup et al., 1998). The  rate equation (Equation 4-4) for the weathering of all minerals
18   within the rooting zone of the soil profile is defined as:

19                              RW = 2(horizonS)2(minerals)/ T> ' AexP ' X' '® ' Z
20   where
21                 rt = dissolution rate of mineral /' (kmolc/m2/s) - sum of the 5 separate reactions
22                 Aexp = exposed surface of mineral matrix (m2/m3)
23                 9 = soil moisture saturation (m3/m3)
     3 Thirteen additional minerals can be added to PROFILE, as necessary (H. Sverdrup personal
     communication).

     March 2010                                  44                   Draft - Do Not Quote or Cite

-------
                                                                                    Appendix B

 1                 Xj = fraction of mineral /' in the mineral matrix of the soil horizon
 2                 z = soil layer thickness (m)
 3   The weathering rate is either increased or reduced by different soil and biotic and abiotic
 4   conditions, many of which are entered as input data by the user. Input data includes site climatic
 5   and deposition attributes, soil physical and chemical characteristics and biological components
 6   that influence the soil chemistry and BCW (input data required by PROFILE discussed further in
 7   Section 4.3.3). As summarized by Jonsson and colleagues (1995), "The weathering rate is
 8   increased by a high H+ concentration, a high soil moisture (water) content, and a high CC>2
 9   pressure. Weathering reactions are product inhibited, i.e., decrease by high concentrations of
10   reaction products in the soil solution such as inorganic aluminum and base cations. The surface
11   activity is calculated as dependent on the mineral surface area, temperature and soil moisture
12   saturation. The soil temperature impact on the weathering rate is expressed as an Arrhenius
13   equation, as dependent on the activation energy. The soil moisture saturation is important for the
14   reaction rate as the reactions will only take place on wetted surfaces.  The degree of surface
15   wetting, and thus surface activity, is considered to be a function of the soil moisture saturation.
16   This is calculated from soil bulk density, the solid particle density and the volumetric water
17   content." For a more detailed description of the theory and calculations behind PROFILE, see
18   Sverdrup (1990), Sverdrup and Warfvinge (1992, 1993a, 1995).
19          A main weakness of the PROFILE model is that it is data intensive and complex, and can
20   be difficult to parameterize. However, PROFILE does offer some significant benefits that set it
21   apart from the other models. As described, it determines current BCW rates from laboratory-
22   derived weathering rates of individual minerals and therefore is not bound to data from a specific
23   location or region. Therefore, it can be used to determine and map BCW over large areas (Miller
24   et al., 1993). Although it was developed in Sweden, it has been successfully applied to the
25   mapping of BCW and critical loads in the Northeastern United States,  Maryland, Minnesota,
26   Pennsylvania, Thailand, China, Argentina, and Greece (Duan et al., 2002; Miller, 2001; Sverdrup
27   et al., 1992; H. Sverdrup personal communication, 2009b).
     March 2010                                  45                   Draft - Do Not Quote or Cite

-------
                                                                                                                         Appendix B
Table 4-1. Review of modeling approaches (and models) to estimate base cation weathering for terrestrial critical acid load determinations.




Model Approach
Budget Studies










Historical Rate
Determinations





Empirical
Models





Description of
Method
mass balance of inputs
and outputs of base
cations within
catchment, watershed
or soil profile






loss of base cations in
soil profile relative to
stable element (Zr, Ti,
quartz or rutile)




modeled relationships
between soil attributes
and abiotic and biotic
site conditions




Data Requirements
medium










low





low- high







Model Complexity
medium










low





low- high



Suitability for
Estimating BCW for
Terrestrial Critical Acid
Load Determinations in
The United States
low; BCw estimate is often
an integrated value for
whole catchment or
watershed







medium; restricted to sites
with young soils of known
age (e.g., soils that have
formed since the most
recent glacial event,
-20,000 years ago)


low- high



Suitability for
Mapping BCw
Over Large
Regions in The
United States
low- medium
(based on data
availability); may
require Sr
isotope ratio of
stream chemistry
to separate
exchangeable
versus
weathered base
cation sources
low; restricted to
sites with young
soils and sites
where historical
rate
determinations
have been
conducted
low- high







References
Sverdrup et al.,
1998; Sverdrup et
al., 1990








Sverdrup et al.,
1998; Sverdrup et
al., 1990









March 2010
46
Draft - Do Not Quote or Cite

-------
                                                                                                                 Appendix B




Model Approach
Skokloster
Assignment








Soil Type -
Texture
Approximation




Description of
Method
BCW rate categorically
determined by relative
abundance of minerals
grouped into 5-6
weathering rate
classes; originally
developed for soils
similar to those found
in the 3 Gardsjon
catchments in Sweden
BCW categorically
determined as a
function of parent
material acidity and
soil texture, modified
by temperature;
developed from data
from European forest
soils




Data Requirements
low









low





Model Complexity
low









low

Suitability for
Estimating BCW for
Terrestrial Critical Acid
Load Determinations in
The United States
low; most accurately
applied to sites similar to
those where the model
was derived






low- medium; most
accurately applied to sites
similar to those where the
model was derived

Suitability for
Mapping BCw
Over Large
Regions in The
United States
low; most
accurately
applied to sites
similar to those
where the model
was derived




low- medium;
most accurately
applied to sites
similar to those
where the model
was derived





References
UNECE, 2004;
Hodson and
Langan, 1999







UNECE, 2004;
Hodson and
Langan, 1999

March 2010
47
Draft - Do Not Quote or Cite

-------
                                                                                                                 Appendix B




Model Approach
Total Base
Cation Content
Correlation








Clay-Substrate
Model







Description of
Method
BCW determined by
correlations between
historical rate
determinations (Zr)
and total content of
base cations in the
undisturbed bottom
soil, corrected for
temperature; based on
data from eleven sites
in Sweden
BCW determined by
one of three equations
based on parent
material acidity and %
clay content; most
likely a modification of
the Soil Type - Texture
Approximation






Data Requirements
low










low








Model Complexity
low










low




Suitability for
Estimating BCW for
Terrestrial Critical Acid
Load Determinations in
The United States
low; restricted to sites with
granitic soils and Nordic
geological histories








low- medium; most
accurately applied to sites
similar to those where the
model was derived (most
likely young soils formed
since the last glaciation)




Suitability for
Mapping BCw
Over Large
Regions in The
United States
low; restricted to
sites with granitic
soils and Nordic
geological
histories






low- medium;
most accurately
applied to sites
similar to those
where the model
was derived
(most likely
young soils
formed since the
last glaciation)




References
UNECE, 2004;
Hodson and
Langan, 1999








original source
unknown; Ouimet et
al., 2006;
Watmough et al.,
2006; McNulty et
al., 2007; Pardo and
Duarte, 2007



March 2010
48
Draft - Do Not Quote or Cite

-------
                                                                                                                 Appendix B




Model Approach
PROFILE









Description of
Method
BCW determined as a
function of weathering
of individual soil
minerals and field-
based soil and biotic
conditions





Data Requirements
high










Model Complexity
high






Suitability for
Estimating BCW for
Terrestrial Critical Acid
Load Determinations in
The United States
medium - high; may have
restrictions in desert
regions and areas that are
lacking necessary data



Suitability for
Mapping BCw
Over Large
Regions in The
United States
medium - high;
may have
restrictions in
desert regions
and areas that
are lacking
necessary data




References
Warfvinge and
Sverdrup, 1992 and
1995; Sverdrup,
1990



March 2010
49
Draft - Do Not Quote or Cite

-------
                                                                                      Appendix B
      Table 4-2. The fourteen dominant minerals modeled within PROFILE.
                            Dominant Minerals
                                K-Feldspar
                               Plagioclase
                                  Albite
                               Hornblende
                                Pyroxene
                                 Epidote
                                 Garnet
                                  Biotite
                                Muscovite
                                Fe-Chlorite
                              Mg-Vermiculite
                                 Apatite
                                 Kaolinite
                                 Calcite
 1
 2    4.3.3  Proposed methodology for estimating and mapping base cation weathering for
 3          terrestrial critical acid load calculations
 4          As has been outlined in the above review, there are multiple approaches to estimate BCW
 5    for terrestrial critical acid loads. However, not all are suitable for both calculating and mapping
 6    terrestrial critical acid loads throughout the United States. Such an approach has to be quick and
 7    easy to apply, be supported by available data and be easily and accurately transferable to sites
 8    within the United States that differ in soil, biotic and abiotic properties and conditions. In
 9    addition, as stated by Miller (2001), "the most promising approach for a logically consistent
10    estimation of the present-day weathering rate over broad regions is the application of model(s)
11    that predict the weathering rate from first principles, given detailed measurements of the soil
12    environment and laboratory-derived rate constants for specific mineral weathering reactions."
13    Therefore, an approach that is based on soil mineralogy and weathering of individual minerals is
14    preferable. Of all the models that are currently available for  determining BCW for terrestrial
15    critical acid load determinations, PROFILE meets these requirements and appears to be the most
16    suitable. Methodologically, it  has few location restrictions and models BCW based on site-
17    specific mineralogy and soil and site conditions. In addition, it has already been successfully
18    applied in both glaciated and non-glaciated regions of the United States to estimate and map BCW
     March 2010
50
Draft - Do Not Quote or Cite

-------
                                                                                 Appendix B

 1   and critical acid loads (Miller et al., 1993; Sverdrup et al., 1992; H. Sverdruppersonal
 2   communication, 2009b/ Although, as with all models, PROFILE does have some weaknesses
 3   and limitations (discussed further in Section 4.3.5) that need to be acknowledged and addressed
 4   prior to application, critical load experts, in general, agree that PROFILE is the best model to
 5   date for estimating and mapping BCW rates for terrestrial critical acid load determinations in the
 6   United States (J. Aherne personal communication, 2009, J. Cosby personal communication,
 7   2009, J. Lynch personal communication, 2009, R. Ouimet personal communication,  2009, H.
 8   Sverdrup personal communication, 2009b).
 9          There are two forms of PROFILE (version 5.0) that can be used for estimating BCW: the
10   single site application which estimates BCW for a single location or  soil profile, and the regional
11   application which PROFILE can be run for a region or conterminous areas (C. Akselsson
12   personal communication, 2009). For mapping BCW in the conterminous United States, the
13   regional application of the model would be applied, and the estimation and mapping of BCW
14   would involve two main steps:
15          Step 1.  Identification of input data required by PROFILE and development of
16                  spatial data layers, national databases and default values for each data
17                  element within the model
18          Step 2.  Determination of polygon layer to spatially define the BCW rates and
19                  development of continuous coverage map of calculated BCW values.
20          These process steps are further illustrated in the flowchart presented in Figure 4-2.
     March 2010                                51                  Draft - Do Not Quote or Cite

-------
                                                                           Appendix B
                Identification of input data required by PROFILE to estimate BCV


In
Existing
National-Level
GIS Coverages
(see table 3a)

P

> u t Dat
a Class
Requires Development
and Delineation
of National-Level GIS
Datalayers
(see table 3b)

1



es
PROFILE Default
Values
Requires review by user
(see table 3c)



F
                                  Construct National GIS
                                Data Layers and convert to
                                      raster format
                          Determine and Delineate BCw polygons
                             Calculate mean values for National
                                     GIS Data Layers
                                        for each
                            	BCw Polygon	
                                   Organize and Format
                                    BCw Polygon data
                                           as
                                     PROFILE model
                                         input file
i
2
                         Calculate BCw
                              For
                         BCw polygons

Figure 4-2. Process Steps for Estimating BCW Using the PROFILE Regional Model
    March 2010
                              52
Draft - Do Not Quote or Cite

-------
                                                                                      Appendix B
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
       Step 1.   Identification of input data required by PROFILE and development of
                spatial data layers, national databases and default values for each data
                element within the model
       PROFILE (version 5.0) is data intensive and requires the user to provide or review a total
of 26 soil, climatic and biological input data (Table 4-3a-c). In addition, to run the regional
application of PROFILE, it would be necessary to have each of these data parameters available
as continuous coverages. A large proportion of these variables are already included in existing
databases in the United States and could be easily converted into continuous coverage data layers
for the conterminous United States, if not currently available in continuous coverage format.
Others, such as soil mineralogy, would need be modeled or constructed from other data. Still
others may need to be represented by default values from the literature, until more unique,
spatially site-specific values are determined.
Table 4-3a. Data required to estimate BCW with the regional PROFILE model (version  5.0). The data in
this table must be input by the user and are currently available as a continuous coverage layers for at
least a portion of the conterminous United States.
PARAMETER
precipitation
cation deposition
anion deposition
number of soil layers
soil layer height
temperature
dry soil bulk density
run-off
UNITS
m/yr
kEq/ha/yr
kEq/ha/yr
#
m
°C
kg/mj
m/yr
DESCRIPTION
30-year long-term average
ammonium (NH4+), Ca^+, Mg^+, K+, Na+, Al - wet and dry deposition
sulphate (SO/"), chloride (Cl~) and nitrate (NO3~) - wet and dry
deposition
up to 5 layers (with the forest floor/organic layer being the first
horizon)
by layer
mean annual soil temperature by layer
by layer
number between 0 and the precipitation rates. If there is no lateral
flow, runoff rate should equal the precipitation rate times the % of
precipitation leaving the last soil layer.
13
      Table 4-3b. Data required to estimate BCW with the regional PROFILE model (version 5.0). The data in
      this table must be input by the user and are not currently available as a continuous coverage layers for at
      least a portion of the conterminous United States (will require development of national coverage layer).
PARAMETER
net uptake
cation uptake
nitrogen uptake
litterfall
soil water content
surface area
UNITS
kEq/ha/yr
%
%
kEq/ha/yr
mj/mj
nf/mj
DESCRIPTION
nitrogen (N), Caz+, Mg^+, K+ - only applied if biomass
through harvesting or fire
% of total soil profile (all soil layers combined should
Can be estimated using root distribution
% of total soil profile (all soil layers combined should
Can be estimated using root distribution
removed
sum to 100%).
sum to 100%).
N, Ca^+, Mg^+ and K+ - input to forest floor
by layer
soil surface area by layer
     March 2010
                                            53
Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B
PARAMETER
logKgibbsite
dissolved organic
carbon (DOC)
mineralogy
UNITS
-
mg/L
%
DESCRIPTION
by layer
by layer
% abundance of 14 dominant mineral groups (K-Feldspar,
Plagioclase, Albite, Hornblende, Pyroxene, Epidote, Garnet, Biotite,
Muscovite, Fe-Chlorite, Mg-Vermiculite, Apatite, Kaolinte, Calcite)
     Table 4-3c. Data required to estimate BCW with the regional PROFILE model (version 5.0). The data in
     this table are used to support calculations within the model and should be reviewed by the user.
PARAMETER
forest canopy
net mineralization
precipitation entering
soil horizon
precipitation leaving
soil horizon
CO2 pressure
immobilization
nitrification
denitrifi cation
nutrient uptake
kinetics
UNITS
kEq/ha/yr
kEq/ha/yr
%
%
xatm
-
-
-
-
DESCRIPTION
N, Ca^+, Mg^+, K+ - nutrients removed by or leached from
canopy
N, Ca , Mg^+, K+ - net accumulation of soil organic matter
expressed as % of precipitation. If no lateral flow, % leaving
top layer should be same as % entering underlying layer
expressed as % of precipitation. If no lateral flow, % leaving
top layer should be same as % entering underlying layer
entered as multiple of atmospheric pressure; typically ranges
from 5 in the organic horizons to 40 in the mineral soil layers
nitrogen immobilization - constant
constant
constant
coupled vs. uncoupled uptake of N and base cations / uptake
mechanism (unspecific, vanselow and none)
 2          A total of eight parameters including climate, deposition, run-off and many of the soil
 3   variables have data available as continuous coverages for the conterminous the United States
 4   (Table 4-4), and for most of these variables, data exist for all 48 states. However, some of these
 5   databases are missing variables and/or data or may need to be modified. Currently, there is no
 6   data that describes wet and dry Al deposition. This data, however, is not available in most
 7   locations where PROFILE is applied, and this parameter is typically left blank within model (H.
 8   Sverdrup personal communication, 2009b). Therefore, the  absence of this datalayer in the United
 9   States should not pose a problem for the BCW estimates. The soil temperature parameter within
10   the SSURGO database is poorly populated and data only exists for seventeen states. However,
11   mean annual air temperature is often used as a surrogate for soil temperature within PROFILE
12   because the two temperature measures are similar in some of regions (Miller, 2001). In some
13   cases, models describing the relationship between air and soil temperature are also available
14   (e.g., Yin and Arp, 1993). Therefore, the use of air temperature instead of soil temperature or
15   modeled soil temperature could be explored with the application of PROFILE in the United
16   States, if necessary.
     March 2010
54
Draft - Do Not Quote or Cite

-------
                                                                                                                       Appendix B
Table 4-4. Available datasets and databases for the conterminous United States that could be used to estimate BCW with the regional application
of the PROFILE model (version 5.0).
DATA
Total annual
precipitation
Average maximum
air temperature
Average minimum
air Temperature
Run-off
Dry cation
deposition
(NH4+, Ca2+, Mg2+,
K+, Na+)
Wet cation
deposition
(NH4+, Ca2+, Mg2+,
K+, Na+)
Dry anion
deposition
(S042-, Cr, N03')
Wet anion
deposition
(SO42", CI", NO3")
NH4+ and NO3-
wet and dry
deposition
Soil horizon depth
Soil bulk density
SOURCE URL or REFERENCE
http://prism.oregonstate.edu/products/matrix.pht
ml?vartype=ppt&view=maps
http://prism.oregonstate.edu/products/matrix.pht
ml?vartype=ppt&view=maps
http://prism.oregonstate.edu/products/matrix.pht
ml?vartype=ppt&view=maps
http://pubs.er.usgs. gov/djvu/HA/ha_710_plt.djvu
http://www.epa.gov/castnet/data.html
http://www.epa.gov/castnet/data.html,
http://nadp.sws.uiuc.edu/maplib/grids/2008/
http://www.epa.gov/castnet/data.html
http://www.epa.gov/castnet/data.html,
http://nadp.sws.uiuc.edu/maplib/grids/2008/
Community Multiscale Air Quality (CMAQ) -
http://www.epa.gov/AMD/CMAQ/
HZDEPT_R field of chorizon table
(httpV/soildatamart.nrcs.usda.gov/documents/S
SURGOMetadataTableColumnDescriptions.pdf)
DB3BAR_R field of CHORIZON table
(http://soildatamart.nrcs.usda.gOv/documents/S
SURGOMetadataTableColumnDescriptions.pdf)
DATE(S) OF
AVAILABLE
DATA
1971-2000
1971-2000
1971-2000
1951-1980
1987-2008
1987-2008,
1994-2006
1987-2008
1987-2008,
1994-2006
2002
1987-2008,
1994-2006
N/A
UNITS
in/yr
°F
°F
in/yr
kg/ha
kg/ha
kg/ha
kg/ha
kg/ha
cm
g/cm3
RESOLUTION
0.64 km2
0.64 km2
0.64 km2
1:2,000,000
86 stations in the
48 conterminous
states
86 stations in the
48 conterminous
states, 6.25 km2
86 stations in the
48 conterminous
states
86 stations in the
48 conterminous
states, 6.25 km2
12km2
1:12,000-
1:63,360
1:12,000-
1:63,360
STATES WITH COVERAGE
all
all
all
all
all (except: ID, SD, NE, NM,
and TX)
Extrapolated 400 km from
each station
all (except ID, SD, NE, NM,
and TX)
Extrapolated 400 km from
each station
all (except ID, SD, NE, NM,
and TX)
Extrapolated 400 km from
each station
all (except ID, SD, NE, NM,
and TX)
Extrapolated 400 km from
each station
all
all
all
March 2010
55
Draft - Do Not Quote or Cite

-------
                                                                                                                 Appendix B
DATA
Soil texture (%
sand, silt, and
clay)
Soil stoniness (%
of soil with
particles >2mm)
Soil temperature
SOURCE URL or REFERENCE
SANDTOT R, SILTTOT R, and CLAYTOT R
fields of CHORIZON table
(httpV/soildatamart.nrcs.usda.gov/documents/S
SURGOMetadataTableColumnDescriptions.pdf)
FRAGVOL_R in CHFRAGS table
(http://soildatamart.nrcs.usda.gOv/documents/S
SURGOMetadataTableColumnDescriptions.pdf)
SOITEMPMM field of the COSOILTEMP table
(http://soildatamart.nrcs.usda.gOv/documents/S
SURGOMetadataTableColumnDescriptions.pdf)
DATE(S) OF
AVAILABLE
DATA
N/A
N/A
N/A
UNITS
%
%
°C
(average
by month)
RESOLUTION
1:12,000-
1:63,360
1:12,000-
1:63,360
1:12,000-
1:63,360
STATES WITH COVERAGE
all
all
AK,CA,CO,GA,ID,KS,MI,MN,
MO,MT,NC,NE,NM,OR,PR,
TX,VA
March 2010
56
Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B

 1          Although a portion of the input data to estimate BCW with PROFILE are already available
 2   as national data coverages, there are nine additional input parameters that are not currently
 3   described by nationwide datasets, and nine parameters that are built into the regional application
 4   of the model and may require review and adjustment prior to applying PROFILE in the United
 5   States. The nine input parameters that would require the development of national GIS coverages
 6   or datasets that could be applied throughout the Unites States include: net uptake, % base cation
 7   and nitrogen uptake, litterfall, soil water content, surface area, logKgibbsite, mineralogy, and
 8   dissolved organic carbon (DOC).
 9          Net Uptake
10          A national dataset of net uptake of nutrients by forest systems could be developed using
11   the approach outlined by McNulty and colleagues (1997). Briefly, the United States Forest
12   Service (USFS) and United States Geological Survey (USGS) dataset describing the 21  different
13   forest types would be used to map forest cover in the 48 states, and nitrogen and base cation
14   (Ca2+, Mg2+, K+) uptake by each forest type would be determined using the average values
15   presented in Table 4-5. These values were calculated by McNulty and colleagues (2007) and
16   incorporate annual volume growth by region from the USFS Forest Inventory and Analysis
17   (FIA) database and nitrogen and base cation contents by tree species and tree component from
18   the Tree Chemistry Database (Pardo et al., 2004). Net uptake would only be necessary for sites
19   that are actively managed and experience removal of biomass through logging and/or fire.
20   Therefore, based on the assumption that only wilderness and conservation areas are not harvested
21   or managed, these nitrogen and base cation uptake estimates would only be applied to forest
22   areas that are not designated as wilderness by the National Wilderness Preservation System of
23   the United States (McNulty et al., 2007).
24
     March 2010                                 57                   Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B
     Table 4-5. Nitrogen and base cation uptake by forest type
     (from McNultyetal., 2007).
FOREST COVER TYPE
white-red-jack pine
spruce fir
longleaf slash pine
loblolly shortleaf pine
oak pine
oak hickory
oak-gum-cypress
elm-ash-cottonwood
maple-beech-birch
aspen-birch
douglas-fir
hemlock-sitka-spruce
ponderosa pine
western white pine
lodgepole pine
Larch
fir-spruce
Redwood
Chaparral
pinyon-juniper
western hardwoods
NITROGEN
UPTAKE
(eq/ha/yr)
59.07
54.27
154.74
140.41
129.71
102.56
124.18
79.74
101.76
81.69
109.89
98.88
75.29
40.69
40.19
65.1
94.65
100.92
106.6
40.87
135.21
BASE CATION
UPTAKE (eq/ha/yr)
77.14
83.72
227.22
208.58
213.75
254.87
235.68
156.3
190.51
125.46
179.03
161.12
174.39
37.11
61.25
77.14
146
156.62
201.61
58.21
263.33
 2          Soil Surface Area
 3          Soil surface area is commonly determined in the laboratory using the Brunauer-Emmett-
 4   Teller (BET) nitrogen absorption technique (Hodson et al., 1997). However, data from such
 5   analyses are not available for all soils in the United States. Therefore, it would be necessary to
 6   estimate surface area from other soil data. Within the PROFILE model, surface area is calculated
 7   with soil texture and particle size distribution data (Equation 4-5) (Alveteg et al., 2004), and
 8   Sverdrup and colleagues (1992), used this equation in their study of critical acid loads in
 9   Maryland. This same approach could be used for mapping soil surface areas in the United States.
10   Soil texture is part of the U.S. Department of Agriculture- Natural Resources Conservation
11   Service (USDA-NRCS) Soil Survey Geographic (SSURGO) database (Table 4.0). Therefore, it
     March 2010
58
Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B
 1   would be possible to produce a continuous coverage map of soil surface areas in the United
 2   States.
                                                                                         (4-5)

 4   where
 5                 Aw = total exposed surface area (m2/m3)
 6                 x = weight fraction of clay, silt and sand when xciay + xs;it + xsand = 1 ;
 7                 p = soil density in kg/m3
 8          Soil Mineralogy
 9          Soil mineralogy is one of the most important and influential variables within PROFILE.
10   However, it is also a very time intensive and expensive measurement. Therefore, soil mineralogy
11   data in the United States is sparse, and a continuous coverage layer of soil mineralogy does not
12   exist. In most regional applications of PROFILE in Europe and other regions, the mineralogy
13   input data are based on a combination of data from soil geochemical and mineralogy analyses
14   and mineralogical composition based on output from a model such as the Analysis to Mineral
15   (A2M) model (Posch and Kurz, 2007). The A2M model estimates all  possible mineral
16   compositions from total chemical analyses (Ca2+, Mg2+, K+, Na+, Ti, Al, phosphorus (P), silicon
17   (Si), iron (Fe)) of the soil and a pre-specified set of minerals that are likely to be present in the
18   soil. The highest probability mineral composition is an output of the arithmetic mean of all
19   extreme mineral modes. The resulting mineralogies are then mapped to "geological provinces"
20   (Sverdrup et al., 1990) that have the same parent material bedrock but may differ in soil
21   mineralogy in a consistent pattern (Sverdrup et al., 1990). Alternatively, the mineralogies can be
22   mapped to "mineralogy polygons" that are delineated based on probable similarities in mineral
23   compositions of the soils. Typically, the spatial borders of mineralogy polygons are determined
24   by underlying parent material geology and/or soil type groupings that are likely to have the same
25   mineralogies (H. Sverdrup personal communication,  2009b). In areas where the soils have
26   formed from transported materials, such as glacial till, it is sometimes necessary to consider the
27   surficial geology and model the origin and transport of materials to determine the parent material
28   geology (McKenzie and Ryan, 1999).

     March 2010                                 59                   Draft - Do Not Quote or Cite

-------
                                                                                     Appendix B

 1          Due to the diverse geological history of the United States, it may be necessary to include
 2    a variety of variables and databases in the characterization and mapping of mineralogy. Parent
 3    materials underlying soils in the conterminous United States vary extremely. These materials
 4    include not only bedrock beneath young soils, but also a variety of young and old regolith
 5    materials that include both residuum formed in place and all varieties of transported sediments.
 6    In addition, the soils are  old and highly weathered in a large portion of the United States, and
 7    therefore, no longer resembles the mineral composition of the parent material. For example, the
 8    mineralogy of soils atop  ancient residuum of the Appalachian region will vary significantly from
 9    the mineralogy of younger residuum of the Western mountain ranges. Also, the mineralogy of
10    soil developed on the older loessal plain in the Mississippi basin will vary from the younger
11    glacial deposits along the northern regions of the United States.  Therefore, determination of soil
12    mineralogy in the United States would require an approach that is able to recognize the varied
13    geological histories, different parent material origins, and soil mineralogies that differ from the
14    original parent material sources. Such an approach would involve the following steps be
15    conducted simultaneously:
16          1.  Delineation of mineralogy polygons based on soil classification at a level supported
17              by available data
18          2.  Determination of mineralogy and geochemical data availability for each mineralogy
19              polygon
20          3.  Comparison of mineralogy polygons with underlying bedrock and surficial geology
21          4.  Testing modeled mineralogy against actual mineralogy measurements
22          Delineation of mineralogy polygons based on soil classification at a level supported by
23    available data -_Mapping and creation of a national GIS coverage of mineralogy in the
24    conterminous United States would require the delineation of "mineralogy polygons".
25    "Mineralogy polygons" are spatially-defined polygons that are delineated based on probable
26    similarities in mineral compositions of the soils. These polygons would need to be large enough
27    in scale to be adequately  covered  by available mineralogy and soil analysis data, yet small
28    enough to  only represent single assemblages of soil minerals. Ideally, each mineralogy polygon
29    should have at least one data point or soil profile analysis that describes the total analysis (Ca2+,
30    Mg2+, K+, Na+, Ti, Al, P, Si, and Fe) and/or mineralogy of the soil  layers. Where data are
31    missing, it would be necessary to  interpolate data from other locations using correlations with

      March 2010                                 60                   Draft - Do Not Quote or Cite

-------
                                                                                     Appendix B
 1   surrounding and adjacent known data and other supporting criteria indicative of similar
 2   mineralogies and weathering patterns (e.g., geologic and physiographic regions, bedrock geology
 3   data, climatic regions, and others).
 4          Within the United States, one of the most suitable coverages for the delineation of the soil
 5   mineralogy polygons is the SSURGO soils database (Table 4-6). The smallest unit within this
 6   database is the soil mapping unit which can consist of up to five individual soil series. A soil
 7   series is defined as "soils that are similar in all major profile characteristics (Brady and Weil,
 8   2002), and soils within the same series have been influenced by similar climate, topographic
 9   location, biota, parent material and pedological time frame. Therefore, the soil within a series,
10   regardless of location would be expected to have identical or sufficiently similar mineralogies
11   (C. Smith personal communication, 2009). The soil groupings within the higher levels of soil
12   taxonomy may also be based on characteristics such as soil mineralogy. For example, soil orders
13   are largely classified by the degree of weathering and soil development, with Entisols
14   representing the youngest, least weathered soils, and Ultisols and Spodosols being more highly
15   weathered. Therefore, it may be possible to group the soil mapping units  at a higher level of
16   taxonomy, such as the great group, family or order, as the "mineralogy polygons". However,
17   since soils are classified based on multiple formative factors, the "mineralogy polygons" could
18   be a mixture of groupings based on different levels of soil taxonomy, with all groupings based on
19   factors indicative of similar mineral assemblages in the soil.
20          Detailed soil delineations have been completed for more than 80% of the conterminous
21   United States and are used in the NRCS Soil  Survey Geographic (SSURGO) dataset (Figure 4-
22   3). Data are missing for many public land areas (e.g., national forest lands), and there are
23   approximately 21,000 soil series delineations within the conterminous United States.
     Table 4-6 Datasets with Geochemical and  Mineralogy Data for U.S. Soils
DATA
Soil Survey
Geographic
Database
(SSURGO)
SOURCE
httpV/soils.usda.gov/survey/g
eography/ssurgo/
RESOLUTION
1:12,000 to
1:63,360
INCLUDED DATA
SSURGO is linked to a National Soil
Information System (NASIS) attribute
database. The attribute database gives
the proportionate extent of the
component soils (i.e., u soil series) and
their properties for each map unit. The
SSURGO map units consist of 1 to 3
components each. There are
approximately 15,000 and 20,000 soil
series polygons delineated across the
United States
     March 2010
61
Draft - Do Not Quote or Cite

-------
                                                                                             Appendix B
      DATA
         SOURCE
RESOLUTION
           INCLUDED DATA
U.S. General Soil
Map (STATSGO
http://soils.usda.gOV/survey/g
eography/statsgo/
1:250,000
The tabular data contain estimated
ranges (low, high, and representative
values) of physical and chemical soil
properties, soil interpretations depicting
the range for the geographic extent of the
map unit. Soil map units are linked to
attributes in the tabular data, which give
the proportionate extent of the
component soils and their properties.
Surficial Geology
of the United
States (1977)
(also Map of
Surficial Deposits
and Materials in
the Eastern and
Central United
States (East of
102° West
Longitude))
http://tin.er.usgs.gov/geology/
state/ or
http://water.usgs.gov/GIS/met
adata/usgswrd/XML/ofr99-
77_geol75m.xml also
http://pubs.usgs.gov/imap/i-
2789/
1:7,500,000
(Eof 102° W
Longitude:
1:2,000,000)
Provides approximate areal extent of
about 45 categories of regolith types
across the conterminous United States.
Compilation East of 102° West Longitude
has further classified deposits generally
within original polygons.
Element
Concentrations in
Soils and Other
Surficial Materials
of the
Conterminous
United States
(Shacklette Data,
1977)
USGS, Denver Federal
Center Offices
Sampling
density: 1
sample per
6,000km2.;
equivalent to
the collection
of samples on
a 75-km grid.
Ultra-low-density geochemical baseline
data from 1,323 samples locations
characterizing soils and other surficial
materials in the conterminous United
States. Elements analyzed included: Ag,
Al, Ba, Be, B, Ca, Ce, Cr, Co, Cu, Ga,
Ge, Hg, Fe, La, Li, Pb, Mg, Mn,  Mo, Na,
Nd, Ni, Nb,  P, K,  Rb, S, Sc, Se,  Sr, Th,
Ti, U,  V, Yb, Y, Zn, Zr, and total carbon.
The National
Geochemical
Survey -
Database and
Documentation
(Version 5.0, on-
going)
http://tin.er.usgs.gov/geoche
m/doc/home.htm and
http://tin.er.usgs.gov/geoche
m/
Nominal grid
spacing of 17
by 17
kilometers (i.e.,
minimum
sample density
of 1 sample
per 289 km2 in
all land areas
of the country
Stream-sediment-based geochemical
survey for the United States; Analytical
methods include a 40-element ICP
package plus single-element
determinations of As, Se, and Hg by
atomic absorption for every sample.
about 60,000 stream-sediment samples
that have been analyzed. Digital data
files are presented in 6 categories.  In
total there are 43 individual data files for
the Unites States. Some of the data has
also been processed into vector data to
produce maps showing the elemental
concentration of As, Se, Hg, Pb, Zn, Cu,
Al, Na, Mg,  P, Ca, Ti, Mn, and Fe at the
county level. Database contains 287
attributes (77,212 records).
Integrated
Geologic Map
Databases for the
United States
(1998-2007)
http://gsa.confex.com/gsa/20
06AM/finalprogram/abstract_
110914.htm and
http://tin.er.usgs.gov/geology/
state/
1:100,000
Seamless national-scale geologic spatial
data-layer and database to support
national and regional level projects,
including mineral resource and
geoenvironmental assessments. Data
include general geologic unit age,
dominant lithology (rocktypel must be
>50% of unit) and second most dominant
lithology (rocktype2).
March 2010
                                62
                           Draft - Do Not Quote or Cite

-------
                                                                               Appendix B
DATA
Soil Pedon Pit
Data (on-going)

























Physiographic
Regions of the
United States
(Fenneman, 1946)


SOURCE
USDA NRCS


























http://water.usgs.gov/GIS/dsd
I/physio. eOO.gz




RESOLUTION
Area covered
by a pedon
varies from 10
- 100 square
feet;
approximately
30,000 soil
pits/pedons in
the NRCS
database

















1:7,000,000





INCLUDED DATA
Geochemical elements: Al, Ca, Fe, K,
Mg, Mn, Na, P, Si, Sr, Ti, and Zr. X-ray
diffraction for clay mineralogy by horizon;
optical mineralogy analysis is performed
on the dominant sand fractions of the soil
from the A-horizon, B-horizon, and C-
horizon, or the most dominant horizon.
More than 60 fields describing the
minerals are listed in the database. The
dataset is not uniform in that elemental
analyses were routinely done through the
1970's but then these analyses were
suspended through the 1980's.
Elemental analyses were resumed during
the early 1990's. It is estimated that as
much as one third of the 30,000 soil
pedons have geochemical data.
Likewise, optical mineralogy is not
performed for all pedons and the NRCS
staff estimate that approximately as
many as one third of the 30,000 soil pits
have optical analysis results. Even
though the number of pedons with data
are similar for geochemical and optical
analysis results, the data are not
necessarily associated with the same set
of pedons or even soil series.
Geomorphic / physiographic broad-scale
subdivisions based on terrain texture, rock
type, and geologic structure and history.
Nevin Fenneman's (1946) three-tiered
classification of the United States - by
division, province, and section.
March 2010
63
Draft - Do Not Quote or Cite

-------
                                                                                                                         Appendix B
           I mage source:
           http://soils.usda.gov/survev/geography/ssurgo/
                                                               Spatial and Tabular
                                                               Tabular Only
                                                               No Data
       Figure 4-3. Map Showing the Distribution and Status of SSURGO Data
March 2010
64
Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B

 1          Determination of mineralogy and total analysis data availability for each mineralogy
 2   polygon - The soils for each spatially-defined "mineralogy polygon" would require %
 3   mineralogy to determine BCW with the PROFILE model. The relative abundances of 14
 4   dominant minerals are required as model input, and this % mineralogy can be based on direct
 5   measurements of soil mineralogy or can be determined with the A2M model. As outlined earlier,
 6   the A2M model is able to estimate the most probable % mineral composition, or proportion of
 7   mineral phases, of a soil based on total analysis data and the identification of the minerals that
 8   are likely to be present in the soil. Therefore, it would be necessary to determine the availability
 9   of such data for each of the "mineralogy polygons" in the conterminous United  States.
10          Currently, there are potentially three consistent national-scale datasets that contain
11   various levels of mineralogy and total analysis data to serve as inputs for the A2M and PROFILE
12   models. These include:
13       •  Chemical Analyses of Soils and other Surficial Materials of the  Conterminous United
14          States (Shacklette dataset) and accompanying Geochemical Landscapes  Project data,
15       •  the more recent National Geochemical Survey data, and
16       •  the United States Department of Agriculture (USDA) NRCSpedon soil pit dataset.
17   A summary  of these datasets is outlined in Table 4-6.
18          Chemical Analyses of Soils and other Surficial Materials of the  Conterminous United
19   States (Shacklette Data) and the Geochemical Landscapes Project datasets provide geochemical
20   baseline data for soils and other surficial materials in the conterminous United States. The
21   original Shacklette dataset contains geochemical data from soils and other regolith collected and
22   analyzed by Hans Shacklette and colleagues beginning in 1958 and continuing until about 1976.
23   This dataset has approximately 1,323 samples, at a sampling density of approximately 1 sample
24   per 6,000 square kilometers (Figure 4-4). The soil  samples within this dataset were analyzed for
25   a large number of elements, including Ca, Fe, Mg, Na, P, K, and Ti (Gustavsson et al., 2001),
26   that are required by the A2M model. However, assessments of mineralogy were not included in
27   these original analyses. An additional drawback with the data set is its extremely low numbers of
28   samples for the entire conterminous United States. However, more recent high-resolution studies
29   (e.g., Smith  et al., 2005) for select elements (e.g., Calcium) have illustrated that  the regional
30   patterns established by the Shacklette data are generally maintained except where areas have
31   been affected by anthropogenic factors (Smith, 2006).

     March 2010                                 65                  Draft - Do Not Quote  or Cite

-------
                                                                                                                    Appendix B
                                                                                                         ••••
                                                                                                                    o
                                                                                                                     Dr.,
      •Blackdotsindicatesamplesfrom sample-collection •* A -.

                                           '°otf
      •whitedotsindicatesamplesfromsample-collection  " •-
      Phase2,                                          *L O
      •gray dotsindicatesampleswhoseplacement into             • Q
      phase 1 or phase 2 is uncertain.
      (All archived sample have been reanalyzed (personal
      communication with David Smith, 1-4-2009)
                                                                      Shacklette, HansfordT., and Josephine G. Boerngen, 1984

       Figure 4-4. Soil Sampling Locations Included in the USGS Shacklette Dataset




March 2010                                                    66                                     Draft - Do Not Quote or Cite

-------
                                                                             Appendix B

       The Geochemical Landscapes Project was begun in 1998 with most work occurring after
2002. The purpose of the data collection is to increase the density of the Shacklette data locations
to produce a high resolution geochemical dataset for North American soils of 6,000 data points
(D. Smith personal communication, 2009). This is an on-going collaborative effort by the USGS,
USDA Natural Resource Conservation Service, other federal agencies, and academia to build a
national-scale soil geochemical survey.  The project has just completed a third year of
continental-sampling and completed sample collection for approximately 80% of the
conterminous United States (D. Smith personal communication, 2009). The USGS anticipates
that sampling may be completed for the conterminous United States in 2010; or 2011 at the
latest. Both total and mineralogy analyses are being performed on these samples. Mineralogy
analyses include x-ray  diffraction on the clay fraction and optical analyses on the fine sands and
silts. In addition, the original Shacklette data have been re-analyzed for mineralogy.
       The National Geochemical Survey (NGS) dataset is being built by on-going efforts by the
USGS to produce a new stream-sediment-based geochemical survey for the United States at a
spacing of 17 by 17 kilometers (i.e., minimum sample density of 1 sample per 289 km2 in all
land areas of the country) (Figure 4-5). The project has sought to capitalize on existing datasets
and archived samples. For this reason the NGS is based primarily on analyses of stream
sediments to build on the massive archives of data and samples from DOE's National Uranium
Evaluation (NURE) program. Much of the survey has entailed reanalysis of approximately
35,000 archival samples from the NURE program. Where NURE samples do not exist, USGS
has been working with cooperators to obtain new samples. In total, the project is expecting to
have more than 60,000 samples. Most or all of the sampling has been completed for the
conterminous United States and only few analyses are left to complete (D. Smith personal
communication, 2009). The samples are being analyzed for 40 elements, including all of the
elements which are necessary input for the A2M model. In addition, for a select number of
samples mineralogy analyses (x-ray diffraction of clay fraction and optimal analyses of fine
sands and silts) are also being conducted.
March 2010                                67                   Draft - Do Not Quote or Cite

-------
                                                                                      Appendix B
      Spacing   Density
         (km)   N/1QOkmA2)
         <3
         3-4
         4-5
         5-7
         7-9
         9-14
         14-20
         >20
63-110
40-63
20-40
12-20
5-12
3-5
<3
     Image source: http://tin.er.usgs.gov/geochem/doc/status.htm

     Figure 4-5. Sample Density of USGS National Geochemical Survey
March 2010
                             68
Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B

 1          USD A NRCS Soil Pedon Pit Data were collected by the USD A NRC S for data required
 2   for delineation of soil series, map units, and associated attributes. The data are contained in the
 3   NRCS USSOILS database that provides data for the SSURGO database. There are currently
 4   approximately 30,000 soil pits/pedons in the NRCS database, and soil samples from these pits or
 5   pedons have been analyzed for a large variety of physical and chemical properties. These
 6   analyses include total chemical analysis, which includes elements required by A2M (e.g., Al, Ca,
 7   Fe, K, Mg, Na, P, Si, Ti). In addition, mineralogy has been characterized through two analyses:
 8   x-ray diffraction, which identifies clay mineralogy, and optical mineralogy which determines the
 9   mineral composition of the fine sand and silt fractions of the soil (C. Smith personal
10   communication, 2009). However, these three analyses have not been conducted on all soils. Only
11   11,747 of the 30,000 soil pits have been analyzed for at least one of the three parameters (Figure
12   4-6), and only 4,710 soils have been analyzed for all three (Figure 4-7).
     March 2010                                 69                  Draft - Do Not Quote or Cite

-------
                                                                                                                          Appendix B
1
2

3

4
Figure 4-6. NRCS Soil Pedon Sample Pit Locations (30,000 total)
(Image created by RTI using data provided by NRCS on 12/2009)
     March 2010
                                             70
Draft - Do Not Quote or Cite

-------
                                                                                                                        Appendix B
                 ?*•*.
                                         . • r
1




2




3
Figure 4-7. NRCS Soil Pedon Pit Sample Locations with Geochemical and Mineralogy Data



(Image created by RTI using data provided by NRCS on 12/2009)
    March 2010
                                                       71
Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B

 1          In summary, three main datasets have been identified that could provide the necessary
 2   total analyses and mineralogy data for each "mineralogy polygon." These datasets would be
 3   combined into a single database and overlaid on top of the "mineralogy polygon" datalayer to
 4   determine the degree to which each polygon is covered by mineralogy and total analysis data. At
 5   the scale of a nationwide analysis, the data from each of these datasets is considered comparable
 6   given the sampling and analysis protocols that have been used (D. Smith personal
 7   communication, 2009). Although the combined database would be large and offer over 75,000
 8   data points, it is not likely that data would be available for all "mineralogy polygons." In such
 9   cases, it would be necessary to determine the mineralogy through alternate methods. Potentially,
10   interpolation between data points could be conducted using numerical probabilistic methods. In
11   addition, it may be possible to determine probable mineralogy based on underlying bedrock or
12   surficial geology (described further in next session). Methods involving professional judgment
13   could also be used to interpret patterns and assign reasonable and appropriate values to express
14   the apparent condition. If such an approach were taken, it would be necessary to work with soils
15   experts who are familiar with the SSURGO database (e.g., NRCS Regional Staff) to make such
16   judgments.
17          Comparison of mineral polygons with underlying bedrock and surficial geology - In
18   many locations in the United States, soils have developed from the underlying bedrock or
19   surficial materials. Therefore, it may be possible to validate, support or identify the mineralogies
20   of each of the "mineralogy polygons" through a comparison with the physiographic regions of
21   the United States and the underlying bedrock and surficial geology. The physiographic provinces
22   are based on geology and topography. Therefore, these provinces relate geology and geological
23   history with expected soil characteristics, and the locations of "mineralogy polygons" should
24   broadly follow the patterns within these province boundaries. Similarly, the "mineralogy
25   polygons" could be compared against the underlying geologies to determine the accuracy of the
26   soil taxonomy groupings that were used to delineate the polygons. In addition, overlays of the
27   "mineraology polygon" datalayer and bedrock or surficial geology could support the estimation
28   of probable mineralogies and percent compositions for "mineralogy polygons" that are missing
29   soil mineralogy and or total analysis data.
30          The USGS 1:100,000  scale bedrock geology GIS cover would be first used for the
31   comparison between the "mineralogy polygons" and bedrock geology (Table 6.0). Most rock

     March 2010                                  72                   Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B

 1   types are typically characterized by less than 4 mineral types. Correlating the "mineralogy
 2   polygons" with the bedrock type can be used to obtain a gross approximation of mineral phases
 3   that would be expected in the residual parent materials and the corresponding soils. This would
 4   be a particularly useful protocol to apply to areas where the soils have formed in place from the
 5   weathering of the bedrock. For example, this approach could be used in unglaciated regions
 6   where Entisols or residuum predominant.
 7          The 1:7,500,000 scale USGS surficial geology layer could be used as the source for
 8   comparison between "mineralogy polygon" and surficial geology (refer to Table 6.0). The
 9   surficial geology layer would identify the type of regolith on which the soil has developed.
10   Regolith is defined here as any unconsolidated materials on top of bedrock, and consists of
11   residuum which has formed in place and transported materials that have been deposited by
12   gravity, wind, water or ice. Therefore, this layer will indicate the type of parent material that
13   supported the development of the soil and will provide an indication of the potential mineral
14   composition of the soil. Specific correlation of mineral types can be more difficult for
15   transported deposits. However, an association is still possible as correlated with general up-grade
16   areas that relate the likely origin, or areas of parent material, for the transported deposit. Even
17   though more generalized approaches to determining the mineralogy are suggested by the
18   available data, modeling the geologic source of parent materials by applying techniques similar
19   to soil-landscape modeling or environmental correlation modeling could be conducted
20   (McKenzie and Ryan, 1999).
21          Test modeled mineralogy against actual mineralogy measurements - To validate and test
22   the accuracy of the %  mineral values assigned to the "mineralogy polygons", comparisons
23   should be made between the "mineralogy polygon" data layer and areas with detailed mineralogy
24   soil analyses. Such sites may include the LTER sites outlined in Table 4-7 or those detailed
25   within the scientific and geological literature. In addition, there may be locations where detailed
26   mineralogy assessments have been conducted by mining companies or research groups that could
27   be used to test the "mineralogy polygon" data layer. Comparisons would be particularly
28   important for mineralogy polygons with % mineralogy determined by the A2M model.
     March 2010                                 73                   Draft - Do Not Quote or Cite

-------
                                                                                    Appendix B
     Table 4-7. Long-Term Ecological Research (LTER) sites that could potentially be suitable as "field test"
     sites to validate BCW estimates generated with the regional application of the PROFILE model (version
     5.0).
LTER STUDY
H.J. Andrews
Experimental Forest
Coweeta LTER
Harvard Forest
Hubbard Brook
Experimental Forest
Kellogg Biological
Station
Konza Prairie LTER
Niwot Ridge
Santa Barbara Coastal
LTER
Sevilleta LTER
LOCATION
Cascade Mountains, Oregon
Southern Appalachian
mountains, North Carolina
Massachusetts
White Mountain National
Forest, New Hampshire
Southwest Michigan
Northeastern Kansas
Colorado
California
New Mexico
ADDITIONAL INFORMATION
http://andrewsforest.oregonstate.edu/
http://www.lternet.edu/sites/cwt/
http://www.lternet.edu/sites/hfr/
http://www.lternet.edu/sites/hbr/
http://www.lternet.edu/sites/kbs/
http://www.lternet.edu/sites/knz/
http://www.lternet.edu/sites/nwt/
http://www.lternet.edu/sites/sbc/
http://www.lternet.edu/sites/sev/
 1
 2          PROFILE Input Parameters Assigned Default Values
 3          Development of national datasets or default values for the % base cation and nitrogen
 4   uptake (by layer), litterfall, soil water content, logKgibbsite, and DOC input parameters would
 5   most likely require the use of data from the literature and research conducted in the United
 6   States. The % base cation and nitrogen uptake by soil layer variables are a function of the
 7   distribution of fine roots, and rooting distributions are typically entered as one of four classes
 8   into PROFILE (H. Sverdrup personal communication, 2009b). These root distribution classes are
 9   based on data from an extensive literature search on the rooting habitats of common tree species
10   in Europe (Sverdrup and  Stjernquist, 2002). A similar literature search could be conducted for
11   the main species within the 21 forest types in the United States, and the four root distribution
12   classes could be adjusted accordingly.
13          The litterfall parameter within PROFILE characterizes the amounts of N,  Ca2+, Mg2+ and
14   K+ returned to the soil with the  senescence of leaves, branches and stems. It is calculated as a
15   function of the site-specific growth rates of individual tree species and the nutrient content of the
16   different litter components. Since PROFILE is a steady-state model, the growth rates are
17   averaged over the rotation of the stand. Litterfall values have been determined for European tree
18   species (Sverdrup et al, 1990; Sverdrup and Stjernquist, 2002), and the same procedure could be
19   used for estimating values for the main species in the 21 forest types  in the United States. The
     March 2010
74
Draft - Do Not Quote or Cite

-------
                                                                                    Appendix B

 1   Tree Chemistry Database (Pardo et al., 2004) could be serve as the source of litter nutrient
 2   content and the USFS FIA database could potentially supply species-specific growth rates.
 3          Soil water content is highly variable. However, since PROFILE is a steady-state model, it
 4   is necessary to use a single value representative of the water content throughout the year. In
 5   Sweden, a default value of 0.2 m2/m3 is often used (Halveteg et al., 2004), and it would be
 6   necessary to establish a similar default value or set of default values for the United States. Such
 7   values could be obtained from the literature. In addition, it may be possible to estimate a set of
 8   soil water content estimates based on a simple water balance model that includes the influences
 9   of precipitation, run-off, soil texture and/or soil drainage classes (H. Sverdrup personal
10   communication, 2009b). Data outlined in Table 4 and soil texture and the six drainage classes
11   (Well Drained; Excessive; Moderately Well; Poorly; Somewhat Excessively;  Somewhat Poorly)
12   included within the SSURGO soils database could potentially be used in this simple water
13   balance model.
14          Soil dissolved organic carbon (DOC) and the logKgibbsite coefficient would also require
15   the use of values from the literature. Currently, with the application of PROFILE within Europe,
16   DOC is entered as 20  mg/L in the organic layers but drops rapidly with depth  in the mineral soil
17   horizons (Alveteg et al., 2004). These values are based on a compilation of data from European
18   field sites (H.  Sverdrup personal communication, 2009b) and are a function of the organic matter
19   content of the soil (Sverdrup et al, 1990). Similar values and relationships would need to be
20   established for forest systems in the United States based on available data and studies outlined in
21   the literature. LogKgibbsite is a coefficient that describes the concentration of Al in the soil
22   solution. It depends on the soil solution pH and differs by soil layer. Two sets of values have
23   been developed for the application of PROFILE within Europe, with one set being used for clay
24   soils and the other for non-clay soils (Sverdrup and Stjernquist, 2002). These values and
25   grouping by soil clay content were based on data from the literature and the consistent trends in
26   the gibbsite coefficients within and between soils (H. Sverdrup personal communication, 2009b).
27   For the application of PROFILE within the United States, it would be necessary to review the
28   logKgibbsite values, review the literature and potentially adjust the values as necessary to be
29   representative of conditions found in the United States.
     March 2010                                  75                  Draft - Do Not Quote or Cite

-------
                                                                                    Appendix B

 1          Regionally-based Built-in PROFILE Input Parameters
 2          There are nine variables that are currently built into the calculations of BCW for the
 3   regional application of PROFILE (version 5.0) and do not require input data from the user. The
 4   values of these variables were determined by field research in Europe, and are thought to vary
 5   minimally between sites or are calculated based on the input data. These variables include: forest
 6   canopy, net mineralization, % precipitation entering layer, % precipitation leaving layer, CC>2
 7   pressure, immobilization, nitrification, denitrification, nutrient uptake kinetic variables (C.
 8   Aksehson personal communication, 2009). Prior to applying PROFILE to map BCW throughout
 9   the United States, the values and equations used for each of these variables should be examined.
10          It may be necessary to modify the model equations and/or replace  the current values with
11   those from the literature to ensure that the values  within PROFILE are representative of
12   conditions  and processes in the United States (H.  Sverdrup personal communication, 2009).
13   Forest canopy, within PROFILE, accounts for the Ca2+, Mg2+, K+ and N (as NH4+) that is
14   absorbed from or leached into the precipitation that is in contact with the canopy. Potassium,
15   Ca2+, Mg2+ are typically leached from the foliage and NH4+ is absorbed. The default values
16   within PROFILE are currently based on the results of field studies in Europe and are divided by
17   forest type (deciduous versus non-deciduous). Net mineralization within the model is a function
18   of soil organic matter content. Currently, net mineralization is set to "0" within PROFILE
19   assuming that the forests are managed sustainably and net mineralization is at an equilibrium;
20   Ca2+, Mg2+, K+ and NH4+ released through mineralization of organic matter is taken up by the
21   vegetation, returned as litter and remineralized. Therefore, there is no net loss or gain of nutrients
22   through mineralization. However, the net mineralization default value of "0" can be changed if
23   forest management is not sustainable and involves short rotations and/or practices such as whole
24   tree harvesting which remove the foliage and a large pool of "mineralized" nutrients from the
25   site.
26          The % of precipitation entering and leaving the soil layers variables within PROFILE are
27   determined based on the fine root distribution in the soil profile. Carbon dioxide pressure in the
28   soil is estimated from a small dataset of measurements conducted in different regions of the
29   world. The values that are used within PROFILE are a function of soil particle size.
30   Immobilization of nitrogen within PROFILE is currently set to range between 0.5 - 1.0 kg
31   N/ha/yr. This range of values was determined by  the amount of the amount of nitrogen that has
     March 2010                                 76                   Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B

 1   accumulated in Northern European soils since the last glaciations. Denitrification and
 2   nitrification are currently determined by mathematical equations that include the influences of
 3   temperature, available soil nitrogen, soil moisture and soil pH. Nutrient uptake kinetics within
 4   PROFILE consists of coupled versus uncoupled uptake of nitrogen and base cations. Within
 5   PROFILE, uptake is set to "coupled" as a default because the uptake of Ca, Mg and Al and K
 6   and NH4+ are coupled (Sverdrup et al., 1990). Uptake kinetics within the model are also
 7   described as unspecified or vanselow depending on the uptake dynamics of base cations and Al
 8   absorbed to the root surface. Currently, within PROFILE, deciduous species and domestic crops
 9   are defaulted to vanselow kinetics and grasses, and conifers use unspecified kinetics (H.
10   Sverdrup personal communication, 2009). Unspecified kinetics indicates that the ion exchange
11   matrix on the root surface is indifferent to the valence of the absorbing ions (Sverdrup and
12   Warfvinge, 1993).
13          Step 2. Determination of polygon layer to spatially define the BCW rates and
14                  development of continuous coverage map of calculated BCW values.
15          Following the establishment of continuous coverage databases and national datasets and
16   default values for the application of PROFILE (version 5.0) within the United States, it would be
17   necessary to construct a spatially-explicit continuous datalayer for mapping BCW throughout the
18   48 states. The  resolution of the datalayer should be small scale and provide the highest level of
19   detail permitted by the data. In addition, the location of individual BCW polygons should be tied
20   to a variable or set  of variables which strongly influence BCW. Since soil attributes including
21   mineralogy, bulk density, volumetric water content and exposed surface area of minerals
22   (discussed further in Section 4.3.6) are the largest sources of variability in the BCW calculations,
23   it may be most appropriate to map BCW according to mineralogy, soil series or a higher level of
24   soil taxonomy. Input data and default values for the 26 PROFILE variables would then be
25   mapped to the delineated BCW polygon layer. When multiple or sections of multiple polygons of
26   the same datalayer are present in a BCW polygon, a weighted average value for the data would be
27   calculated. All the data for each BCW polygon would then be formatted according to the
28   requirements of PROFILE and the PROFILE regional model would be run to produce maps of
29   BCW.
     March 2010                                 77                  Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B

 1   4.3.5  Potential limitations of proposed methodology
 2          Although PROFILE is arguably the most suitable model currently available for
 3   estimating and mapping BCW for terrestrial critical acid load determinations in the United States,
 4   the model does have some limitations that should be acknowledged and potentially remedied
 5   prior to application. The model and algorithms contained therein were developed in Sweden
 6   using Swedish soils as the basis for the soil chemical and physical relationships (Hodson et al.,
 7   1997). The soils  in Sweden are comparatively young, having formed since the last glaciations,
 8   approximately 10,000 years ago (Sverdrup and Warfvinge, 1988). Therefore, there is some
 9   concern that PROFILE may not accurately model base cation release in older soils (C. Smith
10   personal communication,  2009). As discussed by Hodson and Langan (1999), PROFILE does
11   not take into account the decreasing reactivity of minerals with duration of dissolution, and
12   assumes that the reaction  rates are constant regardless of time and duration of dissolution. In
13   addition, the model assumes a constant versus decreasing reactive surface area as total surface
14   area increases. According to the authors, these shortfallings were two of the main reasons that
15   PROFILE did not show a decreased weathering rate with soil age relative to other models.
16   However, at the  same time PROFILE has been used to estimate BCW in multiple locations with
17   older, more weathered soils, such as Maryland, China, Thailand, Argentina and Greece, and has
18   performed with apparent success (Duan et al., 2002; Sverdrup et al., 1992; H. Sverdrup).
19          PROFILE currently accounts for the weathering of 14 different minerals, with the
20   potential to include 13 additional minerals, if necessary. Potentially, there may be minerals
21   within the United States that are not represented within the 27 that are currently included within
22   PROFILE. However, a total of 48 minerals have been investigated by the researchers that
23   developed the model (H. Sverdrup personal communication, 2009b). Therefore,  it may be
24   possible to add additional minerals to PROFILE to ensure that it is able to address BCW in all
25   regions of the United States.
26          Additional limitations and concerns regarding the application of PROFILE to estimate
27   BCW rates have been identified in a  thorough review by Hodson and colleagues (1997). Some of
28   the main issues brought up by the authors include: the need for a more consistent set of constants
29   for the weathering rate  equations; inaccuracies in the mineral compositions; errors in the
30   calculation to determine surface area; and  confounding influences of soil particles greater than
31   2mm in size on soil bulk density. Hodson and colleagues (1997) point out the need to reexamine

     March 2010                                 78                   Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B

 1   the reaction rate coefficients associated with hornblende, tourmaline, staurolite, kaolinite, garnet,
 2   augite, biotite and chlorite, arguing that coefficients assigned to these minerals are not correct.
 3   Similarly, the authors claim that the compositions of the minerals used within PROFILE may be
 4   incorrect in some applications and may need to be modified by the user to more accurately
 5   reflect the soil being modeled. Hodson and colleagues (1997) also demonstrate the potential to
 6   over and underestimate BET surface area using the soil texture equation provided within
 7   PROFILE. They claim that the equation underestimated the surface area of a British soil by 65%.
 8   In part, the authors attributed these inaccuracies to the development of the soil texture - surface
 9   area relationship from only 92 mineral soil samples from Sweden. Lastly, Hodson and colleagues
10   (1997) point out the need to recognize soil particles greater than 2mm in size in the soil bulk
11   density estimates, as such particles can impact the density by as much as 50% for stony soils.
12   The concerns raised by Hodson and colleagues (1997) appear to be valid and should be
13   considered by users of the PROFILE model. However, the authors of the review critiqued an
14   early version of PROFILE (version 3.01) and the most recent version of PROFILE may have
15   already addressed some of these limitations. For example, the abundance of particle sizes greater
16   than 2 mm is included in the current regional model of PROFILE (version 5.0). It should also be
17   noted that Hodson and colleagues (1997) did acknowledge that despite the apparent weaknesses
18   of PROFILE, BCW rates calculated with the model are comparable to those calculated using other
19   methods.
20          In addition to the potential limitations of PROFILE as a model, application of PROFILE
21   to map BCW rates throughout the United States may also present some drawbacks or restrictions.
22   There may be areas  of the United States where input data required by the model is not available.
23   In such situations, it would be necessary to extrapolate data from areas with similar soil, biotic or
24   abiotic conditions. Similarly, if data for specific variables are limited in many areas, it may be
25   necessary to adopt best available default values over large areas, until more data and better
26   coverage across the states is  available.
27   4.3.6  "Field Tests" of model and uncertainty analyses
28          As outlined in the preceding sections, the proposed methodology to map BCW throughout
29   the United States would involve the use of the regional application of PROFILE (version 5.0),
30   continuous coverage data, and in some cases, input and default values from the literature.
31   Therefore, at least a portion of the input data would not be site specific and would be entered as

     March 2010                                 79                   Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B

 1   class values or generated by sub-models or mathematical relationships. Soil water content and
 2   soil mineralogy are examples of such data. It is largely unknown to what degree, if any, this
 3   proposed methodology designed for mapping large areas would influence and potentially distort
 4   the estimates of BCW. Therefore, to validate the weathering estimates from the proposed mapping
 5   methodology, it would be worthwhile to conduct "field tests" of the model output in different
 6   regions of the United States.  Such "field tests" could consist of comparing the regional estimates
 7   of BCW with those determined with the single site version of PROFILE and site-specific data.
 8   (No actual on-the-ground field research required.) In addition, where available, the PROFILE-
 9   generated BCW rate estimates could be compared with weathering rates determined by other
10   methods. Both approaches would provide an indication of the quality and accuracy of estimates
11   from the mapping methodology and regional application of PROFILE.  Sites within the Long-
12   Term Ecological Research (LTER) network would be good locations for the "field tests" due to
13   the large amounts of data available at many of these sites. In addition, at some sites,  such as
14   Hubbard Brook, base cation weathering has been determined using methods other than the
15   PROFILE model. A list of LTER sites within the conterminous 48 states that could potentially
16   serve as "field test" sites is presented in Table 6.0. A sub-set of these sites representing different
17   regions and conditions within the United States should be selected to validate the BCW estimates.
18          In addition to the validating the proposed methodology with "field test" site comparisons,
19   uncertainty analyses should also be conducted on the BCW estimates that are generated with the
20   methodology. There are a total of 26 parameters within the regional application of PROFILE
21   (version 5.0) that require data entry by the user or review prior to applying the model, and each
22   of these parameters could be expected to have a level of uncertainty. Therefore, cumulatively,
23   the uncertainties associated with the BCW estimates could be quite large. In addition, because
24   BCW is one of the most influential terms in the calculation of terrestrial  critical acid loads, and
25   critical loads can be used as a measure of the impact of acidifying nitrogen and sulfur deposition
26   on terrestrial ecosystems, it is important to gain a good understanding of the uncertainty
27   associated with the BCW estimates. Critical acid loads could potentially be used by decision
28   makers to set policy and NOX and SOX emission standards within the United States. Furthermore,
29   uncertainty analyses can reveal which parameters are the most influential in the BCW estimates,
30   thereby guiding which parameters should receive the greatest attention  in the development of the
31   datasets and national coverages for the PROFILE model.

     March 2010                                 80                  Draft - Do Not Quote or Cite

-------
                                                                                    Appendix B

 1          Earlier versions of the PROFILE model (version 3.01) have already been reviewed and
 2   analyzed by researchers based on the application of the model to sites in Norway, Sweden,
 3   Scotland and Wales (Jonsson et al., 1995; Hodson et al., 1996; Zak et al., 1996). Monte Carlo
 4   analyses testing the uncertainty  associated with user defined input variables indicated that
 5   varying input parameter errors individually and simultaneously (within the range of values
 6   reported in the literature) resulted in a variation in model output of+/- 40% (Jonsson et al.,
 7   1995). The authors also determined that bulk density, volumetric water content and exposed
 8   surface area of minerals were the largest source of variation in the output values. The least
 9   sensitive parameters were soil stratification, precipitation and percolation. Similar analyses were
10   conducted by Hodson and colleagues (1996) who determined the influence of single input
11   parameters, one at a time. Based on their analyses, BCW estimates could vary by over 100% using
12   the ranges in parameters values  measured in field studies. The authors also found that some
13   minerals,  such as K-feldspar, were particularly sensitive to variation in input values, and soil
14   temperature,  moisture content and exposed mineral surface area caused the largest amounts of
15   variation in the BCW estimates. These results based on an earlier version of PROFILE suggest
16   that ranges in input values can cause the BCW estimates from the model to vary by moderate to
17   large amounts. However, the level of uncertainty associated with outputs from the most current,
18   regional application of PROFILE (version 5.0) is still unknown. In addition, there has yet to be
19   an assessment of the performance of the model in the United States and a determination of how
20   ranges in  data from different regions in the country would impact the variation in model output.
21   Therefore, uncertainty analyses  should be conducted as a component of the proposed
22   methodology, to provide  bounds to the range of output values associated with the BCW estimates
23   for terrestrial critical acid load calculations in the United States.

24   5.     CONCLUSIONS AND RECOMMENDATIONS
25          The goal of this task was to inform EPA about the tools and data available to develop
26   maximum deposition loads  across the United States for aquatic and terrestrial acidification. In
27   particular, this effort focused on methodologies to estimate Bcw, a parameter that plays a crucial
28   role in predicting an ecosystem's ability to neutralize acid deposition. Based on the findings of
29   this literature review, discussions with experts, evaluation of tools, and assessment of data
     March 2010                                 81                   Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B

 1   availability, two process-based models are recommended: MAGIC for aquatic acidification (with
 2   extrapolation through regional regression modeling) and PROFILE for terrestrial acidification.
 3          It is clear that addressing limitations on soil data availability in the United States will
 4   require considerable effort to populate both models; however, resources invested to satisfy this
 5   data need can be leveraged to the benefit of both terrestrial and aquatic modeling goals. It is also
 6   clear that the MAGIC and PROFILE models' application in the United States has focused on
 7   select regions; however, model developers believe these models can be applied successfully in
 8   other regions, particularly regions with more sensitive ecosystems. Finally, neither MAGIC nor
 9   PROFILE models are readily accessible for public use. Therefore, it will not be practical to
10   assume states and regions could operate the models. Rather, it would be more manageable for the
11   models to be run at the Agency level with states and regional offices providing the needed input
12   data.
13          It is recommended that following EPA review of this report, candidate  regions of the
14   United States be identified for modeling and  levels of effort be estimated to prepare the MAGIC
15   and PROFILE models for operation and to collect and/or predict their input data. As part of the
16   effort, it is recommended that RTI collaborate with recognized experts in the development and
17   application of these two models.

18   6.     REFERENCES
19   Aherne, J. Personal communication. 2009. Communication between Julian Aherne (Trent
20          University, Canada) and Jennifer Phelan (RTI International, USA) by telephone.
21          December 2009.

22   Akselsson, C. Personal  communication.  2009. Communication between Cecilia Akselsson (Lund
23          University, Sweden) and Jennifer Phelan (RTI International, USA) by telephone.
24          December 2009.

25   Akselsson, C., H.U. Sverdrup, and J. Holmqvist. 2006. Estimating weathering  rates of Swedish
26          forest soils in different scales, using the PROFILE model and affiliated databases.
27          Sustainable Forestry in Southern Sweden:  The SUFOR Research Project. Linking Basics
28          and Management, p. 119-131.
     March 2010                                 82                  Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B

 1   Arp, P. Personal communication. 2009. Communication between Paul Arp (University of New
 2          Brunswick, Canada) and Jennifer Phelan (RTI International, USA) during the Canadian
 3          Council of Ministers of the Environment Acid Deposition Critical Loads: Status of
 4          Methods and Indicators Workshop, March 18 and 19, 2009. Ottawa, Ontario, Canada.

 5   Bailey, S.W., S.B. Horsley, R.P. Long, and R.A. Hallett. 2004.  Influence of edaphic factors on
 6          sugar maple nutrition and health on the Allegheny plateau.  Soil Science Society of
 1          America Journal 68:243-252.

 8   Brady, N.C. and R.R.Weil. 2002. The Nature and Properties of Soils:  Thirteenth Edition.
 9          Prentice Hall. Upper Saddle River, New Jersey.

10   Brakke, D. F., A. Henriksen, and S.A. Norton.  1990. A variable F-factor to explain changes in
11          base cation concentrations as a function of strong acid deposition. Verb. Internal.  Verein.
12          Limnol. 24:  146-149.
13   Brakke, D. F., A. Henriksen, and S.A. Norton.  1989. Estimated background concentrations of
14          sulfate in dilute lakes. Water Resources Bulletin 25(2): 247-253.
15   Bricker, O. P., B.F.  Jones, and C.J. Bowser. 2003. Mass-balance Approach to Interpreting
16          Weathering Reactions in Watershed Systems. Treatise on Geochemistry, Volume 5.
17          Editor: James I. Drever. Executive Editors: Heinrich D. Holland and Karl K. Turekian.
18          pp. 605. ISBN 0-08-043751-6. Elsevier, 2003., p.l 19-132
19   Bulger, A.J., B.J. Cosby, C.A. Dolloff, K.N. Eshleman, J.R. Webb, and J.N. Galloway. 1999.
20          SNP:FISH, Shenandoah National Park: Fish in Sensitive Habitats, Volumes 1 through
21          IV. Project final report. Project Completion Report to the National Park Service.
22          Cooperative Agreement CA-4000-2-1007, Supplemental  Agreement #2. University of
23          Virginia, Department of Environmental Sciences, Charlottesville, VA.
24   Chapin, F.S., III., L. Moilanen, and K. Kielland. 1993. Preferential use of organic nitrogen growth by a
25          non-mychorrizal arctic sedge.  Nature 361:150-153.
26   Chen C.W., S.A. Gherini, N.E. Peters, P.S. Murdoch, R.M. Newton, and R.A. Goldstein.  1984.
27          Hydrologic analyses of acidic and alkaline lakes. Water Resources Research 20: 1875-
28          1882.

     March 2010                                 83                   Draft - Do Not Quote or Cite

-------
                                                                                  Appendix B

 1   Chen, L., C. T. Driscoll, S. Gbondo-Tugbawa, M. J. Mitchell, and P. S. Murdoch. 2004. The
 2          application of an integrated biogeochemical model (PnET-BGC) to five forested
 3          watersheds in the Adirondack and Catskill regions of New York. Hydrological Processes
 4          75:2631-2650.
 5   Clair, T.A., Dennis, E.F., Cosby, BJ. 2003. Probable changes in lake chemistry in Canada's
 6          Atlantic Provinces under proposed North American emission reductions. Hydrology and
 7          Earth System Sciences Special Issue 7, 574-583.
 8   Cosby, J. Personal communication. 2009. Communication between Jack Cosby (University of
 9          Virginia, USA) and Jennifer Phelan (RTI International, USA) during the Critical Load
10          Workshop, Fall 2009 National Atmospheric Deposition Program Meeting. October 5 and
11          6, 2009. Saratoga Springs, New York, USA.

12   Cosby, B.J., A. Jenkins, R.C. Ferrier, J.D. Miller, and T.A. B. Walker. 1990 Modeling stream
13          acidification in afforested catchments: Long-term reconstructions at two sites in central
14          Scotland. Journal of Hydrology 120: 143-162.Cosby, B.J., G.M. Hornberger, P.F. Ryan,
15          and D.M. Wolock. 1989a. MAGIC/DDRP final report, vol. 1, Model, calibration, results,
16          uncertainty analysis, QA/QC, internal report, Environ. Res. Lab., U.S. Environ. Prot.
17          Agency, Corvallis, OR.
18   Cosby, B.J., G.M. Hornberger, and R.F. Wright. 1989b. Regional simulation of surface water
19          acidification: Uncertainty due to specification of atmospheric deposition, in Atmospheric
20          Deposition (Proceedings of the Baltimore Symposium, May, 1989), IAHS Publ., 179,
21          153-161.
22   Cosby, B.J., G.M. Hornberger, and R.F. Wright. 1989c. Estimating time delays and extent of
23          regional de-acidification in southern Norway in response to several deposition scenarios,
24          in Regional Acidification Models: Geographic Extent and Time Development,  edited by
25          J. Kamari et al., pp. 151- 166, Springer-Verlag, New York.
26   Cosby, B.J., R.F. Wright, G.M.  Hornberger, and J.N. Galloway.  1985a. Modeling the effects of
27          acid deposition: Assessment of a lumped parameter model of soil water and streamwater
28          chemistry. Water Resources Research 21:51-63.
     March 2010                                84                   Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B

 1   Cosby, B.J., R.F. Wright, G.M. Hornberger, and J.N. Galloway. 1985b. Modeling the effects of
 2          acid deposition: Estimation of long-term water quality responses in a small forested
 3          catchment. Water Resources Research 21:1591-1601.
 4   DeHayes D.H., P.O. Schaberg, GJ. Hawley, and G.R. Strimbeck, 1999.  Acid rain impacts on
 5          calcium nutrition and forest health. Bioscience ¥9(10):789-800.
 6   Dennis, T.E., and AJ. Bulger. 1995. Condition factor and whole-body sodium concentrations in
 7          a freshwater fish: Evidence for acidification stress and possible ionoregulatory
 8          overcompensation. Water, Air, and Soil Pollution 55:377-382.
 9   Driscoll, C.T., G.B. Lawrence, AJ. Bulger, T.J. Butler, C.S. Cronan, C. Eagar, K.F. Lambert,
10          G.E. Likens, J.L. Stoddard, and K.C. Weathers. 2001. Acidic deposition in the
11          northeastern United States: sources and inputs, ecosystem effects, and management
12          strategies. BioScience 57:180-198.
13   Driscoll, C.T., R.M. Newton, C.P. Gubala, J.P. Baker, and S.W. Christensen. 1991. Adirondack
14          mountains. Pp. 133-202 in Acidic Deposition and Aquatic Ecosystems: Regional Case
15          Studies. Edited by D.F. Charles. New York: Springer-Verlag.
16   Drohan, J.R. and W.E.Sharpe, 1997.  Long-term changes in forest soil acidity in Pennsylvania,
17          U.S.A. Water, Air, and Soil Pollution 95:299-311.
18   Drohan, P.J., S.L. Stout, and G.W. Petersen. 2002.  Sugar maple (Acer saccharum Marsh.)
19          decline during 1979-1989 in northern Pennsylvania.  Forest Ecology and Management
20          770:1-17.
21   Duan, L., J. Hao, S. Xie, Z. Zhou, and X. Ye. 2002. Determining weathering rates of soils in
22          China. Geoderma. 110: 205-225.

23   Dupont, J., T.A. Clair, C. Gagnon, D.S. Jeffries, J.S. Kahl, SJ. Nelson, and J.M. Peckenham.
24          2005. Estimation of critical loads of acidity for lakes in northeastern United States and
25          eastern Canada. Environmental Monitoring and Assessment 109: 275-291.
26   Eilers, J.M., G.E. Glass, K.E. Webster, and J.A. Rogalla. 1983. Hydrologic control of lake
27          susceptibility to acidification. Canadian Journal of Fish and Aquatic Sciences 40: 1896-
28          1904.

     March 2010                                 85                  Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B

 1   Fenn, M.E., Huntington, T.G. McLaughlin, S.B., Eager, C., Gomez, A., and Cook, R.B. , 2006. Status of
 2          soil acidification in North America. Journal of Forest Science 52 (special issue):3-\3
 3   Fincher, J., J.R. Cumming, R.G. Alscher, G. Rubin, and L. Weinstein.  1989. Long-term ozone
 4          exposure affects winter hardiness of red spruce (Picea rubens Sarg.) seedlings. New
 5          Phytologist 113:85-96.Fennemaris 1946
 6   Gbondo-Tugbawa, S. S., C.T. Driscoll, MJ. Mitchell, J.D. Aber, and G.E. Likens. 2002. A
 7          model to simulate the response of a northern hardwood forest ecosystem to changes in S
 8          deposition, Ecol. Appl, 12: 8- 23.
 9   Gbondo-Tugbawa, S. S., C. T. Driscoll, J. D. Aber, and G. E. Likens. 2001. Evaluation of an
10          integrated biogeochemical model (PnET-BGC) at a northern hardwood forest ecosystem.
11          Water Resources Research 37:1057-1070.
12   Hartman, M.D., Baron, J.S., Clow, D.W., Creed, IF., Driscoll, C.T., Ewing, H.A., Haines, B.D.,
13          Knoepp, J., Lajtha, K., Ojima, D.S., Parton, W.J., Renfro, J., Robinson, R.B., Van
14          Miegroet, H., Weathers, K.C., and Williams, M.W., 2009, DayCent-Chem simulations of
15          ecological and biogeochemical processes of eight mountain ecosystems in the United
16          States: U.S. Geological Survey Scientific Investigations Report 2009-5150,  174 p.
17   Hartman, M.D., J.S. Baron, and D.S. Ojima. 2007. Application of a coupled ecosystem-chemical
18          equilibrium model, DayCent-Chem, to stream and soil chemistry in a Rocky Mountain
19          watershed. Ecological Mode ling 200: 493-510.
20   Helsel, D.R., and R.M. Hirsch. 1992. Statistical Methods in Water Resources. Studies in
21          Environmental Science. New York: Elsevier Science.
22   Henriksen, A., PJ. Dillon and J. Aherne. 2002. Critical loads of acidity to surface waters in
23          south-central Ontario,  Canada: Regional application of the steady-state water chemistry
24          model. Can. J. Fish. Aquat. Sci. 59 : 1287-1295.
25   Henriksen, A., and M. Posch.  2001. Steady-state models for calculating critical loads of acidity
26          for surface waters. Water, Air, and Soil Pollution: Focus 7:375-398.
27   Henriksen, A., J. Kamari, M. Posch, and A. Wilander.  1992. Critical loads of acidity: Nordic
28          surface waters. Ambio 27:356-363.
     March 2010                                 86                  Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B

 1   Henriksen, A. 1984. Changes in base cation concentrations due to freshwater acidification. Verh.
 2          Int. Verein. Limn. 22: 692-698.
 3   Hodson, M.E., and SJ. Langan. 1999a. The influence of soil age on calculated mineral
 4          weathering rates. Applied Geochemistry 14: 387-394.

 5   Hodson, M.E. and SJ. Langan. 1999b. Considerations of uncertainty in setting critical loads of
 6          acidity of soils: the role of weathering rate determination. Environmental Pollution. 106:
 7          73-81.

 8   Hodson, M.E., SJ. Langan, and MJ. Wilson. 1997. A critical evaluation of the use of the
 9          PROFILE model in calculation mineral weathering rates. Water, Air and Soil Pollution.
10          98: 79-104.

11   Hornberger, G.M., K J. Beven, and P.P. Germann. 1990. Inferences about solute transport in
12          macroporous forest soils from time series models. Geoderma 46: 249-262.
13   Hornberger, G.M., B J. Cosby, and R.F. Wright. 1989. Historical reconstructions and future
14          forecasts of regional surface water acidification in southernmost Norway. Water
15          Resources Research 25: 2009-2018.
16   Horsley, S.G>, R.P. Long, S.W. Bailey, R.A. Hallett, and T.J. Hall .  2000. Facotrs associated with the
17   decline disease of sugar maple on the Allegheny Plateau.  Canadian Journal of Forest Research 30:1365-
18   1378.
19   Jonsson, C., P. Warfvinge, and H. Sverdrup. 1995. Uncertainty in predicting weathering rate and
20          environmental stress factors with the PROFILE model.  Water, Air and Soil Pollution. 81:
21          1-23.

22   Juice, S.M., T J. Fahey, T.G. Siccama, C.T. Driscoll, E.g. Denny, C. Eagar, N.L. Cleavitt, R.
23          Minocha, and A.D. Richardson. 2006. Response of sugar maple to calcium addition to
24          northern hardwood forest.  Ecology 87:1267-1280.

25   Killam, K. 1994, Soil Ecology. Cambridge, UK: Cambridge University Press.

26   Kolka, R.K., D.F. Grigal, and E.A. Nater.  1996. Forest soil mineral weathering rates: use of
27          multiple approaches.  Geoderma. 73: 1-21.
     March 2010                                 87                  Draft - Do Not Quote or Cite

-------
                                                                                  Appendix B

 1   Kretser, W., J. Gallagher, and J. Nicolette. 1989. Adirondack Lakes Study, 1984-1987: An
 2          Evaluation of Fish Communities and Water Chemistry. Ray Brook, NY: Adirondack
 3          Lakes Survey Corporation. Langan, S., Hodson, M., Kennedy, F., Hornung, M.,
 4          Reynolds, B., Hall, J. & Donald, L 2001. The role of weathering rate determinations in
 5          generating uncertainties in the calculation of critical loads of acidity and their
 6          exceedance.  Water, Air and Soil Pollution: Focus 1: 299-312.
 7   Langan, S.J., M. Hodson, D.C Bain, M. Hornung, B. Reynolds, J. Hall, and L. Johnston. 2001.
 8          The role of mineral weathering rate determinations in generating uncertainties in the
 9          calculation of critical loads of acidity and their exceedance. Water, Air and Soil
10          Pollution: Focus 1, 299-312.
11   Langan, S.J., B. Reynolds, and D.C. Bain. 1996. The calculation of base cation release from
12          mineral weathering in soils derived from Paleozoic greywackes and shales in upland UK.
13          Geoderma. 69: 275-285.

14   Langan, S.J., M.E. Hodson,  D.C. Bain,  R.A. Skeffington, and MJ. Wilson. 1995. A preliminary
15          review of weathering rates in relation to their method of calculation for acid sensitive soil
16          parent materials. Water, Air and Soil Pollution. 85: 1075-1081.

17   Li, H. and S.G. McNulty. 2007. Uncertainty analysis on simple mass balance model to calculate
18          critical loads for soil acidity. Environmental Pollution 149:315-326.

19   Lien, L., G.G.  Raddum, and A. Fjellheim. 1992. Critical Loads of Acidity to Freshwater: Fish
20          and Invertebrates. The Environmental Tolerance Levels Programme. Rep. No. 23/1992.
21          Norwegian Ministry of Environment, Oslo, Norway.
22   Lynch, J. Personal communication. 2009. Communication between Jason Lynch (U.S.
23          Environmental Protection Agency, Clean Air Market Division, USA)  and Jennifer Phelan
24          (RTI International, USA) during the Critical Load Workshop, Fall 2009 National
25          Atmospheric Deposition Program Meeting. October 5 and 6, 2009. Saratoga Springs,
26          New York, USA.
     March 2010                                 88                   Draft - Do Not Quote or Cite

-------
                                                                                  Appendix B

 1   Matuszek, I.E., and G.L. Beggs. 1988. Fish species richness in relation to lake area, pH, and
 2          other abiotic factors in Ontario lakes. Canadian Journal of Fisheries and Aquatic
 3          Sciences 45:1931-1941.
 4   McKenzie, NJ. and PJ. Ryan. 1999. Spatial prediction of soil properties using environmental
 5          correlation. Geoderma 89, 67-94.

 6   McNulty, S.G., B.C. Cohen, H. Li, and J.A. Moore-Myers. 2007. Estimates of critical acid loads
 7          and exceedances for forest soils across the conterminous United States. Environmental
 8          Pollution 149:281-292.

 9   Miller, H.G. 1988. Long-term effects of application of nitrogen fertilizers on forest sites. Pp. 07-
10          106 in Forest Site Evaluation and Long-term Productivity. Edited by D.W. Cole and S.P.
11          Gessel. Seattle, WA: University of Washington Press.

12   Miller, E. 2001. Estimating soil weathering rates. Appendix 3 in NEG/ECP Forest Mapping
13          Group (Conference of New England Governors and Eastern Canadian Premiers Forest
14          Mapping Group). 2001. Protocol for Assessment and Mapping of Forest Sensitivity to
15          Atmospheric S and N Deposition: Acid Rain Action Plan - Action Item 4: Forest
16          Mapping Research Project. Prepared by NEG/ECP Forest Mapping Group. Available at
17          http://www.nrs.fs.fed.us/clean_air_water/clean_water/critical_loads/local-
18          resources/docs/NEGECP_F orest_Sensitivity_Protocol_5_21_04.pdf.

19   Nanus, L., M.W. Williams, D.H. Campbell, K.A. Tonnessen, T. Blett, and D.W. Clow. 2009.
20          Assessment of lake sensitivity to acidic deposition in national parks of the Rocky
21          Mountains. Ecological Applications 19(4): 961-973.
22   NEG/ECP Forest Mapping Group (Conference of New England Governors and Eastern
23          Canadian Premiers Forest Mapping Group). 2001. Protocol for Assessment and Mapping
24          of Forest Sensitivity to Atmospheric S and N Deposition: Acid Rain Action Plan - Action
25          Item 4: Forest Mapping Research Project.  Prepared by NEG/ECP Forest Mapping
26          Group. Available at
27          http:www.nrs.fs.fed.us/clean_air_water/clean_water/critical_loads/local-
28          resources/docs/NEGECP_Forest_Sensitivotu_Protocol_5 _21_04.pdf.

     March 2010                                 89                   Draft - Do Not Quote or Cite

-------
                                                                                  Appendix B

 1   Olsson, M., K. Rosen, and P.A. Melderud. 1993. Regional modeling of base cation losses from
 2          Swedish forest soils due to whole-tree harvesting. Applied Geochemistry. Suppl. Issue
 3          No. 2: 189-194.

 4   Ouimet, R. Personal communication. 2009. Communication between Rock Ouiment (Ministere
 5          des Resources naturelles et de la Faune, Canada) and Jennifer Phelan (RTI International,
 6          USA) by email. December 2009.

 7   Ouimet, R. 2008. Using compositional change within soil profiles for modeling base cation
 8          transport and chemical weathering. Geoderma. 145: 410-418.

 9   Ouimet, R., P.A. Arp, S.A. Watmough, J. Aherne, and I. DeMerchant. 2006. Determination and
10          mapping critical loads of acidity and exceedances for upland forest soils in Eastern
11          Canada. Water, Air and Soil Pollution 772:57-66.

12   Ouimet R. and L. Duchesne. 2005. Base  cation mineral weathering and total release rates from
13          soils in three calibrated forest watersheds on the Canadian Boreal Shield. Canadian
14          Journal of Soil Science. 85: 245-260.

15   Pardo, L.H., and N. Duarte. 2007. Assessment of Effects of Acidic Deposition on Forested
16          Ecosystems in Great Smoky Mountains National Park using Critical Loads for Sulfur and
17          Nitrogen. U.S. Department of Agriculture, Forest Service. Prepared for Tennessee Valley
18          Authority, S. Burlington, VT.

19   Pardo, L.H., M. Robin-Abbott, N. Duarte, E.K. Miller. 2004. Tree Chemistry Database (Version
20          1.0). General Technical Report NE-324. U.S. Department of Agriculture, Forest Service,
21          Northeastern Research Station, Newton Square, PA, 45 pp.

22   Posch, M., M. Forsius, and J. Kamari. 1993. Critical loads of sulphur and nitrogen for lakes I:
23          Model description and  estimation of uncertainties. Water, Air, and Soil Pollut.  66, 173-
24          192.
25   Posch, M., J. Kamari, M. Forsius, A.  Henriksen, and A. Wilander. 1997. Exceedance of critical
26          loads for lakes in Finland, Norway and Sweden: Reduction requirements for acidifying
27          nitrogen and sulfur deposition. Environmental Management 21: 291-304.

     March 2010                                 90                   Draft - Do Not Quote or Cite

-------
                                                                                  Appendix B

 1   Rago, P. J., and J.G. Wiener. 1986. Does pH affect fish species richness when lake area is
 2          considered? Transactions of the American Fisheries Society 775:438-447.
 3   Rapp, L. and Bishop, K. 2009. Surface water acidification and critical loads: exploring the F-
 4          factor. Hydrology and Earth System Sciences 13: 2191 -2201
 5   Reynolds, B., and D.A. Norris. 2001. Freshwater critical loads in Wales. Water, Air, and Soil
 6          Pollution: Focus 1:495-505.
 7   Schindler, D.W. 1988. Effects of acid rain on freshwater ecosystems. Science 239:232-239.
 8   Schreck, C.B. 1982. Stress and compensation in teleostean fishes: response to social and physical
 9          factors, pp. 295-321 in Stress and Fish. Edited by A.D. Pickering. New York: Academic
10          Press.
11   Schreck, C.B. 1981. Stress and rearing of salmonids. Aquaculture 28:241-249.
12   Smith, C. Personal communication. 2009. Communication between Chris Smith (U.S.
13          Department of Agriculture Natural Resources Conservation Service, USA) and Jennifer
14          Phelan (RTI International, USA) by telephone. December 2009.

15   St. Clair, S.B., I.E. Carlson, and J.P. Lynch. 2005. Evidence for oxidative stress in sugar maple
16          stands growing on acidic, nutrient imbalanced forest soils. Oecologia 7 ¥5:258-369.

17   Starr, M., AJ. Lindroos, T. Tarvainen, and H. Tanskanen. 1998. Weathering rates in the
18          Hietajarvi integrated monitoring catchment. Boreal Environment Research. 3: 275-285.

19   Stoddard, J., J.S. Kahl, F.A. Deviney, D.R. DeWalle, C.T. Driscoll, A.T. Herlihy, J.H. Kellogg;
20          P.S. Murdoch, J.R. Webb, and K.E. Webster. 2003. Response of Surface Water Chemistry
21          to the Clean Air Act Amendments of 1990. EPA 620/R-03.001. U.S. Environmental
22          Protection Agency, Office of Research and Development, National Health and
23          Environmental Effects Research Laboratory, Research Triangle Park, NC.
24   Sullivan, T.J., BJ. Cosby, J.R. Webb, R.L Dennis, AJ. Bulger, and F.A. Deviney Jr. 2008.
25          Streamwater acid-base chemistry and critical loads of atmospheric sulfur deposition in
26          Shenandoah National Park, Virginia. Environmental Monitoring and Assessment 137:
27          85-99.

     March 2010                                 91                   Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B

 1   Sullivan, T.J., C.T. Driscoll, BJ. Cosby, IJ. Fernandez, A.T. Herlihy, J. Zhai, R. Stemberger,
 2          K.U. Snyder, J.W. Sutherland, S.A. Nierzwicki-Bauer, C.W. Boylen, T.C. McDonnell,
 3          and N.A. Nowicki. 2006. Assessment of the Extent to which Intensively-Studied Lakes are
 4          Representative of the Adirondack Mountain Region. Final report. New York State Energy
 5          Research and Development Authority (NYSERDA), Albany, NY.  Available at
 6          http://nysl.nysed.gOv/uhtbin/cgisirsi/Qcwd6NzFby/NYSL/138650099/8/4298474
 7          (accessed November 1, 2007).
 8   Sullivan, T.J., BJ. Cosby, A.T. Herlihy, C.T. Driscoll, IJ. Fernandez, T.C. McDonnell, C.W.
 9          Boylen, S.A. Nierzwicki-Bauer, and K.U. Snyder. 2007a. Assessment of the Extent to
10          Which Intensively-studied Lakes are Representative of the Adirondack Region and
11          Response to Future Changes in Acidic Deposition. Water, Air, & Soil Pollution 185: 279-
12          291.
13   Sullivan, T J., J.R. Webb, K.U. Snyder, A.T. Herlihy, and B J. Cosby. 2007b. Spatial
14          Distribution of Acid-sensitive and Acid-impacted Streams in Relation to Watershed
15          Features in the Southern Appalachian Mountains. Water, Air, & Soil Pollution 182: 57-
16          71.
17   Sullivan, T.J., BJ. Cosby, K.U. Snyder, A.T. Herlihy, B. Jackson. 2007c.  Model-Based
18          Assessment  of the Effects of Acidic Deposition on Sensitive Watershed Resources in the
19          National Forests of North Carolina, Tennessee, and South Carolina. Report Prepared for
20          USDA Forest Service, Asheville, NC.
21   Sullivan, T. J. Cosby, B J. Herlihy, A.T. Webb, J.R. Bulger, AJ. Snyder, K.U. Brewer, P.F.
22          Gilbert,  E.H. Moore, D.L. 2004. Regional model projections of future effects of sulfur
23          and nitrogen deposition on  streams in the southern Appalachian Mountains. Water
24          Resources Research 40, W02101, doi:10.1029/2003WR001998.
25   Sullivan, T. J., B. J. Cosby, J. R. Webb, K. U. Snyder, A. T. Herlihy, A. J. Bulger, E. H. Gilbert,
26          and D. Moore. 2002. Assessment of effects of acid deposition on aquatic resources in the
27          southern Appalachian Mountains, report, E&S Environ.  Chem., Inc., Corvallis, OR.
28   Sullivan, TJ. 2000.  Aquatic Effects of Acidic Deposition. Lewis Publishers: Washington, D.C.
     March 2010                                 92                  Draft - Do Not Quote or Cite

-------
                                                                                  Appendix B

 1   Sverdrup, H. Personal communication. 2009a. Communication between Harald Sverdrup (Lund
 2          University, Sweden) and Jennifer Phelan (RTI International, USA) during the Canadian
 3          Council of Ministers of the Environment Acid Deposition Critical Loads: Status of
 4          Methods and Indicators Workshop, March 18 and 19, 2009. Ottawa, Ontario, Canada.

 5   Sverdrup, H. Personal communication. 2009b. Communication between Harald Sverdrup (Lund
 6          University, Sweden) and Jennifer Phelan (RTI International, USA) by telephone.
 7          December 2009.

 8   Sverdrup, H. and I. Stjernquist. (editors) 2002. Managing Forest Ecosystems. Developing
 9          Principles and Models for Sustainable Forestry in Sweden. Kluwer Academic Publishers,
10          The Netherlands. 321 pp.

11   Sverdrup, H., W. de Vries, and A. Henriksen. 2001. Mapping Critical Loads. A guidance to the
12          criteria, calculations data collection and mapping of critical loads. Milforapport
13          (Environmental Report) 1990: 14. Nordic Council of Ministers, Copenhagen, NORD:
14          1990: 98, 124 pp.

15   Sverdrup, H., P. Warfvinge, and T. Wickman. 1998. Estimating the weathering rate at Gardsjon
16          using different methods. In: Hultberg, H. and R.  Skeffington editors.  1998. Experimental
17          Reversal of Acid Rain Effects: The Gardsjon Roof Project. John Wiley & Sons Ltd.
18          p. 232-249.

19   Sverdrup, H., W. de Vries, M. Hornung, M.S. Cresser, S.J. Langan, B. Reynolds, R. Skeffington,
20          and W. Robertson.  1995. Modification of the simple mass balance equation for
21          calculating of critical loads of acidity. In: M. Hornung, M. Sutton, and R.B. Wilson
22          (Editors), Mapping and Modeling Critical Loads for Nitrogen. A Workshop Report,
23          October 1994. Published by Institute  of Terrestrial Ecology, Bush Estate, Edinburgh.

24   Sverdrup, H.U. and P. Warfvinge. 1995. Estimating field weathering rates using laboratory
25          kinetics. IN: White, A. and S. Brantley editors. Weathering Kinetics of Silicate Minerals.
26          Volume 8. Reviews in Mineralogy, Mineralogical Society of America, p. 485-541.
     March 2010                                 93                   Draft - Do Not Quote or Cite

-------
                                                                                  Appendix B

 1   Sverdrup, H., De Vries, W., Hornung, M., Cresser, M.S., Langan, S.J., Reynolds, B.,
 2          Skeffmgton, R., and Robertson, W. 1995. Modification of the simple mass balance
 3          equation for calculating of critical loads of acidity.  In: M. Hornung, M. Sutton, and R.B.
 4          Wilson (Editors), Mapping and Modeling Critical Loads for Nitrogen. A Workshop
 5          Report, October 1994. Published by Institute of Terrestrial Ecology, Bush Estate,
 6          Edinburgh.
 7   Sverdrup, H. and P. Warfvinge. 1993a. Calculating field weathering rates using a mechanistic
 8          geochemical model PROFILE. Applied Geochemistry. 8: 273-283.

 9   Sverdrup, H., and P. Warfvinge. 1993b. The effect of soil acidification on the growth of trees,
10          grass and herbs as expressed by the (Ca+ Mg+ K)/Al ratio. Reports in Ecology and
11          Environmental Engineering 2. Lund University, Department of Chemical Engineering,
12          Lund, Sweden.

13   Sverdrup, H., P. Warfvinge, M. Rabenhorst, A. Janicki, R. Morgan, and M. Bowman. 1992.
14          Critical loads and steady-state chemistry for streams in the state of Maryland.
15          Environmental Pollution. 77: 195-203.

16   Sverdrup H.U. 1990. The Kinetics of Base Cation Release Due to Chemical Weathering. Lund
17          University Press, 246pp.

18   Sverdrup, H., W. de Vries, and A. Henriksen. 1990. Mapping Critical Loads. Miljorapport 14.
19          Nordic Council of Ministers, Copenhagen, Denmark.
20   Sverdrup, H. and P. Warfvinge. 1988. Weathering of primary silicate minerals in the natural soil
21          environment in relation to a chemical weathering model. Water, Air and Soil Pollution.
22          38: 387-408.

23   UNECE (United Nations Economic Commission for Europe). 2004. Manual on Methodologies
24          and Criteria for Modeling and Mapping Critical Loads and Levels and Air Pollution
25          Effects, Risks, and Trends. Convention on Long-Range Transboundary Air Pollution,
26          Geneva Switzerland. Available at http://www.icpmapping.org (accessed August 16,
27          2006).

     March 2010                                94                   Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B

 1   U.S. EPA, 2008. Integrated Science Assessment (ISA) for Oxides of Nitrogen and Sulfur-
 2          Ecological Criteria (Final Report) (ISA). U. S. Environmental Protection Agency, Office
 3          of Research and Development, National Center for Environmental Assessment, Research
 4          Triangle Park, NC. EPA/600/R-08/082.
 5   U. S. EPA, 2009. Risk and Exposure Assessment for Review of the Secondary National Ambient
 6          Air Quality Standards for Oxides of Nitrogen and Oxides of Sulfur. Final. U.S.
 7          Environmental Protection Agency, Office of Air Quality Planning and Standards,
 8          Research Triangle Park, NC. EPA-452/R-09-008b.

 9   Van Sickle, 1, J.P. Baker, H.A. Simonin, B.P. Baldigo, W.A. Kretser, and W.E. Sharpe. 1996.
10          Episodic acidification of small streams in the northeastern United States: Fish mortality
11          in field bioassays. Ecological Applications 6:408-421.
12   Velbel, M.A. and J.R. Price. 2007. Solute geochemical mass-balances and mineral weathering
13          rates in small watersheds: Methodology, recent advances, and future directions. Applied
14          Geochemistry 22: 1682-1700.
15   Warfvinge, P. and H. Sverdrup.  1992. Calculating critical loads of acid deposition with
16          PROFILE, a steady-state soil chemistry model. Water, Air and Soil Pollution. 63: 119-
17          143.

18   Watmough S.A., J. Aherne, and P. J. Billion. 2004. Critical Loads Ontario: Relating Exceedance of the
19          Critical Load with Biological Effects at Ontario Forests.  Report 2.  Environmental and Resource
20          Studies, Trent University, ON, Canada.
21   Watmough, S.A., J.  Ahern, and PJ. Dillon. 2005. Effect of declining base cation concentrations
22          on freshwater critical load calculations. Environmental Science & Technology. 39: 3255-
23          3260.
24   Watmough, S., J. Aherne, P. Arp, I. DeMerchant, and R.  Ouimet. 2006. Canadian experiences in
25          development of critical loads for sulphur and nitrogen. Pp. 33-38 in Monitoring Science
26          and Technology Symposium: Unifying Knowledge for Sustainability in the  Western
27          Hemisphere Proceedings RMRS-P-42CD. Edited by C. Aguirre-Bravo, PJ. Pellicane,
28          D.P. Burns, and S. Draggan. U.S. Department of Agriculture, Forest Service, Rocky
29          Mountain Research Station, Fort Collins, CO.
     March 2010                                 95                   Draft - Do Not Quote or  Cite

-------
                                                                                    Appendix B

 1   Webb, J.R., Deviney, F. A., Galloway, J. N., Rinehart, C. A., Thompson, P. A., & Wilson, S.
 2          (1994). The acid-base status of native brook trout streams in the mountains of Virginia; a
 3          regional assessment based on the Virginia trout stream sensitivity study. Charlottesville,
 4          VA: University of Virginia.
 5   Webster K.L., I.F. Creed, N.S. Nicholas, and H.V. Miegroet. 2004. Exploring interactions between
 6          pollutant emissions and climatic variability in growth of red spruce in the Great Smoky
 7          Mountains National Park.  Water, Air, and Soil Pollution 759:225-248.
 8   Wedemeyer, G.A., B.A. Barton, and DJ. MeLeay. 1990. Stress and acclimation, pp. 178-198 in
 9          Methods for Fish Biology. Edited by C.B. Schreck and P.B. Moyle. Bethesda, MD:
10          American Fisheries Society.
11   Whitfield, C.J., S.A. Watmough, J. Aherne, and PJ. Dillon. 2006. A comparison of weathering
12          rates for acid-sensitive catchments in Nova Scotia, Canada and their impact on critical
13          load calculations. Geoderma. 136: 899-911.

14   Yin, X. and P.A. Arp. 1993. Predicting forest soil temperatures from monthly mean air
15          temperature and precipitation records. Canadian Journal of Forest Research, 23: 2521-
16          2536.

17   Zhai, J., C. T. Driscoll, T. J. Sullivan, and B. J. Cosby. 2008.  Regional application of the PnET-
18          BGC model to assess historical acidification of Adirondack lakes. Water Resources
19          Research 44, W01421, doi:10.1029/2006WR005532.
20
     March 2010                                  96                  Draft - Do Not Quote or Cite

-------
                                                                                  Appendix B

 i                                       APPENDIX 1
 2                Potentially Applicable National-Scale Geochemical Data

 3          Currently, there are potentially three consistent national-scale data sets that are most
 4   appropriate for use in this project: the Shacklette data, the more recent National Geochemical
 5   Survey data, and the NRCS pedon soil pit (i.e., LIMS database).

 6   Chemical Analyses of Soils and other Surficial Materials of the Conterminous United
 7   States (Shacklette Data) & the Geochemical Landscapes Project

 8          These data provide an ultra-low-density geochemical baseline for soils and other surficial
 9   materials in the conterminous United States. It is the most widely cited reference for
10   geochemical background data and the data are most appropriately used to provide information on
11   background concentrations of elements in soil for areas represented by small map scales.

12          The data set contains geochemical data from soils and other regolith collected and
13   analyzed by Hans Shacklette and colleagues beginning in 1958 and continuing until about 1976.
14   Originally compiled as a paper record, the data was later included as part of the original USGS
15   PLUTO database. Approximately 1,323 samples were collected through 1976. The 1,323 sample
16   locations that comprise the Shacklette data represent a sampling density of approximately 1
17   sample per 6,000 square kilometers (metadata); equivalent to the collection of samples on a 75-
18   km grid across the country.

19          The sampling protocol called for removal of loose organic debris from the surface and
20   then collection of soil from a depth of 0-20 cm (Smith et al., 2005). Where possible, sample
21   locations were selected where surficial materials had been altered very little from their natural
22   condition as evidenced by the presence of native plants. The sample material at most sites could
23   be termed "soil" because it was a mixture of disintegrated rock and organic matter. Some of the
24   sampled deposits, however, were not soils as defined above, but were other regolith types. These
25   included desert sands, sand dunes, some loess deposits, and beach and alluvial deposits that
26   contained little or no visible organic material.

27          This national-lev el geochemical data set of 1,323 samples has been collected and
28   analyzed according to standardized protocols. This is considered one of the principal strengths of
29   the data set overall. The samples were chemically analyzed by various but compatible techniques
30   in the U.S. Geological Survey laboratories in Denver, CO. Geochemical point-symbol maps were
31   plotted for 40 elemental results and published as USGS. Professional Paper 1270 (Shacklette  and
32   Boerngen, 1984). The original elements analyzed included: Ag,  Al, Ba, Be, B, Ca, Ce, Cr, Co,
33   Cu, Ga, Ge, Hg, Fe, La, Li, Pb, Mg, Mn, Mo, Na, Nd, Ni, Nb, P, K, Rb, S, Sc, Se, Sr, Th, Ti,  U,
34   V, Yb, Y, Zn, Zr, and total carbon. A newer set of national-level interpolated maps displaying
35   the geochemical distribution for 22 elements using the Shacklette data has since been published
36   (Gustavsson, et al, 2001). Using weighted-median and Bootstrap procedures for interpolation and

     March 2010                                97                  Draft - Do Not Quote or Cite

-------
                                                                                  Appendix B

 1   smoothing, full-color maps were produced for seven major elements (Al, Ca, Fe, K, Mg, Na, and
 2   Ti) and 15 trace elements (As, Ba, Cr, Cu, Hg, Li, Mn, Ni, Pb, Se, Sr, V, Y, Zn, and Zr).

 3          The major drawback with the data set is its extremely low numbers of samples for the
 4   entire conterminous United States. However, more recent high-resolution studies (e.g., Smith et
 5   al., 2005) have illustrated that the regional patterns established by the Shacklette data are
 6   generally maintained except where areas have been affected by anthropogenic factors (Smith,
 7   2006).

 8          Efforts are also on-going to build upon the Shacklette data by increasing the density of
 9   the sample locations and producing a high resolution geochemical data set for North America.
10   Also referred to as the Geochemical Landscapes Project, this is a collaborative effort by the
11   USGS, USDA Natural Resource Conservation Service, other federal agencies, and academia to
12   build a national-scale soil geochemical survey that will eventually increase the sample density of
13   the Shacklette data set. The Geochemical Landscapes project began in October 2002 in
14   collaboration with partners in Canada (Geological Survey of Canada; Agriculture and Agri-Food
15   Canada) and Mexico (Consejo de Recursos Minerales/Servicio Geologico de Mexico; Institute
16   Nacional de Estadistica Geografia e Informatica) that has as its long-term goal a soil
17   geochemical survey of North America (Smith et al., 2005). A 3-year pilot project was completed
18   n 2004. During the pilot project soil samples were collected for major- and trace-elements from
19   265 soil samples collected from two continental-scale transects in North America (Smith et al.,
20   2005). The project has just completed a third year of continental-sampling and completed sample
21   collection for approximately 60% of the conterminous United States (D. Smith personal
22   communication, 2009). The state areas that have been completed to date are: ME, NH, VT, CT,
23   RI, MA, NY, MO, AR, MS, LA, NV, UT, CO, WY, KS, NJ, MD, WV, DE, NE, FL, SC, GA,
24   AL, OK, NM, MT, ID, MN, and SD. The USGS anticipates that sampling may be completed for
25   the conterminous US in 2010; or 2011 at the latest. However, funding doesn't allow for analyses
26   to be completed for a number of samples  and several hundred grams of each sample is being
27   archived for on-going and future analysis.

28   National Geochemical Survey (NGS)

29          Efforts are on-going by the USGS to produce a new stream-sediment-based geochemical
30   survey for the United States at a nominal  spacing of 17 by 17 kilometers (i.e., minimum sample
31   density of 1 sample per 289 km2 in all land areas of the country). Project mapping shows that the
32   work is either complete or nearly completed. Unlike other national geochemical data collection
33   efforts, the analytical routines and standards will be consistent throughout the survey. Analytical
34   methods include a 40-element ICP package plus single-element determinations of As, Se, and Hg
35   by atomic absorption for every sample.

36          The project has sought to capitalize on existing datasets and also achieved  samples. For
37   this reason the NGS is based primarily on analyses of stream sediments to build on the massive

     March 2010                                98                   Draft -  Do Not Quote or Cite

-------
                                                                                 Appendix B

 1   achieves of data and samples from DOE's National Uranium Evaluation (NURE) program.
 2   Much of the survey has entailed reanalysis of approximately 35,000 archival samples from the
 3   NURE program. Where NURE samples do not exist, USGS has been working with cooperators
 4   to obtain new samples. The project website reports a total of about 50,000 stream-sediment
 5   samples that have been analyzed for 42 elements, including arsenic, selenium, and mercury. Last
 6   reported during 2004, only about 10,000 more samples needed to be collected and analyzed to
 7   complete the national survey. Samples are generally categorized as follows:

 8       >  Inherited Data: Much of the RASS and PLUTO data were inherited into the NGS;
 9       >  Independent Reanalyses of NURE samples: These sample were reanalyzed by USGS
10          projects other than the NGS. Prior to the NGS, numerous USGS projects reanalyzed
11          samples from the NURE archives. Other USGS projects have continued to reanalyze
12          NURE samples in parallel with the NGS. In the majority of these cases, most or all of the
13          NURE samples in an area were reanalyzed.
14       >  NURE-Systematic. Systematic reanalyses of NURE samples done by the NGS. An
15          archive of stream sediment and soil samples collected by the NURE program is stored at
16          the USGS in Denver, Colo. Rules were established to select a subset of samples for
17          reanalysis that maintains the NGS coverage.
18       >  NURE-Targeted. Targeted reanalyses of NURE samples done by the NGS for various
19          reasons.
20       >  USGS-Re sampling. Reanalyses of USGS archived project samples done by the NGS. The
21          archive includes most of the samples for which there are analytical data in the National
22          Geochemical Database, including those collected by USGS programs.
23       >  Collaborative Sampling with State Programs. Collaborative sampling programs by the
24          USGS and states.

25          Digital data files are presented in 6 categories. In total there are 43 individual data files
26   for the United States. Some of the data has also been processed into vector data to produce maps
27   showing the elemental concentration of As, Se, Hg, Pb, Zn, Cu, Al, Na, Mg, P, Ca, Ti, Mn, and
28   Fe at the county level.

29   USDA NRCS  Soil Pedon Pit Data

30          The USDA NRCS measures soil geochemical characteristics along with performing
31   quantitative and bulk mineralogy tests and other physiochemical measurements for soil series
32   delineated across the United States. This data set and associated detail was discovered through
33   communication with NRCS staff (C. Smith personal communication, 2009; T. Reinsch personal
34   communication, 2009). Most of the geochemical and mineralogy data is associated with
35   individual soil pedons. The NRCS defines a pedon as the smallest unit that can be called a soil. It
36   is a three-dimensional sample that extends from the soil surface to the deepest roots or genetic
37   soil horizons. The area covered by a pedon varies from 10-100 square feet, depending on
38   changes in soil properties. Pits are dug to expose the pedons and the NRCS generally refers to
     March 2010                                99                  Draft - Do Not Quote or Cite

-------
                                                                                    Appendix B

 1   the data associated with the pedons as soil pit data. There are currently approximately 30,000 soil
 2   pits/pedons in the NRCS database.

 3           Groups of pedons with very similar characteristics that are closely associated in the
 4   landscape are called polypedons. Polypedons that have a common set of characteristics that fall
 5   within a particular range are delineated as a basic soil unit referred to as a soil series which have
 6   been identified as the basic unit of the proposed data framework, as previously discussed. The
 7   same soil series delineations can occur in different and distant areas (i.e., across county areas,
 8   states, or regions). A variety of data are used to define a soil (e.g., geomorphic position in the
 9   landscape,  relationship to the water-table, supported flora, geology, number and type of horizons,
10   sediment texture, sediment color variations, etc.), and therefore geochemical and mineralogy
11   data has not been collected from every soil pedon associated with an individual series of the
12   same name since associations can be made based on a number of these other related
13   characteristics. However, geochemical and quantitative mineralogy data has been measured for a
14   significant number of pedons and soil series locations across the country.

15           Since soils of the same series name possess enough similarities to be classified as similar
16   soils it is thought that the geochemistry data can also be extrapolated to pedons of like soil series
17   (C. Smith personal communication, 2009). Assigning mineral phases to the soil series that do not
18   have either geochemical or mineralogy data associated with their pedons will require
19   professional judgment by researchers familiar with the soil pit data and soil taxonomy to make
20   geochemical data extrapolations with a degree  of confidence. In these cases the characteristics of
21   surrounding soils would be used to extrapolate geochemistry or mineralogy, or another data set
22   could be used to aid in the characterization. GIS tools would be used to help automate these
23   determinations where necessary. NRCS staff would aid RTI in determining rules and developing
24   database relationship tables that could be used in automating any extrapolation of this data. The
25   NRCS would also aid RTI in evaluating the reasonableness of the results.

26           Since soil series are delineated across the conterminous United States the pit data could
27   potentially provide a complete geochemical and mineralogy data layer for determining
28   mineralogy. Since mineralogy is already associated with the geochemical data a more accurate
29   assignment of mineral modes may be possible using this data set. Laboratory analysis includes
30   the major geochemical elements: Al, Ca, Fe, K, Mg, Mn, Na, P, Si, Sr, Ti, and Zr. In addition, x-
31   ray diffraction is used to indentify clay mineralogy generally for each horizon of a pedon, and
32   optical  mineralogy analysis is performed on the dominant sand fractions of the soil  from the A-
33   horizon, B-horizon, and C-horizon, or the most dominant horizon. More than 60 fields describing
34   the minerals are listed in the database. The dataset is not uniform in that elemental analyses were
35   routinely done through the 1970's but then these analyses were suspended through the 1980's.
36   Elemental  analyses were resumed during the early 1990's. It is estimated that as much as one
37   third of the 30,000 soil pedons have geochemical data. Likewise, optical mineralogy is not
38   performed  for all pedons and the NRCS staff estimate that approximately as many as one third of
39   the 30,000 soil pits have optical analysis results. Even though the number of pedons with data are

     March 2010                                 100                   Draft - Do Not Quote or Cite

-------
                                                                                  Appendix B


1   similar for geochemical and optical analysis results the data is not necessarily associated with the
2   same set of pedons or even soil series.

3   References

4   Smith, Chris, 2009. Personal communication between Chris Smith, (affiliation) and Scott
5          Guthrie, RTI International, 12/16/09.

6   Reinsch, Thomas, 2009. Personal communication between Thomas Reinsch, (affiliation) and
7          Scott Guthrie, RTI International, 12/18/09.

8

9
    March 2010                                 101                   Draft - Do Not Quote or Cite

-------
                                                                                  Appendix B

 i                                       APPENDIX 2
 2             References for Table 3-2: Applications of the MAGIC Model
 3   Aherne, J, PJ. Dillon, and BJ. Cosby. 2003. Acidification and recovery of aquatic ecosystems in
 4          south central Ontario, Canada: regional application of the MAGIC model. Hydro!. Earth
 5          Syst. Sci 7: 561-573.

 6   Bernett, J.A., J.M. Eilers, and BJ. Cosby. 1997. Overview, Libby Lake Modeling Workshop.
 7          E&S Environmental Chemistry, Inc., Corvallis, OR.

 8   Bulger A.J., BJ. Cosby,  and J.R. Webb. 2000. Current, reconstructed past, and projected future
 9          status of brook trout (Salvelinus fontinalis) streams in Virginia. Canadian Journal of Fish
10          and Aquatic Science 57:1515-1523.

11   Bulger, A.J., C.A. Dolloff, B J. Cosby, K.N. Eshleman, J.R. Webb, and J.N. Galloway. 1995.
12          Sensitivity of Blacknose Dace (Rhinichthys Atratulus) to Moderate Acidification Events
13          in Shenandoah National Park, U.S.A. Water, Air, & Soil Pollution 753(1-4): 125-134.

14   Church, M. R. and J. Van Sickle. 1999. Potential relative future effects of sulfur and nitrogen
15          deposition on lake chemistry in the Adirondack Mountains, United States. Water
16          Resource. Res. 35:2199-2211.

17   Cosby, B J. and TJ. Sullivan. 2001. Quantification of dose-response relationships and critical
18          loads of sulfur and nitrogen for six headwater catchments in Rocky Mountain, Grand
19          Teton, Sequoia, and Mount Rainer national parks. E&S Report 97-15-01.

20   Dennis, IF., T.A. Clair, and B J. Cosby. 2005. Testing the MAGIC acid rain model in highly
21          organic, low-conductivity waters using multiple calibrations. Environmental Modeling
22          and Assessment 70(4): 303-314.

23   Eilers J.M., B J. Cosby, J.A. Bernet, and T.A. Sullivan. 1998. Analysis of the Response of Shasta
24          Lake, Idaho to Increases in Atmospheric Sulfur and Nitrogen Using the MAGIC
25          MODEL. E&S Environmental Chemistry, Inc., Corvallis, OR.
     March 2010                                102                  Draft - Do Not Quote or Cite

-------
                                                                                   Appendix B

 1   Ellis, H., and M. Bowman. 1994. Critical Loads and Development of Acid Rain Control Options.
 2          Journal of Environmental Engineering 120(2): 273-289

 3   Sinha, R., MJ.  Small, P.P. Ryan, T.J. Sullivan, and BJ. Cosby. 1998. Reduced-Form Modelling
 4          of Surface Water and Soil Chemistry for the Tracking and Analysis Framework.  Water,
 5          Air, & Soil Pollution 705(3-4): 617-642.

 6   Sullivan, T.J. and B.J. Cosby. 2004. Aquatic Critical Load Development for the Monongahela
 7          National Forest, West Virginia. Report prepared for the USDA Forest Service
 8          Monongahela National Forest.

 9   Sullivan, T.J. and B.J. Cosby. 2002. Critical Loads of Sulfur Deposition to Protect Streams
10          within Joyce Kilmer and Shining Rock Wilderness Areas from Future Acidification.
11          Report for the USDA Forest Service.

12   Sullivan, T. J., and J.M. Eilers. 1996. Assessment of Deposition Levels of sulfur and Nitrogen
13          Required to Protect Aquatic Resources in Selected Sensitive Regions of North America.
14          E and S  Environmental Chemistry, EPA/600/R-96-123, Corvallis Environmental
15          Research Lab, OR.

16   Sullivan, T.J., B.J. Cosby, K.A. Tonnessen, and D.W. Clow. 2005.  Surface water acidification
17          responses and critical loads of sulfur and nitrogen deposition in Loch Vale watershed,
18          Colorado. Water Resources Research 41: WO 1021.

19   Sullivan, T.J., B.J. Cosby, A.T. Herlihy, J.R. Webb, A.J. Bulger, K.U.  Snyder, P.F. Brewer, E.H.
20          Gilbert,  and D.L. Moore. 2004. Regional model projections of future effects of sulfur and
21          nitrogen deposition on streams in the southern Appalachian Mountains.  Water Resour.
22          Res. 40: W02101.

23   Sullivan, T.J., B.J. Cosby, J.A. Bernert, and J.M. Eilers. 1998. Model Evaluation of
24          dose/response relationships and critical loads for nitrogen and sulfur deposition to the
25          watersheds of lower saddlebag and white dome lakes.

26

     March 2010                                103                  Draft - Do Not Quote or Cite

-------
United States                             Office of Air Quality Planning and Standards             Publication No. EPA-452/P-10-006
Environmental Protection                  Health and Environmental Impacts Division                                  March, 2010
Agency                                          Research Triangle Park, NC

-------