xvEPA
              United States
              Environmental Protection
              Agency
              Air And Radiation
              (6204J)
EPA430-R-95-001a
October 1995
Acid Deposition Standard
Feasibility Study
Report To Congress
                                     PROPERTY Qf
                                        OF

                                     METRJROLOGV

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY


             REPORT TO CONGRESS
          U.S. Environmental Protection Agency
              Office of Air and Radiation
                 Acid Rain Division
                   October 1995

-------
                                    ACKNOWLEDGMENTS
The development of this Acid Deposition Standard
Feasibility Study involved the cooperative efforts
of  individuals  within  the   U.S.  Environmental
Protection Agency (EPA), especially the  Office of
Air and Radiation's Acid  Rain Division  and the
Office of Research  and  Development's Effects
Research  Laboratory  in  Corvallis,  Oregon, and
Atmospheric  Research Exposure and  Assessment
Laboratory in Research Triangle Park, North Caro-
lina. Many individuals at EPA, particularly in the
Office of Air Quality  Planning  and Standards,
contributed useful  comments to this report. EPA
also  acknowledges The Cadmus Group, Inc., for
their  assistance  in  preparing  this  report  under
Contract   Number  68-D2-0168.   Finally,  EPA
appreciates public comments  received and  the
technical review by  the Acid  Deposition  Effects
Subcommittee of the  Ecological Processes and
Effects Committee  of  the EPA  Science Advisory
Board. Both review processes provided important
insight and improved this report.
                                               in

-------
                                        TABLE OF CONTENTS
Acknowledgments	iii

List of Exhibits	vii

List of Acronyms and Abbreviations	xi

Executive Summary	xiii

1  Introduction	1

   What acidic deposition  levels are neces-
   sary to protect sensitive regions'1	2

   What degree of protection is  provided  by
   Title IV?  What  is the residual risk? What
   additional emissions limitations would  be
   required to protect sensitive regions?	4

   How  would an acid deposition standard(s)
   be  implemented?  What  are  the different
   implementation approaches? What are the
   feasibility  and  effectiveness  relative  to
   other  approaches?	4

2  Environmental Goals	7

   2.1    Introduction	7

   2.2    Basic Relationships in Surface Water
         Acidification and Recovery	8
         2.2.1   Relationship  of  Base Cations,
               Sulfur,    and   Nitrogen   in
               Surface Water Acidification	10

         2.2.2  Episodic Acidification	14

         2.2.3  Cumulative Loading Effects	15

         2.2.4  Recovery of  Acidified Ecosys-
               tems 	16

   2.3   Characterizing   Resources  at  Risk
        from Acidic Deposition	17
         2.3.1   Defining Sensitive Resources	1 8

         2.3.2  Identifying Resources at Risk	20

   2.4    Identification  of  Resource  and  Re-
        gional  Priorities	21
         2.4.1   United States	21

         2.4.2  Qualitative   Assessment   of
               Sensitive Aquatic Resources in
               Three  Regions of the United
               States	25

         2.4.3  Qualitative   Assessment   of
               Sensitive Aquatic Resources in
               Canada	27
   2.5  Assessing Protection Needs and  Re-
        source Responses in  the Control of
        Acidic Deposition	28
        2.5.1   Model      Selection      and
               Application	28
        2.5.2  Direct/Delayed  Response Pro-
               ject	
.30
        2.5.3  Nitrogen Bounding Study	36

        2.5.4  Overview of International and
               State Acidic Deposition Crite-
               ria and Standards	50

        2 5.5  Spatial and  Temporal Issues in
               Development of a  Standard	55

   2.6  Controlling  Sulfur  and  Nitrogen  to
        Reduce Surface Water Acidification	56

3  Source-Receptor Relationships and  Depo-
   sition Reductions under Various  Emissions
   Scenarios	59

   3.1  Introduction	59

   3.2  The Regional Acid Deposition Model
        (RADM)	60
        3.2.1   Emissions   and  Atmospheric
               Chemistry	63

        3.2.2  Modeling Source-Receptor Re-
               lationships and Source Attribu-
               tion	65

        3.2.3  Transport,    Chemistry,   and
               Source-Receptor Relationships	67

        3.2.4  Confidence  in Results	69

   3.3  Source Attribution	70
        3.3.1   Changes from 1985 to 2010	70

        3.3.2  Regional  Emissions   Distribu-
               tion  in 2010	71

   3.4  Emissions Reductions Scenarios	72

   3.5  Deposition Reductions under Various
        National  Emissions Reductions  Sce-
        narios	76
        3.5.1   Impact  of   SO2  Allowance
               Trading on Sulfur Deposition	76

        3.5.2  Effect of Additional SO2 Emis-
               sions  Reductions   on  Sulfur
               Deposition	78

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
        3.5.3  Decrease  in  Total  Nitrogen
               Deposition from Decreases  in
               NOX Emissions	79

   3.6  Emissions  Reductions  Strategies  to
        Achieve   Geographically  Targeted
        Sulfur Deposition Loads	82

4  Potential  Benefits  of  an  Acid Deposition
   Standard on Visibility, Human  Health, Ma-
   terial, and Cultural Resources	89

   4.1  Introduction	89

   4.2  Relationship  of Visibility to Acidic
        Deposition	89
        4.2.1  Visibility Impairment	89

        4.2.2  Visibility  Protection Laws and
               Class I Areas	90

        4.2.3  Visibility  Metrics and the Pro-
               jected  Impact of  CAAA  on
               Visibility	90

        4.2.4  Potential   Impact  of  Further
               Sulfur Dioxide Reductions on
               Visibility	93

   4.3  Relationship  of  Human  Health  to
        Acidic Deposition	93

   4.4  Relationship  of  Materials  Damage
        and Cultural   Resources  to  Acidic
        Deposition	95
        4.4.1  Acidic  Deposition Effects on
               Materials and  Structures	95

        4.4.2  Material Life-Cycle and  Dam-
               age Estimates	96

5  Implementation Issues	99

   5.1  Introduction	99
   5.2  Regional Targeted Approach	100
        5.2.1  Description    of    Regional
               Targeted Approach	100

        5.2.2  Integration with Title IV	100

        5.2.3  Impediments  to  Implementa-
               tion .
.100
             5.3   National Emissions-Based Approach	101
                  5.3.1  Description    of    National
                        Emissions-Based Approach	101

                  5.3.2  Integration with Title IV	101
                  5.3.3  Impediments  to Implementa-
                        tion	
                                              .101
             5.4  Economic Impacts	101
                  5.4.1  2010  CAAA  Scenario  (with
                        Trading)	102

                  5.4.2  50 Percent Utility SO2 Reduc-
                        tion Scenario	102

                  5.4.3  50 Percent Utility and Indus-
                        trial SO2 Reduction Scenario	105

                  5.4.4  Geographically Targeted  Re-
                        duction Scenario	105

                  5.4.5  NOX  Reductions—50 Percent
                        Utility and Industrial	105
                  5.4.6 Summary  of  Economic  Im-
                        pacts	
                                              .107
   5.5  Monitoring Program Effectiveness	108

   5.6  Conclusions	109

6  Integration and Conclusions	111

   6.1  Introduction	111

   6.2  Determining Environmental Goals	111

   6.3  Projected    Environmental    Conse-
        quences of Acidic Deposition Reduc-
        tion Scenarios	114

   6.4  Selecting Deposition Goals	116
   6.5  Feasibility of Establishing and Imple-
        menting an Acid Deposition Standard	119

Appendices
   A  Summary of Selected NAPAP Reports	A-1
   B  Plots  from  EPA's  Nitrogen  Bounding
      Study	B-1
   C  Range  of  Influence of Emissions  from
      RADM Tagged Subregions	C-1

   D  Summary  of Science Advisory  Board
      Review and  Public  Comments   and
      Responses	D-1
                                                     VI

-------
                                       LIST OF EXHIBITS
 1. Empirically  determined   relationship
    between ANC and pH for three sensi-
    tive  regions, including cross-regional
    mean and standard deviation	
 2. Principal watershed and surface water
    characteristics that influence  resource
    sensitivity to acidification	19

 3. Critical  pH  for selected taxa in  lakes
    and streams	23

 4. MAGIC	31

 5. Study  regions  included  in  the  Di-
    rect/Delayed Response Project and the
    Nitrogen Bounding Study	33

 6. Target populations  included  in  the
    NSWS, DDRP, and NBS Studies	34

 7. NBS model  projections  for year 2040
    percentage of target  population Adi-
    rondack lakes with ANC<0 u.eq/1	39
 8. NBS model  projections for year  2040
    percentage   of   target    population
    mid-Appalachian  streams  with ANC<
    50u,eq/l	
.40
 9. NBS model  projections for year  2040
    percentage of target population South-
    ern Blue Ridge streams with ANC<50
    u.eq/1	41

10. Interpreting NBS plots	42

11. Time to watershed nitrogen saturation	43

12. Summary  of NBS  results: Range  of
    minimum  (background deposition)  to
    maximum (implementation of CAAA)
    percentages of acidic and sensitive tar-
    get waters	45
13.  Surface water responsiveness to reduc-
    tions in deposition beyond the CAAA:
    Detectible improvements in  long-term
    ANC by 2040	
.47
14.  Impact of CAAA  on sensitive  surface
    waters: NBS model projections for Year
    2040	
.51
15.  LRTAP	53

16.  Physical and chemical  processes con-
    tributing to acidic deposition	61

17.  The RADM modeling domain	62
 18. Map of annual sulfur emissions density
     in 1985 (tons/year)	65

 19. Tagged RADM subregions	66

 20. Proportion of annual sulfur deposition
     contributed  by RADM  Subregion  15
     (OH/WV/PA border region)	67

 21. Percentage cumulative range of  influ-
     ence   of   RADM   Subregion   15
     (OH/WV/PA border region)	67

22a. Source-receptor  relationships  in the
     Northeast: Cumulative percent  sulfur
     deposition	68

22b. Source-receptor  relationships  in the
     lower Ohio Valley: Cumulative percent
     sulfur deposition	68

22c. Source-receptor  relationships  in the
     Southeast: Cumulative percent  sulfur
     deposition	68

 23. Percent  contribution  to  sulfur  emis-
     sions of 53 tagged RADM regions	70

 24. Percent  reduction in  tagged  regions
     from  1985  to 2010 as  a  function of
     relative contribution of each region to
     all tagged emissions	70

 25. Percentage   of  tagged emissions  by
     tagged regions for 1985 and 2010	71

 26. Contribution  of  top-10  SO2  emitting
     regions to sulfur deposition in sensitive
     regions	71

 27. Comparison  of proximate and major
     emitting regions to sulfur deposition in
     sensitive areas in 2010	72

 28. Estimated U.S. SO2 emissions with and
     without Title IV from 1980 to 2015	73

 29. Predicted SO2  utility  emissions  from
     1990 to 2010	75

 30. SO2 emissions in the U.S. RADM do-
     main (eastern United States)	76

 31. Annual  average  RADM  total  sulfur
     deposition (kg-S/ha): 1980	76

 32. Annual average RADM-predicted total
     sulfur deposition  (kg-S/ha): Post-2010
     full CAAA implementation	77

 33. Percentage reductions  in sulfur deposi-
     tion from CAAA implementation	77
                                                VII

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
 34. Impact of trading on  sulfur deposition
     in sensitive regions	
 35.
                                             77
    Annual  average  RADM  total  sulfur
    deposition   (kg-S/ha)   by   geographic
    area in 2010:  Ranges of differences in
    deposition   between   post-2010  full
    implementation   and   no    trading
    scenarios	

36. RADM-predicted  annual average total
    sulfur deposition (kg-S/ha) in 2010 un-
    der additional utility SO2 emissions re-
    duction scenario	
 37. RADM-predicted annual average total
     sulfur deposition (kg-S/ha) in 2010 un-
     der additional utility and industrial SO2
     emissions reduction scenario	
 38. Percentage reductions in sulfur deposi-
     tion from post-2010 full  implementa-
     tion under  additional  utility  SO2  re-
     duction scenario	
 39. Percentage reductions in sulfur deposi-
     tion from post-2010 full  implementa-
     tion under additional utility and  indus-
     trial SO2 reduction scenario	
 40. Sulfur  deposition to sensitive  regions
     under various emissions scenarios	
 41. Percent reductions in sulfur deposition
     to sensitive regions  from 1980 levels
     under various emissions scenarios	
 42. Estimated U.S. NOX emissions with and
     without Title IV from 1980 to 2010.
 43. Percent contribution of utility sources
     to nitrogen deposition in 1990	
 44. Percent   contribution  of   industrial
     sources to nitrogen deposition in 1990 ..

 45. Percent contribution of mobile sources
     to nitrogen deposition in 1990	
 46. Annual average RADM total  nitrogen
     deposition (kg-N/ha) in 1990	
 47. RADM-predicted annual average total
     nitrogen  deposition  (kg-N/ha)  under
     utility  and  industrial  NOX emissions
     reductions scenario	
 48. Percentage   reductions   in   nitrogen
     deposition under utility and  industrial
      NOX emissions reductions scenario.
 49.  Nitrogen  deposition  to sensitive  re-
      gions under  base case and additional
      utility  and  industrial  NOX emissions
      reduction scenario	
.78




.80




.80




.80




.80


.81



.81


.81


.82


.82


.82


.83




.83



.83




.83
          50. Example:  Selection  of  maintenance
              loads	85

         51 a. Geographically targeted additional util-
              ity SO2 reduction in contiguous RADM
              subregions	86
                                                      54.
                                                      51 b. Geographically targeted additional util-
                                                           ity SO2  reduction  in  major  RADM
                                                           subregions contributing to deposition
                                                           (not contiguous)	
                                             .86
 52. Map  of extent of  contiguous  geo-
     graphic regions for achieving targeted
     deposition  loads  equivalent  to addi-
     tional  nationwide  utility  SO2  reduc-
     tions  	87

53a. Geographically targeted additional util-
     ity and industrial SO2 reduction  in con-
     tiguous RADM subregions	88
                                                     53b. Geographically targeted additional util-
                                                          ity and industrial SO2 reduction in ma-
                                                          jor RADM subregions contributing to
                                                          deposition (not contiguous)	
                                             .88
     Extent  of contiguous  geographic re-
     gions for achieving targeted deposition
     loads equivalent to additional nation-
     wide utility  and industrial SO2 reduc-
     tions 	
                                                      55. Geographically   targeted   reductions
                                                          with a maintenance load of 5 kg-S/ha
                                                          in major RADM  subregions contribut-
                                                          ing to deposition (not contiguous)	
                                                      56. Anthropogenic contributions to visibil-
                                                          ity impairment	,
                                                      57. Annual average visual range (km) pro-
                                                          jected  for  2010  without  Title   IV:
                                                          50th-percentile visibility	
                                                      58. Annual average visual range (km) pro-
                                                          jected for 2010 with Title IV, including
                                                          trading: 50th-percentile visibility	
                                                      59. Annual average improvement in 50th-
                                                          percentile visibility (dv) from  1980 to
                                                          2010 with Title IV, including trading	

                                                      60. Specific provisions of Titles I,  II, and IV..

                                                      61. Average annual visual range  estimates
                                                          for representative Class I  areas in the
                                                          Southwest	
                                                      62.  Percent increase in visual  range from
                                                           1985 to 2010 with full CAAA  imple-
                                                           mentation 	
                                                      63.  Percent increase in visual range from
                                                           1985 to 2010 with additional SO2 re-
                                                           duction beyond CAAA	
.88




.88


.89



.91



.91



.91

.92



.93



.93



,.94
                                                  VIM

-------
                                                                                        LIST OF EXHIBITS
 64. Relationship of acidic deposition proc-
     esses to health effects	
 65. Percentage of metal corrosion attrib-
     uted to atmospheric factors	
 66. 2010 annual  costs and SO2 emissions
     by Census region: CAAA scenario	

 67. 2010 annual  costs and SO2 emissions
     by Census region: CAAA scenario ver-
     sus additional 50 percent utility emis-
     sions reduction scenario	
  ...94


  ...96


  .103




  .104
 68. 2010 annual  costs and SO2 emissions
     by Census region: CAAA scenario ver-
     sus additional 50  percent  utility and
     industrial emissions reduction scenario .

69a. Annual  costs of  geographically tar-
     geted reductions equivalent to nation-
....106
     wide   50%  utility   SO2   reduction
     (contiguous RADM subregions)	107

69b. Annual  costs  of  geographically tar-
     geted reductions equivalent to nation-
     wide 50% utility SO2 reduction: Major
     RADM  subregions  contributing   to
     deposition (not contiguous)	107
 70. Summary of costs of various emissions
     reductions scenarios	108

 71. Year 2040 NBS projections for Adiron-
     dack lakes	115

 72. Year 2040 NBS projections  for mid-
     Appalachian streams	115

 73. Year 2040  NBS projections for  South-
     ern Blue Ridge Province streams	116
                                                  IX

-------
                                      LIST OF ACRONYMS
ADIR      Adirondack Mountain region
ADOM     Acid Deposition and Oxidants Model
ALim       inorganic monomeric aluminum
ANC       acid neutralizing capacity
AQRV     Air Quality Related Values
BACT      best available control technology
BEA       Bureau of Economic Affairs
Btu        British thermal unit
Ca2+       calcium
CAA       Clean Air Act
CAAA     Clean Air Act Amendments
CCE       Coordination Center for Effects
CCT       Clean Coal Technology
CEUM     Coal and Electric Utilities Model
Ch         chlorine
cm         centimeter
CO        carbon monoxide
DDRP     Direct Delayed Response Project
DOE       Department of Energy
DOI       Department of the Interior
dv         deciview
EMEFS     Eulerian Model Evaluation Field Study
EPA        U.S. Environmental Protection Agency
FIP         Federal Implementation Plan
H+         hydrogen ion
H2O2      hydrogen peroxide
ha         hectare
H2SO4     sulfuric acid
HCHO     formaldehyde
HNO3     nitric acid
IMPROVE  Interagency  Monitoring  of  Protected
           Visual Environments
kg         kilogram
km         kilometer
LNB       low-NOx burner
LRTAP     Long-Range Transboundary Air Pollution
LTM       Long-Term Monitoring
m          meter
M-APP     mid-Appalachian region
mi
MM4
N
NAAQS
NADB
NAPAP

MAS
NBS
NH3
NH4+
NO
NO2
NO3-
NOAA

NOX
NRC
NSPS
NSS
NSWS
NURF
NYSDEC

03
PAN
PM10

P043+
ppm
PSD
RADM
RIA
S
SBRP
SCR
SIP
SNCR
SO2
magnesium
mile
Mesoscale Model
nitrogen
National Ambient Air Quality Standard
National Allowance Data Base
National Acid  Precipitation  Assessment
Program
National Academy of Sciences
Nitrogen Bounding Study
ammonia
ammonium ion
nitric oxide
nitrogen dioxide
nitrate
National   Oceanic  and  Atmospheric
Administration
nitrogen oxide
National Research Council
New Source Performance Standard
National Stream Survey
National Surface Water Survey
National Unit Reference File
New   York   State    Department   of
Environmental Conservation
ozone
peroxyacetyl nitrate
particulate  matter   smaller  than  10
micrometers
phosphate
parts per million
prevention of significant deterioration
Regional Acid Deposition Model
Regulatory Impact Analysis
sulfur
Southern Blue Ridge Province region
selective catalytic reduction
State Implementation Plan
selective non-catalytic reduction
sulfur dioxide
                                                XI

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
SO42'      sulfate                                   jieq/l       microequivalents per liter
SOMA     Sulfur Oxide Management Area             u.g/1        micrograms per liter
SOS/T     state-of-science/technology                jim        micrometer
SOX        sulfur oxide                              4DDA     four-dimensional data assimilation
UNECE     United  Nations  Economic Commission
           for Europe
yr         year
                                                 XII

-------
                                    EXECUTIVE SUMMARY
Scientific  evidence has  shown  that  atmospheric
deposition of sulfur and  nitrogen compounds can
harm ecosystems.   Title  IV of the  Clean Air Act
Amendments of 1990 (CAAA or the Act) addresses
the problem of such  effects by mandating reduc-
tions in emissions of sulfur and  nitrogen oxides,
the major precursors  of  acidic deposition.  Cou-
pled with  Titles I and II of the Act,  which address
new and existing stationary and mobile sources of
sulfur and nitrogen oxides, implementation of Title
IV is expected to provide significant benefits to the
United States and Canada. These benefits include
decreases in the acidity of lakes and streams, con-
comitant improvements in fish population diversity
and  health, decreases in  soil degradation and for-
est stress,  improvements  in visibility (especially to
scenic vistas), decreases  in damage  to  materials
and cultural resources, and a reduction in human
health effects.  Congress included Section 404 in
Title IV (Appendix B of the Act) requiring the Envi-
ronmental Protection Agency (EPA or  the Agency)
to provide a report to Congress  on the feasibility
and effectiveness of an acid deposition standard or
standards  to protect sensitive and critically sensi-
tive aquatic and terrestrial resources.  Specifically,
Congress listed six areas to be addressed in the re-
port:

  *  Identification of sensitive and critically sen-
     sitive aquatic  and terrestrial resources in
     the  U.S.  and  Canada  which  may  be af-
     fected by the  deposition of acidic  com-
     pounds;

  *  Description and specification of a numeric
     value for an acid deposition standard suffi-
     cient to protect such resources;

  *  Description of the use of such standard or
     standards in other Nations or by any of the
     several States in  acidic  deposition control
     programs;

  *  Description of measures that would be
     needed to integrate  such standard or stan-
     dards with the control program required by
     Title  IV of the Clean Air Act;

  *  Description of the state of knowledge with
     respect  to source-receptor  relationships
     necessary to develop a  control program on
     such  standard or standards  and additional
     research  that is on-going  or would  be
     needed to  make such a  control program
     feasible;

   * Description of impediments to implemen-
     tation  of such  control program and the
     cost-effectiveness of deposition standards
     compared  to  other  control strategies  in-
     cluding ambient air quality standards, new
     source performance  standards and the  re-
     quirements of Title IV of the Clean Air Act.

This report fulfills the requirement of Section 404
by integrating state-of-the-art ecological effects  re-
search,  emissions and  source-receptor modeling
work, and evaluation of implementation and cost
issues to address the six areas and other issues  re-
lated to the feasibility of establishing and imple-
menting an acid deposition  standard or standards.
Congress   also   requires   the  National   Acid
Precipitation  Assessment  Program  (NAPAP)  to
conduct a study similar to the technical portions of
this  report,  on the reduction in  deposition  rates
needed  to  prevent  adverse  ecological  effects.
NAPAP is required to  submit its report to Congress
in 1996   (Section 901 [j]  of the  Clean   Air Act
Amendments).

DEVELOPING A STANDARD TO PROTECT
SENSITIVE RESOURCES
An acid deposition standard is a level of deposi-
tion  (most likely in units of  kilograms of pollutant
per  hectare per  year) that provides a predeter-
mined level of protection to specific ecological  re-
sources.  The  natural  resources most at risk  from
acidic deposition and those most amenable  to a
quantitative  assessment  are  aquatic  systems.
Therefore, a standard designed to protect against
the ecological effects of acidic deposition would
most likely be  developed  based  on effects  to
aquatic systems. Other  ecological resources  such
as high elevation red  spruce forests in the eastern
United States and Canada may  also be at risk, but
less  is  known about the effects process,  and the
rate  and extent of impacts on those resources. Re-
search  conducted  under the auspices of the Na-
tional  Acid  Precipitation  Assessment  Program
(NAPAP)  concluded that regions in the United
States most at risk from continued acidic deposi-
tion  are located  along the Appalachian Mountain
chain stretching from  the Adirondacks  to the
                                               XIII

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
Southern Blue Ridge. Although many surface wa-
ters in western North America are as sensitive as,
or more sensitive than, aquatic systems in the East,
deposition  levels in  the West are sufficiently low
that the risk of chronic (long-term) acidification to
resources in most of the West is low at present and
is  expected to  remain low in the foreseeable fu-
ture.   Episodic acidification (from  spring snow-
melts) adversely affects some eastern surface wa-
ters.   It may be affecting  high-elevation  western
surface waters, as well.

An acid deposition standard or standards could be
designed to achieve a variety of  environmental
protection goals. For example, the goal of a stan-
dard may be to (a) maintain specific conditions as
observed at a particular point in time (e.g., condi-
tions observed in 1984 during the National Surface
Water Survey);  (b) protect all  systems from any
harmful  anthropogenic  effects  (i.e.,  return  to
pre-industrial  conditions);  or (c) balance effects,
costs, and other societal values. A standard can be
designed to address  chronic or episodic acidifica-
tion  and could vary by region based on the re-
gional variability  of  ecological  sensitivity.  No
guidance is provided in the statutory language re-
garding the degree of protection desired by a stan-
dard or standards.

Target populations  of Adirondack  lakes,  Mid-
Appalachian streams,  and Southern Blue Ridge
streams were selected for detailed analysis in this
study because  they  represent areas that receive
fairly high levels of acidic deposition, are sensitive
to acidic deposition, have the best historical data,
and  have been extensively studied  by scientists.
Potential future impacts of acidic deposition  are
estimated by modeling the  response ol a target
population  of  aquatic systems (lakes and streams)
in these areas to various levels of deposition.  Tar-
get populations  are selected to represent sensitive
surface waters  over broad geographic regions. In-
dividual  target  populations  of  sensitive surface
waters used for various acidic deposition studies
have  become  progressively smaller  over time  as
investigators have refined their work  to study more
intensively the acidification processes  influencing
the most sensitive surface waters.  For example, of
the total  population of lakes in the eastern United
States, 18,156 of these lakes  potentially most sen-
sitive to acidification  were  included  within  the
target  population of the National Surface Water
Survey (NSWS,  1984).  In turn, further refinement
of sensitivity characteristics for the NSWS lakes led
to targeting a total  population of 3,227 lakes in the
Northeast  during  the  Direct/Delayed Response
Project (DDRP, 1988)  and 703 lakes in the Adi-
rondack Mountains during the Nitrogen Bounding
Study (NBS, 1994).  Similar refinements also oc-
curred for lakes and stream reaches  in other re-
gions.  This approach  is illustrated diagrammati-
cally  in Exhibit I  (not to scale).   In this  report,
analyses of risks from acidic deposition to sensitive
lakes  and  streams focuses extensively on extrapo-
lations regarding the highly sensitive, but limited,
target populations used during the NBS studies.


  EXHIBIT I. TARGET SURFACE WATER POPULATIONS
    INVESTIGATED BY NSWS, DDRP, AND NBS
SELECTING AN ACID DEPOSITION STANDARD OR
STANDARDS
The analyses presented in this study provide model
projections of direction and magnitude responses
for modeled target watersheds under various sulfur
and nitrogen  deposition scenarios.  Based  on the
remaining scientific  uncertainties,  particularly re-
garding the effects of nitrogen and the rate of those
effects on the watershed, selection of an appropri-
ate  level or levels for a standard  to achieve any
particular environmental  goal   is very  difficult.
Therefore, this study does not recommend an acid
deposition standard or standards at this time. The
modeling analyses in this report however, do indi-
cate important watershed responses  to emissions
reductions in the Clean Air Act and provide esti-
mates  of deposition  reductions  that  would be
needed to achieve a range  of environmental goals
within the uncertainty of the modeling results.

IMPACT OF THE CAAA ON  SENSITIVE SURFACE
WATERS
Model projections  from the Nitrogen  Bounding
Study (NBS) indicate that sulfur deposition reduc-
tions  mandated  by Title IV of  the Clean Air Act
Amendments would benefit sensitive  surface wa-
ters by the year 2040. Exhibits II-IV show ranges
                                                XIV

-------
                                                                                  EXECUTIVE SUMMARY
   EXHIBIT II. PERCENT OF NBS TARGETED ADIRONDACK LAKES PROJECTED
        TO BE CHRONICALLY ACIDIC (ANC<0 (IEQ/L) IN 2040
            PERCENT OF TARGET POPULATION ADIRONDACK
              LAKES OBSERVED ACIDIC IN 1 984 = 1 9%
                TARGET POPULATION = 700 LAKES
Time to Watershed
Nitrogen Saturation
Never
250 years
1 00 years
50 years
Percent of Targeted Waters
Projected to be Acidic in 2040
Without CAAA
25%
23%
36%
50%
With CAAA
11%
15%
26%
43%
    EXHIBIT III. PERCENT OF NBS TARGETED MID-APPALACHIAN STREAMS
    PROJECTED TO BE CHRONICALLY ACIDIC (ANC<0 I1EQ/L) IN 2040
          PERCENT OF TARGET POPULATION MID-APPALACHIAN
              STREAMS OBSERVED ACIDIC IN 1 985 = 4%
               TARGET POPULATION = 4,300 STREAMS
Time to Watershed
Nitrogen Saturation
Never
250 years
1 00 years
50 years
Percent of Targeted Waters
Projected to be Acidic in 2040
Without CAAA
8%
21%
23%
33%
With CAAA
0%
4%
5%
9%
 EXHIBIT IV. PERCENT OF NBS TARGETED SOUTHERN BLUE RIDGE PROVINCE
 STREAMS PROJECTED TO BE CHRONICALLY ACIDIC (ANC<0 LIEQ/L) IN 2040
        PERCENT OF TARGET POPULATION SOUTHERN BLUE RIDGE
         PROVINCE STREAMS OBSERVED ACIDIC IN 1985 = 0%
              TARGET POPULATION = 1,300 STREAMS
Time to Watershed
Nitrogen Saturation
Never
250 years
1 00 years
50 years
Percent of Targeted Waters
Projected to be Acidic in 2040
Without CAAA
0%
1%
2%
13%
With CAAA
0%
0%
0%
4%
of percent target population  lakes or streams in
each sensitive region projected to be chronically
acidic (acid neutralizing capacity [ANC] < 0 Lieq/l)
by 2040 with and without implementation  of the
Clean Air  Act.  The scenarios are described ac-
cording to the extent and rate of nitrogen deposi-
tion  effects on   watersheds (e.g.,  percent  target
waters in Adirondacks projected to be  acidic in
2040 with  implementation of  the  CAAA  range
from 11  to 43 percent).  Nitrogen saturation may
be defined as the condition reached at which the
          supply of nitrogenous compounds to a
          watershed   exceeds  the   ability  of
          biogeochemical  processes  within  the
          watershed to retain those compounds on
          a net basis.  That is, the point at which
          the supply of nitrogen exceeds demand.
          When this capacity is reached,  nitrogen
          losses   from    watersheds   increase,
          principally  in   the  form  of  nitrate
          leaching.     The   time  to   nitrogen
          saturation varies  among  and  within
          regions due  to  differences in  tempera-
          ture, moisture, length of  growing season,
          soil  fertility,  forest  age,  and   historic
          nitrogen deposition.  There is significant
          uncertainty regarding times to watershed
          nitrogen  saturation  in  each  sensitive
          region.  As reflected in  the exhibits, in
          each modeled region, the proportion of
          targeted  acidic  surface  waters would
          have been higher,  in some cases  dis-
          tinctly, without  the sulfur dioxide (SO2)
          reductions in the CAAA.

          ENVIRONMENTAL ANALYSIS OF
          EMISSIONS TRADING
          Projections of sulfur deposition in 2010,
          when  the SO2  emissions trading pro-
          gram is fully implemented and  utilized,
          is part of the analysis of the impact of
          the CAAA on sensitive  regions.  Atmos-
          pheric modeling projects no more than
          a 10 percent difference in sulfur deposi-
          tion at any location in  2010 with  and
          without trading.  Over most of the east-
          ern  United  States,  the difference  in
          deposition is less  than  5 percent,  and
          there is no difference in eastern  Canada.
          Exhibit V is a map that  shows the pro-
          jected annual average difference in sul-
          fur deposition between  trading and  no
          trading over the eastern United States
          and  Canada.  Differences in deposition
          of less than 10 percent are projected not
          to measurably change ecological effects.
          Furthermore,  a  recently released Gen-
eral Accounting Office report estimated  that the
allowance trading  program  will  reduce  control
costs by over 40 percent  and up to 70 percent if
the trading program is  used to the fullest extent.*
Therefore, while the allowance trading program is
expected  to reduce costs  of  control, it  is not
* U.S. General Accounting Office. December 1994. Air
  Pollution Allowance Trading  Offers an Opportunity
  to Reduce Emissions at Less Cost. Washington, DC.
                                                xv

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
    EXHIBIT V. ANNUAL AVERAGE RADM TOTAL SULFUR DEPOSITION (KG-S/HA) BY GEOGRAPHIC AREA IN 2010:
  RANGES OF DIFFERENCES IN DEPOSITION BETWEEN IMPLEMENTATION OF THE CAAA WITH AND WITHOUT TRADING
                                                                -0.80 TO -1.16

                                                             H -0.50 TO -0.80

                                                             H -0.20 TO -0.50

                                                             CH -0.20 TO 0.20

                                                             0 0.20  TO  0.50

                                                             ^ 0.50  TO  0.90

                                                             • 0.90  TO  1.53
projected to have a measurable negative environ-
mental impact.

IMPLICATIONS FOR FURTHER DEPOSITION
REDUCTIONS
Scientific analysis indicates that nitrogen as well as
sulfur  deposition are  important  contributors  to
chronic and episodic acidification of surface waters.
Further reductions in nitrogen as  well  as  sulfur
deposition may  be  necessary in order to realize
protection of target sensitive systems.  Model pro-
jections indicate that, if the time to nitrogen satu-
ration  in the  Adirondacks  is  100 years or less,
maintaining  the  proportion of chronically acidic
target  surface waters in  the Adirondacks  in  the
year 2040 near proportions observed in 1984 may
require reducing anthropogenic sulfur and nitrogen
deposition by  40 to 50 percent or more  below
levels  achieved  by  the  CAAA.   In  the mid-
Appalachians,   implementation   of  the   CAAA
should maintain 1985  proportions of chronically
acidic target streams in the year 2040 if the time to
nitrogen saturation is 250 years or longer; more
rapid  nitrogen saturation  (in the  range of 100
years)  may require reductions  in  anthropogenic
sulfur and nitrogen deposition by 25 percent be-
low levels achieved by the CAAA.  With  imple-
mentation of the CAAA,  no chronically  acidic
streams are  expected within the Southern  Blue
Ridge  target population  in  the year  2040.   In
                                               XVI

-------
                                                                                EXECUTIVE SUMMARY
addition to acidity (ANC<0 u.eq/1), another useful
measure, particularly for streams at risk of episodic
acidification, is the sensitivity of a lake or stream
to becoming acidic (i.e., very low ANC such as 50
u.eq/1).  Modeling projections using  this measure
(as described in Chapter 2) also  indicate  that
further deposition reductions may be necessary for
full protection of target sensitive surface waters.

Exhibit VI compares deposition levels produced by
several  sulfur dioxide emissions  scenarios.   The
additional reduction scenarios were  chosen to il-
lustrate the effect of  further emissions reductions
and to serve as examples for cost and implementa-
tion analyses; they do not represent a  reduction
necessary to  meet any particular level of protec-
tion.  In comparison with 1980 deposition levels,
implementation of the CAAA is projected to re-
duce deposition by 30 to 40 percent by 2010,
standards   may  consider  reduction   of  effects
resulting  from  episodic  acidification   a   key
environmental endpoint.

NATIONAL OR TARGETED EMISSIONS
REDUCTIONS
To achieve an  acid  deposition standard  or  stan-
dards for particular sensitive  regions,  some  have
suggested  targeting  emissions  reductions  rather
than reducing national emissions.  By 2010, Title
IV will produce the largest emissions reductions in
the highest emitting  regions (i.e.,  Ohio,  Indiana,
West Virginia,  and  western  Pennsylvania).    An
analysis of geographically targeted emissions  re-
ductions using  the   Regional  Acid   Deposition
Model (RADM) shows that to  achieve deposition
reductions beyond the CAAA (equivalent to those
achieved by an additional national 44 percent SO2
        EXHIBIT VI. SULFUR DEPOSITION TO SENSITIVE REGIONS UNDER VARIOUS SO2 EMISSIONS SCENARIOS
Emissions Scenario
1980
1985
2010 after CAAA implementation
CAAA plus additional 50% utility^SO2 reduction
CAAA plus additional 50% utility and industrial SO2 reduction
Annual Average Deposition Level (kg-S/ha)
Adirondacks
11
9.8
6.9
5.5
4.7
Mid-Appalachians
19
17
11
8.1
6.9
Southern
Blue Ridge
14
13
9.7
6.8
5.5
Exhibit  VII  shows  the  projected percentage de-
crease in sulfur  deposition  between  1980  and
2010 with full  implementation of Title IV.  If an
additional 50 percent reduction  in utility  and in-
dustrial sulfur dioxide emissions beyond the CAAA
were to occur, then sulfur deposition would be re-
duced by about 60 percent compared to 1980.

EPISODIC ACIDIFICATION
Episodic acidification occurs when pulses of acidic
waters enter lakes  and  streams with  stormwater
runoff and spring snowmelt.  Both nitrates and sul-
fates originating from atmospheric deposition can
contribute significantly  to  episodic  acidification
events.  Rapid, acutely  toxic  changes in  surface
water chemistry  can result from  such episodic
acidification. Such events often occur at the most
biologically  significant time of year (i.e., during
spawning).   Significantly more lakes and  streams
become episodically  acidic than are  chronically
acidic.  Lower levels of acidic deposition would
decrease the number and severity of acidic and
toxic episodes driven  by  sulfate and  nitrate.
Development of  an acid deposition standard or
     EXHIBIT VII. PERCENTAGE REDUCTIONS IN
     SULFUR DEPOSITION FROM 1980 TO 2010
      FROM IMPLEMENTATION OF THE CAAA
      AND CANADIAN ACID RAIN PROGRAM
                                   LEQgMQ;
                                    0-23
                                  -*25 - 30
                                  W30 - 35
                                  »35 - 40
                                  •  > 40
                                               XVII

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
emissions reduction) in sensitive receptor regions,
zones targeted for emissions reductions may po-
tentially include 6 to 11 states and require source-
specific,   sulfur  dioxide  reductions  of   about
95 percent.  To achieve deposition  loadings in all
three sensitive  receptor  regions  (equivalent  to
those  achieved   by   an   additional  national
44 percent SO2 emissions reduction),  both  geo-
graphically targeted and national emissions reduc-
tions strategies would require about the same total
emissions reductions  at approximately the same
total cost.  It  may be  appropriate to assume  that
some level of cost savings associated with an un-
restricted national trading program (as assessed for
implementation of the Acid  Rain Program  under
Title  IV)  could  also   result  in reduced costs of
compliance with broad emission reductions be-
yond the current program.  This could widen the
cost difference between a geographically targeted
and national emissions reduction strategies.  Thus,
for  this level of additional reductions, there does
not  appear to be an economic or  environmental
advantage to  geographically targeting regions for
emissions reductions.

IMPLEMENTING AN ACID DEPOSITION
STANDARD
In order to determine  the effectiveness of an acid
deposition  standard or  standards  for  protecting
sensitive resources, it is necessary to describe how
the standard might be implemented. This study de-
scribes two basic approaches to using an  acid
deposition standard.   Under  the first approach (a
regional  target approach),  EPA would set a  stan-
dard or standards, either using existing authority (if
adequate) or seeking  further authority from Con-
gress to set such  standards and provide deadlines
for  their attainment. Then, similar to Title I, states
would  determine  source-specific    limits  using
source-receptor  models  and  technical  and  cost
analyses, incorporate those limits in State Imple-
mentation Plans (SIPs), and enforce  them.  If one
or more states failed to do the above, EPA would
promulgate a Federal Implementation Plan (FIP).

Under the second approach (a national emissions-
based approach), Congress would direct EPA to set
a deposition  standard  or standards  and to  deter-
mine the national (or regional) emissions levels for
sulfur dioxide  and nitrogen oxides that would meet
those standards.   Congress  would  then  set an
emissions cap and allowance allocations for nitro-
gen oxides and,  if necessary,  adjust the cap for
sulfur dioxide in  Title  IV, and provide a timetable
for meeting the new caps. EPA would use Title IV
provisions to implement the emissions program.

To provide a rough comparison of the cost-effec-
tiveness of the two approaches for  sulfur reduc-
tions, estimates were made of the cost of achieving
the same reduction in  sulfur deposition from utility
sources at the three sensitive areas under each ap-
proach.  Total compliance costs were similar,  al-
though the national emissions reduction approach
resulted in slightly larger and more widespread
emissions reductions.   The costs of  further emis-
sions reductions characterized in this report could
lead to costs that are more than double those of
the current acid rain control  program.   However,
the  emissions  reductions achieved  under  either
implementation  scenario may also provide  ancil-
lary  benefits in multiple effect areas, such as visi-
bility, human health, and material resources. The
level of potential benefits in these areas will de-
pend on the level and type of standard developed
and  cannot be determined until such information
is  available.  Development of an  acid deposition
standard or standards would require further analy-
sis of costs and  benefits.  This is necessary to de-
termine the level of incremental benefits in a range
of effect  areas as compared to the deposition  re-
ductions  necessary to meet  a  range of standard
levels and the costs associated with these levels.

Environmental resources have ranges of sensitivi-
ties and risks to potential effects caused by acidic
deposition. Resources having equivalent sensitivi-
ties have different risk potentials for harmful effects
that  depend on  how much acidic deposition they
receive. Aquatic modeling results presented in this
report indicate that additional reductions in sulfur
and/or nitrogen would reduce regional proportions
of chronically acidic  surface waters and propor-
tions of surface  waters most  sensitive to episodic
effects.  The magnitude  of benefits  varies by  re-
gion, and modeling uncertainty is too significant to
indicate more than direction and magnitude at this
time.

Scientific  uncertainties  make  setting   an  acid
deposition  standard or  standards at a  particular
level difficult.   However, even when  the uncer-
tainties have been resolved or reduced, setting a
single, uniform standard may be  an  inappropriate
approach in view of the differing sensitivities and
risks associated with resources in different regions
of the country.  Many have suggested that acidic
deposition goals (rather than standards) established
through consideration and  analysis  of resource
sensitivity and risk would provide useful informa-
                                                XVIII

-------
                                                                                  EXECUTIVE SUMMARY
tion and guidance.  Such goals could provide ref-
erence points to assess the effectiveness of pollu-
tion control efforts and would serve  as guides for
environmental policymakers at various levels until
scientific  uncertainties that inhibit setting a stan-
dard or standards are better understood.

FEASIBILITY AND EFFECTIVENESS OF AN ACID
DEPOSITION STANDARD
This report responds to the request by Congress to
assess the feasibility and effectiveness of an acid
deposition standard or standards for  protection  of
sensitive aquatic and terrestrial resources.  The re-
port concludes that establishing such  standards for
sulfur and nitrogen deposition  are technically fea-
sible, but two critical  areas of uncertainty advise
against completion of  the  task at this time.  First,
policy decisions regarding appropriate or desired
goals for protecting sensitive aquatic and terrestrial
resources are needed to  help guide the Agency in
continued analyses and decisions  regarding possi-
ble establishment of acidic deposition standards.
Such policy decisions would  be  based on Con-
gressional guidance and continued  efforts  to ad-
dress social  science  uncertainties  related  to the
level  of protection desired by  the public and the
costs and benefits associated with incremental  or
significant changes.   Second, key  scientific un-
knowns,   particularly regarding watershed proc-
esses leading to nitrogen acidification and remain-
ing times to watershed  saturation  with nitrogen,
limit the ability  to recommend a specific standard
for any region of North America at this time.  With
this level of scientific uncertainty, a policy goal  of
protecting all  ecosystems  could only be  assured
through standards reflecting pre-industrial  deposi-
tion levels.  This scientific  uncertainty can only be
reduced through additional research  and environ-
mental monitoring.

Determining specific requirements for an appro-
priate deposition  standard  or  standards calls  for
additional research to understand the simultaneous
effects of sulfur and nitrogen deposition in produc-
ing acidifying effects on  sensitive aquatic and ter-
restrial receptors.  Establishing specific needs and
appropriate  limits for deposition standards requires
advancing the  scientific understanding of factors
key to (1) defining times to  watershed  nitrogen
saturation and (2) the interaction of sulfur and ni-
trogen in acidifying watersheds and surface waters.
Additional scientific studies to determine the po-
tential impact   of acidic  deposition  on  climate
change may also be reliant to fully characterize
ecological effects.

If a deposition standard or standards were devel-
oped, an  extensive monitoring program would be
necessary to (1) ensure compliance with the stan-
dard^), (2) determine the effectiveness of the stan-
dard(s), and (3) verify whether standards were cor-
rectly defined or an  alternative standard may be
more appropriate.

EPA's Science Advisory Board (SAB) review of this
report emphasized that policy concerns will influ-
ence which science questions will drive  research
priorities regarding the need and feasibility of set-
ting acid  deposition standards.  Results from  new
research addressing the issues described above are
needed to provide the scientific  confidence and
assurance to develop and implement effective acid
deposition standard  or standards.  The  SAB em-
phasized  the importance of environmental moni-
toring  of  deposition,  ecological  indicators,  and
ecological endpoints  as a parallel  and  comple-
mentary strategy to modeling  in  order  to assess
ecological resource issues.f Furthermore, guidance
from Congress and the public is critical to defining
the extent of protection desired for acid- sensitive
aquatic and terrestrial resources.

Finally, the effectiveness  of an  acid deposition
standard or  standards depends heavily on  the ap-
proach  used to  implement it.  Although  the two
basic  approaches discussed in  this report could
have  similar  compliance   costs   and effects  on
aquatic  resources,  the   national  market-based
emissions  reduction approach may be less admin-
istratively  cumbersome and more compatible with
the existing Title IV and is more likely to be imple-
mented.  Further analysis is necessary to determine
what level of  ancillary benefits to human health,
visibility,  and  material and  cultural resources
either approach would provide and the  extent to
which one approach provides greater benefits than
the other.   Title  IV  was designed to address re-
gional air pollution problems, especially those in-
volving long-range transport of pollutants and their
transformation  products.    The  likelihood  of
achieving deposition   reductions   is  viewed as a
critical factor in judging effectiveness.
t An SAB  Report: Review  of the  Acid Deposition
  Standard Feasibility Study  Report to Congress. Acid
  Deposition  Effects Subcommittee of the  Ecological
  Processes    and   Effects   Committee,    United
  States Environmental  Protection  Agency  Science
  Advisory Board, EPA-SAB-EPEC-95-019, July, 1995.
                                                 XIX

-------
                                        CHAPTER 1

                                     INTRODUCTION
Title IV of the Clean Air Act Amendments of 1990
(CAAA  or the  Act)  addresses  the  problem  of
harmful effects on ecosystems from acid  rain by
mandating reductions in emissions of sulfur and ni-
trogen oxides—the major precursors of acid rain.
Coupled with Titles I and II  of the Act, which ad-
dress new and existing  stationary and  mobile
sources of sulfur and nitrogen oxides, implementa-
tion  of  Title IV is expected  to provide significant
benefits to the United States and  Canada. Those
benefits include decreases in the acidity of lakes
and  streams,  concomitant improvements  in fish
population diversity  and health,  decreases  in soil
degradation  and forest  stress,  improvements  in
visibility (especially to scenic vistas), decreases  in
damage to materials  and cultural resources, and a
reduction   in  human   health  effects.  Congress
included Section 404  in  Title IV (Appendix B  of
the Act) which  requires the U.S.  Environmental
Protection Agency (EPA or the Agency) to provide
a report to Congress on the feasibility and effec-
tiveness of an acid deposition standard to protect
sensitive and  critically sensitive aquatic and terres-
trial  resources. Specifically, Congress listed six ar-
eas to be addressed in the report:

  »  Identification of sensitive and critically sen-
     sitive  aquatic and terrestrial  resources in
     the U.S. and Canada  which  may be af-
     fected by the  deposition of  acidic  com-
     pounds;

  »  Description and specification of a numeric
     value for an acid deposition standard suffi-
     cient to protect such resources;

  *  Description of the use of such standards or
     standards in other Nations or by any of the
     several States in acidic deposition control
     programs;

  *  Description of measures  that  would be
     needed to integrate such standard or stan-
     dards with the control program required by
     Title IV of the Clean Air Act;

  4  Description of the state of knowledge with
     respect  to source-receptor  relationships
     necessary to develop a  control program on
     such standard or standards and additional
     research  that is  on-going  or  would be
     needed to  make such a control program
     feasible;

   * Description of impediments to implemen-
     tation  of such control  program  and  the
     cost-effectiveness  of deposition  standards
     compared to other control strategies  in-
     cluding ambient air quality standards, new
     source performance standards and the re-
     quirements of Title IV of the Clean Air Act.

Congress also requires the  National Acid Precipi-
tation Assessment  Program  (NAPAP) to conduct a
study similar to the technical portions of this re-
port, on the reduction in deposition rates needed
to prevent  adverse ecological effects.  NAPAP is
required to submit its report  to Congress in  1996
(Section 901 [j] of the Clean Air Act Amendments).

To achieve  significant reductions in  emissions of
sulfur and  nitrogen oxides, Title IV targets  emis-
sions from  electric utilities—the major source of
sulfur dioxide (SO2) emissions and a major source
of nitrogen oxide  (NOX)  emissions.  Annual  emis-
sions of SO2 are to be reduced by 10 million tons
from  1980 levels  through an innovative market-
based allowance trading program. The program es-
tablished an SO2 allowance trading system that al-
lows utilities to  minimize the cost  of  complying
with SO2 emissions reduction requirements,  while
maintaining a cap  on SO2 emissions from utilities.
The trading program encourages energy conserva-
tion and technological innovation, which should
yield  pollution prevention benefits and minimize
compliance costs.

Annual allowances for SO2 emissions  have been
allocated to affected utility units based  on their
historic  emissions  and  fuel use.  Each  allowance
permits  a  utility to emit  1 ton of SO2. Each unit
must hold a sufficient number of allowances  at the
end of the year to cover its emissions for that year.
Emissions reductions will be implemented for 263
units  under Phase I  beginning in 1995  and  for
approximately 2,200 units affected under Phase II
in 2000. Utilities may buy, sell, trade, or save al-
lowances for future use. When the program is fully

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
implemented  in  2010,  nationwide emissions of
SO2 from affected utilities  (i.e., units  generating
over 25 MW) will be capped at 8.95 million tons
per year. In  addition, nationwide emissions of SO2
from industrial sources are  capped at 5.6  million
tons per year. These emissions are not included in
the allowance trading program, but  some  indus-
trial sources may be allowed to elect to participate
in the program under rules  currently  being devel-
oped by EPA.

Title IV also specifies that standards be set for NOX
emissions from utility boilers, with the goal of re-
ducing  nationwide emissions  by  2  million tons
from 1980  levels.  Unlike  the  SO2 program, the
NOX program does not use a tradable allowance
system and  does  not cap  emissions,  but  instead
calls for  low-NOx burner  technology  to  reduce
emissions.

Concerns  exist regarding how allowance  trading
will affect protection of sensitive aquatic and ter-
restrial resources. Some critics argue that the cau-
tious nature of the electric utility industry and lo-
cal pressure in high-sulfur coal regions to continue
use of  regional coals to preserve local economies
will not  allow  trading to  achieve economically
meaningful  benefits. Others argue that  extensive
trading could  result in a geographic distribution of
emissions that would prevent achievement  of the
Act's  goal  of  reducing the  effects  of   acidic
deposition in  sensitive  areas, and that emissions
levels designed to provide a level of protection are
necessary.

The process for determining the degree of  protec-
tion afforded by Title IV to sensitive and critically
sensitive aquatic and terrestrial  resources is com-
plicated by  a number of scientific and technical
uncertainties.  For example,  there are  gaps  in eco-
logical effects research,  particularly regarding ni-
trogen  cycling and retention  in  forested   water-
sheds.  The  scientific  community  is  still learning
how the  multiple impacts of sulfur and nitrogen
acidic  deposition affect ecosystems.  In addition,
meteorological   variability,   uncertainties   in
emissions inventories,  and  the  complexity of at-
mospheric chemistry limit the ability to relate spe-
cific ecosystem damage to specific point sources.

The purpose of this study  is to integrate state-of-
the-art ecological effects research, emissions and
source-receptor modeling work, and  implementa-
tion and   cost  issues  when  considering  the
feasibility of setting and implementing a standard
to protect aquatic and terrestrial resources from the
effects of acidic deposition. This report addresses
three broad themes:

  1.  What acidic deposition levels are necessary
     to protect sensitive regions?

  2.  What degree of protection is provided by
     Title IV? What  is the residual  risk? What
     additional emissions limitations would be
     required to protect sensitive regions?

  3.  How would an acid deposition standard(s)
     be  implemented? What  are the different
     implementation approaches? What are their
     relative feasibility and effectiveness?

A  common thread running through each theme is
uncertainty  in  (1)data  and  models,  (2)  future
ecosystem behavior, and (3) future economic and
policy decisions  that may influence decisions
regarding feasibility.

WHAT ACIDIC DEPOSITION LEVELS ARE NEC-
ESSARY TO PROTECT SENSITIVE REGIONS?
The report outlined in Section 404 (Appendix B) of
the Act requires identification of sensitive aquatic
and terrestrial resources and description of the na-
ture and  numerical value for a deposition standard
that  would  protect  these  resources.  An  acid
deposition standard or standards can be designed
to achieve a variety of environmental  goals. For
example, the goal of a standard may be (a) mainte-
nance of specific conditions as observed at  a par-
ticular point  in time (i.e.  conditions observed in
1984 during the National  Surface Water Survey),
(b) return to pre-industrial conditions, or (c) a level
which  balances effects,  costs,  and other societal
values.   A standard can  be designed to address
chronic or episodic  acidification  and could vary
by region  based  on  the  regional  variability of
ecological sensitivity. No guidance is provided in
the  statutory  language  regarding the  desirable
degree of protection afforded by a standard or
standards.

Chapter 2 of this feasibility study brings  together
the most current scientific understanding regarding
the  relationship between  acidic  deposition  and
ecological    effects,   specifically   effects   on
watersheds. The information  comes from  research
conducted  by EPA's Office of Research and De-
velopment, peer-reviewed  literature, and efforts to
define appropriate acid deposition standards in the
United States and other countries. Best understood
from the scientific point of view, is the relationship
between acidic deposition of sulfur and its effects

-------
                                                                            CHAPTER 1: INTRODUCTION
on stream and lake ecosystems. Effects of nitrogen
deposition  on aquatic systems and of combined
sulfur and  nitrogen deposition on terrestrial sys-
tems are less clear and  poorly quantified for re-
sources in the United States.

Regions of North America differ in their sensitivity
to acidic deposition (i.e., ability of a watershed to
buffer acidity) and in the amount of acidic deposi-
tion they receive. Some parts of the eastern United
States are highly sensitive and chronically or epi-
sodically   receive  damaging  concentrations  of
acidic deposition. Other sensitive regions, such as
the western  United States,  are unlikely  to  suffer
adverse chronic effects  at current or  projected
rates of acidic deposition. Certain high-elevation
western lakes, however,  are subject to episodes of
acidic  deposition.  Chapter 2  identifies  sensitive
aquatic and terrestrial resources in specific regions
of the  United  States and  describes the  effects
caused  by  acidic  deposition  in  each  region.
Aquatic resources  of concern  include  fish and
other species, as well as the water quality of lakes
and  streams.  Terrestrial  resources   of  primary
concern include trees and forest soils,  although
fewer data  are currently available in these areas.

The relative  contributions and importance of sul-
fur- and nitrogen-containing compounds to the ef-
fects of acidic  deposition differ among regions.
The importance of each  group of compounds de-
pends on  its relative deposition  level and on the
capacity of individual watersheds to retain depos-
ited nitrogen and  sulfur. Sulfur appears to be the
principal cause  of ongoing,  chronic  acidification
of aquatic  systems within  most  affected  areas in
eastern  North America.  The importance  of  nitro-
gen deposition, however, cannot be overlooked for
several  reasons, including:   nitrogen, as well  as
sulfur,  produces episodic surface water acidifica-
tion  effects,  especially during spring snowmelts;
and  some  watersheds in the  Northeast  may be
approaching the limit of their  ability to sequester
nitrogen, leading to  increased  acidification from
nitrogen deposition.  The importance of consider-
ing the effects  of  nitrogen   deposition  on  eu-
trophication of estuarine  bodies is identified in this
report with fuller discussion provided in other EPA
reports.1

Chapter 2 assesses these  issues by analyzing  avail-
able scientific data  within a risk-based context and
produces a set of environmental goals. The analy-
ses concentrate on the three regions in the United
States most extensively characterized: the Adiron-
dacks,  the mid-Appalachians, and the Southern
Blue Ridge Province, each  of which is subject to
deposition from sources in the East.  Knowledge of
current and historic differences in deposition levels
and watershed  sensitivities  in these three  regions
makes it possible to discern  differences in ongoing
effects and remaining risk for each. Risks for other
regions  of the United  States and Canada are  also
described, but in  a more qualitative sense. Effects-
based (critical and target load) control  strategies
and approaches adopted by Europe, Canada,  and
several  states in  the United  States are also  dis-
cussed.  (Note that an  effects-based analysis  and
development of  an acid deposition standard  or
standards  does  not necessarily  imply emissions
reductions  associated  with  implementation of a
standard or target loads.)

Although this report focuses on aquatic and terres-
trial systems  at risk,  acidic deposition and its  pre-
cursor emissions  also can affect  visibility, human
health,  and materials.  Decreasing acidic  deposi-
tion can also provide  benefits in  these areas,  de-
pending on what  size emissions reductions may be
involved.  Visibility,  especially   in  the  eastern
United States, is markedly degraded by sulfate par-
ticles  in the atmosphere.   Human  health  effects
from exposure to  SO2,  NO2,   and ozone  (O3,
formed  by chemical  reactions involving  nitrogen
dioxide) are well  known, and  effects from particu-
late matter, including acidic  aerosols,  are docu-
mented  as well. Damage to materials and cultural
resources  by acidic  deposition  has been docu-
mented  by  the National Acid  Precipitation  As-
sessment Program (NAPAP).2 Visibility degradation
has been addressed more fully in another Report to
Congress3 and  is covered under Section 169A of
the Act. Also, primary  National Ambient Air Qual-
ity  Standards (NAAQS) have  been  established to
protect  the public health from adverse  effects of
criteria pollutants, including SO2, NO2,  particulate
matter  (including sulfates and nitrates), and   O3.
Nevertheless, any control program or standard es-
tablished to  reduce acidic  deposition could  also
1  U.S. Environmental Protection Agency  May  1994.
  Deposition of Air Pollutants to the Great Waters. First
  Report to Congress. EPA-453/R-93-055.
2 Irving,  P.M., ed.  1991 Acidic Deposition: State of
  Science and Technology Summary Report. National
  Acid Precipitation Assessment Program. Washington,
  DC.
3 Office of Air Quality Planning and Standards   Octo-
  ber 1 993.  Effects of the 1990 Clean Air Act Amend-
  ments on Visibility in Class I Areas:  An EPA Report
  to Congress. U.S. Environmental  Protection Agency,
  Washington, DC.

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
further impact these other related areas. Chapter 4
of this report highlights potential impacts  a  stan-
dard may have on visibility, human  health,  and
materials in a primarily qualitative manner.  How-
ever, additional research is required to determine
the extent to which a standard would  provide an-
cillary benefits in these areas.

WHAT DEGREE OF PROTECTION is PROVIDED BY
TITLE IV? WHAT is THE  RESIDUAL RISK? WHAT
ADDITIONAL EMISSIONS LIMITATIONS  WOULD BE
REQUIRED TO PROTECT SENSITIVE REGIONS?
The complex relationship between emissions and
deposition depends on a great number of physical,
chemical, and biological processes. Acidic  deposi-
tion results from  a complex series of  interactions
among chemicals in the atmosphere. Airborne sul-
fur and nitrogen species can be transported hun-
dreds of kilometers by meteorological  forces. Dur-
ing transport these species can remain unchanged
or react with other atmospheric pollutants,  such as
volatile organic compounds (VOCs), to form new
compounds, some of which are acidic. These pol-
lutants are then  deposited to the earth through
either wet or dry deposition.  To understand the
environmental impact of the Act and to develop
and analyze strategies to protect ecosystems from
acidic deposition, the relationship between emis-
sions and deposition (i.e., the source-receptor rela-
tionship) should be addressed not only in the pre-
sent, but also in the future.

SO2 and NOX reduction mandates established by
Title IV provide  for  a  nationwide  decrease  in
acidic deposition  precursors.  Geographic or  re-
gional restrictions  do  not exist.  The  inherent
flexibility in  choosing compliance strategies and
the market-based allowance trading program  pro-
mote the most cost-effective strategy for achieving
SO2 reductions, rather  than  requiring a specific
type of control  on certain  sources.  Variation  in
sensitivity to  acidic deposition among geographic
regions raises the question of whether targeted or
regional  standards are needed to protect sensitive
resources.  Before such  a question  can  be an-
swered, determining  the  level  of  protection that
will be provided by full  implementation of Title IV
in 2010 and in subsequent years is necessary.

To answer questions regarding the effectiveness of
Title IV in  protecting sensitive areas (i.e., the re-
sidual risk after implementation of Title IV)  and the
impact of  additional control, several alternative
emissions scenarios are developed in Chapter 3:
  * A scenario that achieves the SO2 emissions
    reductions mandated by the Act,

  * A scenario to assess the environmental im-
    pact of trading SO2 allowances, and

  * Scenarios  that  achieve additional  reduc-
    tions of SO2 and NOX emissions from utili-
    ties  and  industrial  sources  beyond those
    required by the Act.

The  scenarios chosen   were   limited  by  the
availability of emissions inventories at the time this
report was developed.

The Regional Acid Deposition  Model (RADM)4 is
used to translate each emissions scenario to depo-
sition values for the  eastern  United States. Deposi-
tion of sulfur and nitrogen  species are then com-
pared for each scenario, with particular emphasis
on the three key sensitive areas—the Adirondacks,
the mid-Appalachians,  and the  Southern  Blue
Ridge Province.

HOW WOULD AN ACID DEPOSITION STAN-
DARD^)  BE IMPLEMENTED?  WHAT ARE THE
DIFFERENT  IMPLEMENTATION APPROACHES?
WHAT ARE THE FEASIBILITY  AND EFFECTIVENESS
RELATIVE TO OTHER APPROACHES?
In Chapter  5, two broad approaches are reviewed
for degree of protection,  geographic coverage, im-
plementation difficulty, and cost.  The approaches
follow the emissions reductions scenarios modeled
in Chapter 3. The two broad implementation ap-
proaches are  (1)  a national,  emissions-oriented,
market-based  approach  and (2) a regional,  stan-
dard-oriented, source- (region-) specific limit ap-
proach.  Control  of both  utility  and   industrial
sources is assessed.

A variety of factors should be considered in im-
plementing  an acid deposition  standard. To be
successful,  an implementation approach must have
clear goals and must provide certainty as to the re-
sponsibilities of the regulated  community,  EPA,
and states.  Chapter  5 identifies and describes the
factors that may affect implementation of a stan-
dard under both national and regional approaches.
Four general categories of factors are considered:
4 Chang, J.S., P.B.  Middleton, W.R. Stockwell, C.J.
  Walcek, J.E. Pleim, H.H. Lansford, F.S. Binkowski, S.
  Madronich,   N.L.  Seaman,  and  D.R.  Stauffer.
  December 1990.   The Regional  Add Deposition
  Model and Engineering  Model.   NAPAP  SOS/T
  Report 4.  In: Acidic Deposition: State of Science and
  Technology. National Acid Precipitation Assessment
  Program.

-------
                                                                      CHAPTER 1: INTRODUCTION
* STATUTORY AUTHORITY:  Is existing authority
  adequate, or would Congress need to pro-
  vide additional authority to  implement an
  acid deposition standard or standards?

* ADMINISTRATION/COMPLIANCE: How enforce-
  able and  administratively complex are al-
  ternative  regulatory   approaches?   How
  would the approach be administered and
  enforced? Would new  administrative enti-
  ties be needed? What level of federal and
  state resources would be needed?

» INTERACTION AND  INTEGRATION WITH OTHER
  ENVIRONMENTAL PROGRAMS: An acid deposi-
  tion  standard may impose additional limits
  on  SO2  and NOX  emissions from  point,
  area, and/or mobile sources. Existing fed-
  eral, state, and  local regulations (including
     the Title IV Acid Rain Program, Title I  Am-
     bient Air Standards,  and Title  II Mobile
     Source regulations at the federal level) ad-
     dress emissions from  these sources. What
     effect would  implementation  of an  acid
     deposition standard have on  these  and
     other environmental programs?

  *  Economic  Impacts: What would be  the
     costs and  economic  impacts  of an  acid
     deposition standard to the regulated com-
     munity, as well  as the  national and local
     economies?

Chapter 6 integrates  analyses of  environmental
goals, emissions reductions,  and implementation
issues and  provides conclusions concerning  the
feasibility of developing and  implementing  a stan-
dard or standards for acidic deposition.

-------
                                           CHAPTER 2
                                   ENVIRONMENTAL COALS
2.1  INTRODUCTION
Title IV of the 1990 Clean Air Act Amendments
(CAAA) addresses the problem of effects of acidic
deposition on environmental resources by mandat-
ing nationwide reductions  in  emissions of sulfur
and nitrogen oxides  from electric utility generating
units, the major contributor to acidic deposition.
While reductions in total emissions will  benefit
many aquatic  and terrestrial  resources, Congress
mandated a  study of whether more specific acid
deposition standards may be appropriate.  Under
Section  404 (Appendix B) of the CAAA, EPA must
assess the feasibility  and effectiveness of establish-
ing an acid  deposition standard, or standards, to
protect  sensitive  aquatic and  terrestrial resources.
This chapter addresses three  specific Section 404
requirements:

   * Identification of the sensitive and critically
     sensitive aquatic and terrestrial resources in
     the United States and Canada  which may
     be  affected by the deposition of acidic
     compounds;

   * Description of the nature and numerical
     value of a deposition standard or standards
     that would be sufficient to protect such re-
     sources;

   * Description of  the use of such  standard or
     standards  in other Nations or by any of the
     several  States in acidic deposition control
     programs.

Section 2.2  of  this  report reviews surface water
acidification  and recovery  processes. The two
most common measures of surface water acidifica-
tion are pH and acid neutralizing capacity (ANC).
Low ANC is a  common indicator of sensitivity to
acidification; other parameters, including pH, dis-
solved aluminum, and sensitive biological species,
also provide useful information on resource health.
Atmospheric deposition of  sulfur  and nitrogen
compounds that form acids is the principal cause
of surface water  acidification.  Most recent atten-
tion has focused  on  the effects and control of sul-
fur deposition  (Section 2.2.1).  Although  many
studies have  focused primarily on long-term acidi-
fication  processes,  recent  EPA  research supports
the contention that short-term acidification caused
by rainstorms and snowmelt may often be the in-
itial cause of many of the most severe acidification
effects in surface waters.  Consideration of acid
deposition standards may take into  account impli-
cations to both the long- and short-term acidifica-
tion processes (Section 2.2.2). Recent research also
indicates that acidification  effects caused by nitro-
gen deposition are increasingly important in some
areas. The increasing degree of nitrogen saturation
in some watersheds may be leading to long-term
and short-term increases in  nitrate  concentration
and  concomitant  acidification  of   some  surface
waters (Section 2.2.3). Section 2.2.4 notes that al-
though a number  of studies indicate that  surface
water acidification can be reversed by reducing
emissions and, at least temporarily, by practices
such as liming (e.g., the application of powdered
limestone),  restoration  of  ecological  systems to
their predisturbance conditions may not be possi-
ble.

Section 2.3 introduces a risk-based approach to as-
sessing the need for an acid deposition standard.
Alternative approaches for defining resource sensi-
tivities are reviewed, and environmental and land-
use characteristics affecting these sensitivities are
described  (Section 2.3.1).  Section  2.3.2  empha-
sizes that using a risk-based approach requires as-
sessing four  central concerns: (1) sensitivities of
potentially affected resources; (2) factors extrinsic
to selected resources that may alter their sensitivi-
ties;  (3) geographic  location  of the selected  re-
sources; and (4) exposure intensity,  duration,  and
timing of acidic deposition at the  selected loca-
tions.

Section 2.4  addresses  the  first Congressional  re-
quirement listed  above in  italics.  Section 2.4.1
emphasizes  that assessments of the  need for acid
deposition  standards should  focus  primarily  on
potential benefits to sensitive  aquatic resources in
six regions of the eastern half of the United States
and to sensitive stands of red spruce  at high eleva-
tions in two  of those regions.  An EPA-sponsored
literature review confirmed that the potential  ef-
fects of nitrogen deposition is an increasing con-
cern  in  the western   United  States  as  well
(Section 2.4.2). In  Canada,  roughly south of James
Bay and east of the  Manitoba-Ontario border, re-

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
source concerns similar to concerns for the eastern
United States are emphasized (Section 2.4.3).

Section 2.5  addresses the second Congressional
requirement by describing the potential  need to
protect acid-sensitive resources and the potential
benefits derived from additional control of  acidic
deposition. Simulation modeling is the  best avail-
able way to project future ecological effects of
possible  changes  in deposition   rates.  This  ap-
proach has  limitations, however,  as  results carry
considerable uncertainty (Section  2.5.1).  EPA has
completed two major effects modeling studies. The
first  study,  conducted  under  the  National  Acid
Precipitation Assessment  Program  (NAPAP), pro-
jected the impact of sulfur deposition on long-term
soil  and  surface water acidification,  with conse-
quent loss of aquatic habitat for sensitive fish spe-
cies, in three broad  geographical  regions of the
eastern  United States. Adverse effects  were pro-
jected to continue unless sulfur deposition was re-
duced, and  sufficient reductions  in sulfur deposi-
tion  were projected  to  likely reverse  these effects
(Section 2.5.2). The  second major EPA modeling
study illustrated the  role  that  nitrogen  deposition
may play as an important cause of soil and water
acidification, and its importance was  projected as
likely to  increase in future years unless deposition
rates  decreased (Section  2.5.3).  This study also
projected that the 1990 CAAA would provide clear
benefits to surface water in three sensitive regions
of the eastern United States. The  accuracy  of the
model  projections is highly uncertain,  however,
largely because researchers and available predic-
tive  models lack the ability to precisely  estimate
how long it  takes a sensitive watershed  to become
saturated with deposited nitrogen.

In addressing the third Congressional  requirement,
Section 2.5.4  reviews  various acidic  deposition
criteria and standards that have  been  developed
internationally  and by  individual  states, many of
which  are being revised  as additional and more
accurate information becomes available. This sec-
tion  is not,  however, a comprehensive list of all
state or international efforts. Important ecological
and  geographical  concerns that  raise questions
about whether the application of  deposition stan-
dards should take into  account regional and sea-
sonal conditions are  discussed in Section 2.5.5. Fi-
nally, assessments of the need and options for de-
veloping deposition standards address concerns re-
lated to  both  sulfur  and  nitrogen  deposition
(Section 2.6).
2.2  BASIC RELATIONSHIPS IN SURFACE WATER
     ACIDIFICATION AND RECOVERY
Understanding the need for, and feasibility of,  es-
tablishing acid deposition standards and  under-
standing the sensitivities of resources to acidifica-
tion, requires a knowledge of how  environments
assimilate atmospherically deposited acids and
acid-forming  chemical  compounds.  This  knowl-
edge  is  key to assessing whether surface water
acidification  effects are occurring  in  a  region,
when these effects  may have begun, how exten-
sive the effects are and may become, and the criti-
cal periods when these effects may be most severe.
This section briefly  reviews the process of  surface
water  acidification, important considerations  in
identifying at-risk resources, episodic acidification,
and useful information for setting regional and  re-
source priorities for  acidic deposition controls. The
section concludes with a brief discussion of the  re-
sponses of acidified ecosystems to  reductions in
acidic deposition  levels.   First,  a  few  common
terms and concepts, key to understanding  surface
water acidification processes,  are introduced.

The common measure of acid-base  conditions in
solutions is the pH  scale. On this scale, neutrality
(i.e., neither acidic nor basic) occurs at pH 7.0;
acidic conditions  have  lower values  (pH<7.0),
while basic (also termed alkaline) conditions have
higher values  (pH>7.0).  The  most acidic condi-
tions  occur  near  pH 0.0  and the most  alkaline
conditions occur near  pH  14.0.  On this  scale,
each full unit decrease in pH (e.g., from 7.0 to 6.0)
represents a ten-fold increase in acidity and in the
concentration of hydrogen ions that cause acidity.

Some surface waters  are naturally acidic.  This is
largely due to (1) carbon dioxide from the atmos-
phere dissolving to  form carbonic acid and (2)  or-
ganic acids produced by the  decay of dead plant
materials. Waters in some bog lakes,  for example,
can have natural  pH levels below 5.0. Similarly,
pure rain water and distilled  water in equilibrium
with atmospheric concentrations of carbon dioxide
naturally have pH  levels near 5.6. Dissolution of
natural organic  acids from the atmosphere into
rain water has been found to increase rain water
acidity and lower pH to near 5.2 in  some pristine
open ocean areas.

The natural tendency toward acidic  conditions is
countered in  most  surface waters by the dissolu-
tion of common, alkaline  minerals such as lime-
stone that dissolve into them. This dissolution neu-
tralizes the acidity, often  producing  slightly alka-
line conditions (alkalinization). The  dissolution of

-------
                                                                    CHAPTER 2:  ENVIRONMENTAL GOALS
many minerals not only neutralizes acidic condi-
tions in  waters, it  produces a buffering capacity
that enables these waters to minimize pH changes,
while allowing  their mass of dissolved acids to
vary within certain limits. Additional buffering ca-
pacity can be produced by solutions of weak ac-
ids, including carbonic acid and many organic ac-
ids.  The extent of acid-base buffering within any
watershed in the environment is determined by the
specific combinations of dissolved materials.

In total,  the interaction of these natural acidifica-
tion, alkalinization, and buffering processes causes
the pH in most surface waters to range from about
6.5  to 8.0. Much  of the  concern  about surface
water acidification focuses on the effects that may
occur with decreases, especially below pH 6.5, of
0.5 to 2.0 pH units or more (i.e., increasing
acid concentrations in the environment by
500 to 10,000 percent or more).
        significantly moderated potential for pH fluctua-
        tions below 6.0. Also, they generally have minimal
        development of acidic water qualities that can be
        stressful, or even toxic, to aquatic organisms. In
        turn, waters with ANC of 50 ueq/l or less tend to
        be the most sensitive to severe and long-term pH
        depressions below  6.0, which  can produce the
        most  severe effects  on aquatic life. The general
        empirical  relationship between pH  and ANC can
        be characterized  (e.g., as within EPA's Nitrogen
        Bounding  Study  [NBS]   [see Section 2.5.3]) for
        three  regions  in  the eastern United  States  that
        contain sensitive  ecological  resources (Exhibit 1).
        This exhibit shows, for example, that an ANC of
        50 ueq/l correlates to a pH of approximately 6.5
        across these regions.
Acid neutralizing capacity  (ANC)  is  the
term  commonly  used  to  describe  the
concentration  of  dissolved  compounds
present in fresh water that collectively tend
to neutralize water pH, creating less acidic
and more alkaline conditions. Greater ANC
generally correlates with greater buffer ca-
pacity in the water.  In  most fresh  waters
ANC   is   determined   primarily   by
concentrations    of     carbonate    and
bicarbonate, which generally dissolve into
the  water  from  calcium  carbonate  (the
predominant   chemical   constituent   of
limestone, for example). Concentrations of
borates, phosphates, silicates, sulfides, and organic
anions can also contribute to total  ANC in surface
water. In earlier literature, the term alkalinity was
often used  in place  of  ANC.5   In  most  recent
literature, however, alkalinity is used primarily in
discussing  total   dissolved   concentrations  of
bicarbonates, carbonates, and hydroxides in water.
The total capacity of a surface water to neutralize
acidity can  include other chemical and biological
processes—the most  important of which are the
biologically mediated  processes of sulfate  (SO42~)
and nitrate (NO3-) reduction.

Surface waters with   higher ANC  are  generally
more resistant to acidification and  have higher pH
levels. That  is, lakes and streams with ANC greater
than 200 microequivalents per liter (ueq/l)  have
EXHIBIT 1. EMPIRICALLY DETERMINED RELATIONSHIP BETWEEN
 ANC AND PH FOR THREE SENSITIVE REGIONS, INCLUDING
    CROSS-REGIONAL MEAN AND STANDARD DEVIATION
ANC
(ueq/l)
-10
0
10
40
50
60
Empirical pH for Sensitive Regions
Adirondacks
4.96
5.28
5.69
6.31
6.41
6.50
Mid- Appalachi-
ans
4.98
5.30
5.72
6.36
6.47
6.55
Southern
Blue Ridge
4.95
5.27
5.78
6.53
6.65
6.73
   Drever, J.I. 1982. Geochemistry of Natural Waters.  Prentice-
   Hall, Inc., Englewood Cliffs, NJ.
        Acidic deposition can lead to two kinds of acidifi-
        cation processes. First, over the longer term, the
        fundamental character of soil and water  chemis-
        tries can shift to chronically acidic conditions due
        to the input and accumulation of deposited acidic
        ions.  Such  conditions  can  produce  long-term,
        chronically toxic, and lethal environmental effects.
        Second,  acutely acidic   conditions  can  rapidly
        develop during periods leading to,  accompanying,
        or  following  episodic  events,  which primarily
        accompany discharges  of storm  and  snowmelt
        water runoff. Pulses of highly acidic water flushing
        into and through soils, streams, and  lakes often
        expose  soil  and  aquatic  biota  to  short-term,
        acutely toxic, lethal chemical conditions.

        When considering acidification effects, it is impor-
        tant to recognize that the earliest  effects  on bio-
        logical  components  of   an  aquatic  ecosystem
        commonly  accompany early episodic acidification
        events.  For acid-sensitive fish  species in  some
        lakes and streams, for example, these events can

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
cause complete spawning or recruitment failures.
As  chronic  acidification  becomes  more  pro-
nounced, such effects can become more frequent
and may result in further harmful effects to overall
species richness. In contrast, for surface waters in
regions  recovering  from  acidification,  this  se-
quence reverses as  occasions of episodic  effects
become less and less frequent, until acidification
effects  apparently end.   In some recovering re-
gions,  however, proportions of lakes and streams
affected by episodic acidification may increase for
a time before decreasing. This temporary increase
would likely result wherever recovery of lakes and
streams  affected   predominantly  by   long-term
chronic acidification progressed more rapidly then
recovery of  those waters  affected predominantly
by short-term episodic acidification.

As more fully discussed in several of the following
sections, available  information  indicates that sur-
face  waters  with   ANC<50  ueq/l  include  the
aquatic resources most sensitive to potential effects
from episodic acidification.  Thus, ANC is  a very
important response variable for use in evaluating
acidification-related  changes  in  surface  waters,
particularly streams. As such, ANC is a primary fo-
cus of discussion throughout this chapter. The next
four subsections review potential chemical rela-
tionships, episodic  acidification, cumulative  ef-
fects,  and recovery processes associated with soil
and  water acidification.   Biological  implications
are discussed in reviewing of identification  of re-
source priorities. Additional issues are summarized
in Appendix  A, Summary of NAPAP reports.

2.2.1   Relationship of Base Cations, Sulfur,
        and Nitrogen in  Surface Water
        Acidification
Acidic deposition increases total load of hydrogen
ions (H+) and acidic anions (primarily SO42~ and
NO3-)  in watersheds. A vast majority of these  de-
posited ions interact  within  the  watersheds,  ex-
changing with and displacing ions of other chemi-
cal species  from  watershed receptors primarily
through soil  weathering and chemical equilibrium
processes in soil waters,  and through  biological
uptake  processes adjacent to plant roots and soil
microbes.  Watershed ions exchanged for  depos-
ited ions enter soil water  solutions and can subse-
quently drain into streams and lakes.

In a report  that provided much  of the  basis for
EPA's modeling research  on  watershed responses
to acidic deposition, the National Academy of Sci-
ences (NAS)  identified two geochemical processes
as  the dominant  watershed  factors  mediating
long-term surface water acidification.6  The first is
the rate at which watershed sources  exchange
base cations, especially calcium (Ca2+) and mag-
nesium (Mg2+), for H+ through neutralization and
buffering  processes.  Essentially  all base  cations
within a watershed are supplied initially through
the relatively slow process of mineral weathering,
while  much more rapid supplies of base cations
can  be  available  through  exchange  processes
within soil  solutions  and by soil  biota.  Acidic
deposition can accelerate each of these processes.

The second dominant factor that NAS identified as
affecting acidification is the capacity of a  water-
shed  to  retain deposited  sulfur-containing com-
pounds. This process is important  because a vast
majority of atmospherically deposited sulfur  is in
the form of SO42~ or other inorganic sulfur-contain-
ing compounds that rapidly oxidize to  SO42~. The
process of SO42' adsorption by soils directly affects
the mobility of SO42' in watersheds and, thus, the
mobility of assorted base cations and acidic cat-
ions (e.g., H+ and aluminum, AI3+).

The NAS report concluded that the external proc-
ess of acidic ion deposition balanced  against in-
ternal  watershed processes  of base cation  supply
(i.e., acid assimilation) and SO42~ adsorption  (i.e.,
anion  mobility) critically determine the rates and
degrees  of  long-term acidification  of soil  and
water. The production of base cations and internal
retention of sulfur in watersheds can decrease as
available  supplies of base cations and SO42~ ad-
sorption  abilities are depleted.  Surface waters in
watersheds with insufficiently available base cat-
ion supplies and minimal net annual sulfur reten-
tion tend to be at  greater risk from  acidic deposi-
tion.

EPA initiated the Direct Delayed Response Project
(DDRP) (see Section 2.5.2) to help understand how
these  factors interact  and affect  environmental re-
sponses to acidic deposition. This project primarily
addressed the question of whether watersheds tend
to acidify immediately in proportion to the inten-
sity of deposition (i.e., "direct" acidification) or lag
in time due to internal watershed  processes  (i.e.,
"delayed" acidification).
6  National  Academy  of Sciences.  1984.   Acid  Deposition:
   Processes of Lake Acidification, Summary of a  Discussion.
   National Research Council Commission on Physical Sciences,
   Mathematics, and Resources. Environmental  Studies Board,
   Panel on  Processes of Lake Acidification.  National Academy
   Press, Washington, DC. 11 pp.
                                                 10 .

-------
                                                                     CHAPTER 2:  ENVIRONMENTAL GOALS
Watershed  processes  regulating  base cation ex-
change and retention of atmospherically deposited
sulfur  are two primary  controls on surface water
acidification rates that must be  understood to al-
low projection  of  potential acidic deposition ef-
fects.  In addition, as discussed later in this report,
an expanding body of research shows that nitrogen
deposition often also can be an important compo-
nent and potentially a  primary cause of present
and future soil and surface water acidification ef-
fects  within  some regions  of  North  America.
Other   important   research  further  reveals  that
chronic nitrogen loading and ozone  are the two
most  important air pollutants adversely affecting
forest ecosystems in North America.7

Past assessments minimized considering the effects
of nitrogen  deposition as a  potential cause of sur-
face water  acidification, largely because nitrogen
is an  essential nutrient  for  many biological proc-
esses.  The frequent scarcity of nitrogen in many
environments, relative to other required  nutrients,
often  limits plant growth and  production of other
organisms.  Thus, because  nitrogen  is commonly
thought of as a fertilizer that often quickly incorpo-
rates  into organisms,  many researchers  held the
view, and some continue to do so, that an insuffi-
ciently small mass of atmospherically deposited ni-
trogen would typically remain, after the fertiliza-
tion effects of nitrogen were maximized, to acidify
soils and surface waters  or to cause adverse effects
to resident  plants  or  animals.  In fact,  nitrogen
deposition to some  areas, particularly parts  of the
West, occurs primarily as the non-acid, "buffered"
compound ammonium nitrate. Second, biological
demand for nitrogen is  also highly dependent on
such factors as tree species  composition,  forest
age, disease, fire, and land management practices.
This produces high geographical variability among
watersheds  and high  seasonal variability within
watersheds  and makes  it difficult  to  understand
and model potential watershed acidification proc-
esses associated with nitrogen deposition.

Two lines of evidence now suggest that we need to
more  thoroughly  address acidification  problems
caused by  nitrogen, regardless  of the  chemical
form of its deposition from  atmosphere.  First, in-
creasing evidence  reveals  that dry deposition  is
usually a significant, and sometimes the dominant
portion of total atmospheric deposition of both sul-
fur and nitrogen.  For example, across all sites in-
cluded  in a recent review, dry deposition ranged
from   9  to  59   percent  of  total  deposition
(wet+dry+cloud) for S, 25 to 70 percent for NO3,
and 2 to 33 percent for NH3.8 Thus, in many areas
nitrogen deposition especially attaches to foliage
primarily in dry chemical form (e.g., as nitric acid
vapor  or ammonium  nitrate),  rather than  with
deposition  in precipitation.  Most affected vegeta-
tion  can not readily  derive  nutrient  benefits  from
such deposits of dry nitrogen.  Because of this,
Taylor et al. (see  footnote  7) suggested  that the
analogies equating atmospheric nitrogen deposi-
tion to nitrogen  additions accompanying broadcast
fertilization are  often  inappropriate.  This is espe-
cially the case when considering  adverse acidifi-
cation effects from  dry deposition of nitric  acids at-
tached to forest  and other terrestrial vegetative sur-
faces.

Second,  both seasonal conditions  and physiologi-
cal processes produce limits  on  potential  maxi-
mum nitrogen use by terrestrial organisms  in many
low productivity watersheds. These limits restrict
the total mass of deposited nitrogen that can be in-
corporated  into organic matter  by the combined
plant and  microbial  growth  needs within water-
sheds.   When these needs are met, i.e., when  ni-
trogen  is no longer the limiting nutrient  for bio-
logical  production and  growth,  nitrogen  com-
pounds can increasingly accumulate through a wa-
tershed process termed nitrogen saturation.9   In
other words, when nitrogen inputs are non-zero,
many low productivity forests that  are not affected
by  widespread  management  or  natural distur-
bances have the potential to reach  nitrogen satura-
tion at some time.10  Moreover, increasing rates of
nitrogen loadings (e.g., through acidic deposition)
would  tend to shorten times required to  achieve
watershed nitrogen saturation and  increase the  ni-
trogen  leaching  losses expected  after saturation
occurs (see footnote 10).
7  Taylor, G.E., D.W. Johnson, and C.P. Andersen. 1994.  Air
   pollution  and forest  ecosystems:  a regional to global
   perspective. Ecological Applications 4:662-689.
8  Lovett, C.M. 1994  Atmospheric deposition of nutrients and
   pollutants in North America:   an ecological perspective.
   Ecological Applications 4:629-650.

9  Aber, J.D., K.J. Nadelhoffer, P. Steudler, and J.M.  Melillo.
   1989.  Nitrogen saturation in northern forest ecosystems.
   Bioscience 39:378-386.

10 Aber, J.D., J.M. Melillo, K.J. Nadelhoffer, J. Pastor, and R.D.
   Boone.  1991.  Factors controlling nitrogen cycling and
   nitrogen saturation in norther temperate forest ecosystems.
   Ecological Applications 1:303-315.
                                                  11

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
Beyond the biological growth processes,  potential
for nitrogen saturation can also be affected through
physical processes that incorporate and store   ni-
trogen within watersheds.   These processes  are
poorly understood and generally thought to be a
relatively minor component of the nitrogen cycle
in watersheds.  Potential for saturation also can be
extended due to microbial denitrification.  This
process, which  emits nitrogen gases from water-
sheds,  most  commonly  occurs  in  .anaerobic
(lacking oxygen) water-saturated soils and deep
aquatic sediments, and denitrification rates tend to
increase with increases in environmental  tempera-
tures.11   Disturbances produced  by forest man-
agement and  natural events, e.g., timber windfalls,
insect infestations, or disease infections, can effec-
tively either shorten  or lengthen times to  potential
watershed nitrogen  saturation, depending  on  the
nature of the disturbance.

As suggested  in the literature cited above, initial
symptoms indicating the  potential onset of water-
shed  nitrogen saturation can include higher nitro-
gen concentrations in plant foliage and forest litter,
faster litter  decay  rates,  enhanced  nitrification
rates, increased  soil  acidity, and increased loss of
nitrogen gases  from watershed  soils.   With  con-
tinuing saturation,  nitrogen compounds  may  in-
creasingly leach to  layers below the  root zones
and from watersheds through surface waters drain-
ages, principally as  NO3~.  Further, excess avail-
ability of NO3- in watersheds can lead to depletion
of base  cations and  surface water  acidification
through the same processes as those involving ex-
cess SO42-.  Overall,  other environmental factors
equal, times  to nitrogen  saturation tend to  be
shorter for forested  watersheds  with lower water
availability and higher nitrogen deposition rates.

Forest watersheds  in  diverse  regions of  North
America show clear symptoms of developing wa-
tershed nitrogen saturation.12  In particular, early
stages of saturation  have been  noted  for  surface
waters  and/or   watersheds  in  New  Hampshire,
Vermont, the Adirondack and Catskill  mountains
of New York, and West Virginia; advance stages of
saturation have appeared in several high  elevation
streams  within  the  Great  Smoky  Mountains in
Tennessee and North Carolina.13   Symptoms of
ongoing watershed nitrogen saturation have also
be reported for areas  of the Blue  Ridge Mountains
in North Carolina,14 Colorado  Rocky Mountains,15
California  mountain ranges down wind from Los
Angeles,16 and in  the Algoma district of northern
Ontario.17 Other data suggest that many European
forests  apparently  are  becoming  nitrogen  satu-
rated; additional emission  controls options  to pro-
tect  European  forests  and surface waters from the
detrimental effects of excessive nitrogen deposition
are being evaluated.18 Some analyses suggest that
to reverse trends of adverse effects in some  regions
of Europe, nitrogen deposition may need to be  re-
duced  by 85  percent.19   With  considering the
European  situation, however, it  is important to
note  that  nitrogen deposition rates  of  concern
there are many times greater than occurs to areas
containing sensitive resources in North America.
11 Campbell, R. 1977. Microbial Ecology.  John Wiley and
   Sons, New York.

12 Johnson,  D.W.  and  S.E.   Lindberg,  editors.    1992.
   Atmospheric Deposition and  Forest  Nutrient  Cycling:  a
   Synthesis of the Integrated Forest Study.  Ecological Series
   91.  Springer-Verlag.  New York, USA.
13 Stoddard, J.L.  1994.   Long-term changes  in watershed
   retention of nitrogen: its causes and aquatic consequences.
   Pages  223-284  in  L.A.  Baker,  editor.   Environmental
   Chemistry of Lakes and Reservoirs. Advances in chemistry
   series   number  237,   American   Chemical   Society,
   Washington, DC.

14 Aneja, V.P., and A.B. Murthy.  1994.  Monitoring deposition
   of nitrogen-containing compounds in a high-elevation forest
   canopy. Journal of Air  & Waste  Management Association
   44:1109-1115.

15 Baron, J.S., D.S.  Ojima, E.A.  Holland, and  W.j.  Parton.
   1994.  Analysis of nitrogen saturation potential in Rocky
   Mountain  tundra and  forest.   implications for aquatic
   systems. Biogeochemistry 27:61-82.

16 Fenn, M.E., and A. Bytnerowicz.  1993.  Dry deposition of
   nitrogen and sulfur to ponderosa and Jeffrey pine in the San
   Bernardino   National   Forest   in  southern  California.
   Environmental Pollution 81:277-285.

   Fenn, M.E., and M.A. Poth.  1994. Preliminary evidence of
   nitrogen saturation in  the San Bernardino  Mountains  in
   southern California.   Presented at the  16th  International
   Meeting for Specialists  in Air  Pollution Effects on Forest
   Ecosystems, Sept. 7-9, 1994, New Brunswick, Canada.

   Riggan, P.J.,  R.N. Lockwood, and E.N.  Lopez.   1985.
   Deposition and processing of airborne nitrogen pollutants in
   Mediterranean-type  ecosystems  of southern California.
   Environment Science and Technology 781-789.

17 Mitchell, M.J., N.W. Foster, J.P. Shepard, and I.K. Morrison.
   1992.  Nutrient cycling in Huntington Forest and Turkey
   Lakes  deciduous stands: nitrogen and  sulfur.  Canadian
   Journal of Forest Research 22:457-464.

18 Sullivan, T.J.   1993.   Whole-ecosystem  nitrogen effects
   research in Europe. Environmental Science and Technology
   27(8):1482-1486.

19 Freemantle, M. 1995.  The acid test for Europe.  Chemical
   and Engineering News.  May 1, 1995:10-17.
                                                    12

-------
                                                                       CHAPTER 2: ENVIRONMENTAL GOALS
Shifting roles between nitrogen and sulfur deposi-
tion appear both in regions of North America and
in Europe. Studies from Europe first showed that
increased deposition  of nitrogen compounds ap-
pear  to offset potential benefits from reduced
deposition of sulfur  compounds.20  Similarly in
North America, for example, lakes included in the
Adirondack  Long-Term Monitoring program have
shown long-term  temporal declines in concentra-
tion of SO42 "consistent with declining SO2 emis-
sions and  declining  SO42' concentrations  in  pre-
cipitation falling across the eastern United States.21
The decreased lake SO42~ concentrations have not,
however, coincided with a recovery in lake levels
of ANC.  Rather, ANC values have  continued to
decline in some lakes. This decrease accompanies
increasing concentrations of NO3~ in most  of the
lakes  monitored,  and  decreasing  atmospheric
deposition of  base cations.   To illustrate further,
soil NO3~ in  the Woods Lake watershed within the
Adirondacks was positively and SO42~ was nega-
tively correlated with  calcium and aluminum con-
centrations.  Thus, from recent data, NO3~ concen-
trations appear to exert greater relationship to spa-
tial and seasonal temporal patterns of calcium and
aluminum in the assessed surface waters.  Further,
NO3" concentrations appear to have  increased in-
fluence on soil acidification in this watershed than
does SO42" concentrations.22   The effect that de-
creasing deposition of base cations has on water-
shed acidification processes remains poorly under-
stood and requires additional research.

Additional research  is  needed  to evaluate more
completely the extent that soil nitrogen accumula-
tions  are now and may potentially become a pri-
mary cause of acidification in watersheds, not only
in the Adirondack Mountains, but in other regions
of the North America.  Recent research finds that
major uncertainties regarding the capacities of wa-
tersheds to  assimilate  nitrogen deposition  limit
abilities to project directly surface water  acidifica-
20 Grennfelt,  P. and H. Hultberg.  1986.  Effects of nitrogen
   deposition  on the acidification of terrestrial  and aquatic
   ecosystems. Water, Air, and Soil Pollution 30:945-963.

   Henriksen,  A.  and  D.F.  Brakke    1988.   Increasing
   contributions of nitrogen to the acidity of surface waters in
   Norway. Water, Air, and Soil Pollution 42:183-201

21 Driscoll, C.T., and R. Van  Dreason.  1993.  Seasonal and
   long-term temporal patterns in the chemistry of Adirondack
   lakes. Water, Air and Soil Pollution 67:319-344.

22 Geary, R.J., and C.T. Driscoll.  1995.  Forest  soil  solution:
   acid/base  chemistry  and  response to calcite treatment.
   Biogeochemistry (in press).
tion rates from acidic deposition, and  that com-
puter modeling will be vital to assess concerns re-
garding  watershed  nitrogen saturation.23   Addi-
tionally, short-term and seasonal  changes in depo-
sition and processing patterns often tend to mask
true long-term trends in both atmospheric and wa-
tershed  processes, including nitrogen  saturation.
Therefore, true long-term  trends are best  deter-
mined  through  consistently  implemented  and
regularly sampled monitoring programs.24

Of additional concern are  episodes of storm flow
or snowmelt runoff that can expose organisms to
short-term, acutely lethal, acidic  water.25 Episodic
events (described  in detail in the following section)
occurring during spring snowmelt often  tend to be
the most acidic and contain the highest concentra-
tions of inorganic monomeric aluminum, which is
highly toxic to fish.26 NO3~ tends to be  more  mo-
bile in watershed soils at this time of the year be-
cause most plants are dormant. This fact  and the
prevailing  cold temperatures through winter  and
early spring tend  to promote increasing NO3~ ac-
cumulations in soil and overlying snowpack. Espe-
cially during  these  periods,  snowmelt  and  storm
water runoff can flush NO3~ through the  watershed
at flow  rates that  exceed the assimilative capacity
of terrestrial plants to capture the  rapidly passing
nutrients. Cold water temperatures also slow the
ability of  aquatic organisms to  incorporate  the
newly added NO3~. As a result, NO3~ can be a sig-
nificant seasonal cause of episodic acidification in
surface waters in  some  regions, often occurring at
the most biologically significant time of year (i.e.,
during fish spawning and reproduction).
23 Taylor, G.E., D.W. Johnson, and C.P. Andersen. 1994.  Air
   pollution  and forest  ecosystems: a  regional to global
   perspective. Ecological Applications 4:662-689.

24 Likens, G.E.  1992. The Ecosystem Approach: Its Use and
   Abuse. Volume 3, O.  Kinne (editor), Excellence in Ecology.
   Ecological Institute, Germany.

25 Wigington, P.J., Jr., J.P. Baker, D.R. DeWalle, W.A.  Kretser, P.S.
   Murdoch, H.A. Simonm, J. Van Sickle, M.K. McDowell, D.V.
   Peck, and W.R.  Barchet.  1993.   Episodic  Acidification of
   Streams in the Northeastern  United States:  Chemical and
   Biological Results of the Episodic Response Project. EPA/600/R-
   93/190.   Office   of  Research  and   Development,  U.S.
   Environmental Protection Agency, Washington, DC.

26 Baker, J.P., and S.W. Christensen. 1991.  Effects of acidification
   on biological communities in  aquatic ecosystems.  Pages 83-
   106 in D.F. Charles (editor).  Acidic Deposition and Aquatic
   Ecosystems -  Regional Case  Studies.  Springer-Verlag,  New
   York, NY.
                                                   13

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
2.2.2  Episodic Acidification
Most remaining sections of this chapter primarily
focus on chronic effects to surface waters associ-
ated with long-term exposure to acidic deposition,
because much past research has emphasized proc-
esses leading to long-term chronic acidification. In
surface waters that have not completed processes
leading to chronic acidification or are in the proc-
ess of recovering from chronic acidification, the
largest impacts of acidic deposition most  com-
monly accompany episodic acidification. Episodic
acidification (temporary loss  of ANC) can occur
when pulses of low-ANC or acidic  waters enter
streams and lakes  as  the result of rainstorms or
snowmelt.  Acid  anions (i.e.,  sulfate  and nitrate)
that reach surface waters during these events may
originate  from  immediate deposition  or,  more
likely, may  be the  result of prior  atmospheric
deposition (i.e., the previous year) that has cycled
within the watershed and  is flushed from the sys-
tem during  the high storm flows.  Acids or toxic
aluminum compounds leached from soils may ac-
company the acid anions during  leaching.  Both
sulfates and nitrates originating from  atmospheric
deposition can  contribute significantly to episodic
acidification events.27  Episodic acidification  can
cause  lakes and streams that  have positive ANC
during most of the year to become acidic (ANC<0
ueq/l) and have  high  toxic aluminum concentra-
tions for periods of hours to days.

Preliminary  results (pending technical peer review)
of a recent  modeling  study compared projections
for the minimum observed ANC during the  worst
annual episodes to pre-episode ANC for lakes and
streams in the Northeast and  mid-Appalachian re-
gions.28  Data on which this study was based were
collected between the early-mid  1980s to Spring
1990.   (Appropriate data were not  available to
conduct  a  similar  analysis  for  streams of the
Southern Blue Ridge Province.) Because the sever-
ity of episodes is  influenced  markedly by deposi-
tion loadings that  occurred  during the  study pe-
riod, empirical  relationships  reported in  the fol-
27 O'Brien, A.K., K.C. Rice, M.M. Kennedy, and O.P. Bricker.
   1993.  Comparison of episodic acidification of mid-Atlantic
   upland  and  coastal plain  streams.   Water  Resources
   Research 29(9):3029-3039.

28 Church, M.R., ). VanSickle,  P.J. Wigmgton, Jr., and  B. J.
   Cosby, Jr.  1994. Combining process and empirical models
   to predict future episodic acidification of streams and lakes.
   Presented  at NATO Advanced Research Workshop on
   "Ecosystem Modeling: Delineating  the  Possible from the
   Impossible", Bishofsgrun, Germany, February 20-25, 1994.
lowing comparisons should be considered appli-
cable only  for the conditions  (e.g.,  deposition
loadings) during which the data  were collected.
That is,  these  calculated  empirical  relationships
should not  be extrapolated to  different  deposi-
tional loadings  occurring at other times.

Across the sample of lakes and streams studied and
over the time  period of this modeling study, ap-
proximately 70 percent of lakes within the target
population were  projected to  be affected during
the worst annual episode in the Adirondacks.  This
was about 3.5  times the  number of lakes in this
target population observed to be chronically acidic
during the NSWS survey.  Similarly, for streams in
the mid-Appalachian, approximately 30 percent of
the target population  was  projected as  likely to
become  acidic during the worst annual  episode
occurring over the  study  period.  This projection
was approximately  7 times the number of chroni-
cally acidic stream reaches found  for the target
population  in  this region during  spring index
sampling conducted by the National Stream  Sur-
vey.  Further decreases in  levels of acidic deposi-
tion in these regions would likely lead to decreases
in the number and severity of acidic episodes.

EPA recently completed its Episodic Response Pro-
ject.29  Major  conclusions from that  project in-
clude:

   * Acidic deposition  episodes,  evidenced by
     stream water containing elevated SO42~ and
     NO3- concentrations  during the  episodes,
     were a common  occurrence in the  study
     streams of all three regions  investigated
     (i.e.,   the   Adirondack   and   Catskill
     Mountains of  New York and the Northern
     Appalachian Plateau of Pennsylvania).

   » Acidic episodes were common in streams
     of  each   region wherever  and whenever
     ANC values were 50  ueq/l or less immedi-
     ately before the episode. When acidic epi-
     sodes occurred, they were accompanied  by
     depressed pH  levels and elevated concen-
     trations of inorganic monomeric aluminum
29 Wigington, P.J., Jr., J.P. Baker, D.R. DeWalle, W.A. Kretser, P.S.
   Murdoch, H.A. Sinonin, J. Van Sickle, M.K. McDowell, D.V.
   Peck, and W.R.  Barchet.  1993.  Episodic Acidification of
   Streams in the  Northeastern  United States:  Chemical and
   Biological Results of the Episodic Response Project. EPA/600/R-
   93/190.  Office  of   Research and   Development,  U.S.
   Environmental Protection Agency, Washington, DC.
                                                  14

-------
                                                                   CHAPTER 2:  ENVIRONMENTAL GOALS
   * Elevated  concentrations of SO42~ in  Penn-
     sylvania  streams  and of NO3~ in Catskill
     and Adirondack streams augmented natural
     processes during episodes to produce lower
     ANC  and pH and higher Alim levels than
     would  have  occurred  due  to natural
     processes alone.

   * Even when SO42~ and NC>3~ concentrations
     did not markedly increase during episodes,
     elevated  baseline concentrations of  SO42~
     in all regions and of NO3~ in the Catskill
     and Adirondack streams lowered minimum
     ANC  and pH below levels naturally ex-
     pected during episodes.

   * Fish in all three regions studied, exposed to
     low pH  and high Alim over longer periods
     of episodic exposure had higher short-term
     mortality   rates   and   showed  greater
     long-term adverse population-level effects.
     Time-weighted median  Alim concentration
     was the single best predictor of brook trout
     mortality found during  these studies. Fur-
     thermore,  the  authors  concluded  that
     stream assessments based solely on chemi-
     cal measures during low flow do not  accu-
     rately predict the status of fish communities
     in small streams.

   * The ability  of fish to avoid episodic acidic
     water conditions by moving to less affected
     waters only partially mitigated the adverse
     effects in small  streams. Such behavioral
     adaptations  were not sufficient  to sustain
     fish density or biomass at  the  levels ex-
     pected in the absence of acidic episodes.

   » Brook trout density  and biomass were not
     different   between   chronically   acidic
     streams and streams with episodes where
     ANC decreased to less than 0 ueq/l.   Rela-
     tive to streams that  did  not  episodically
     acidify, both density and biomass of  brook
     trout  were   significantly lower  in stream
     reaches that became episodically acidic.

This last point supports  the  hypothesis that epi-
sodic acidification can be  a  primary cause of ad-
verse effects to brook  trout (and other ecological
components) in acid-sensitive streams  (i.e., ANC<
50 ueq/l). These  episodic  effects potentially equal
those seen in chronically acidic  streams  (ANC<
0 ueq/l).  Indeed,  effects  from  severe  episodic
acidification (i.e., events  leading to ANC<0 ueq/l
in surface waters) are  likely the first source of bio-
logical  damage  to  most  aquatic populations and
communities inhabiting waters that  have become
chronically  acidified.  The continuing  ecological
effects from episodic events often blend with and
become indistinguishable from all other effects ac-
companying chronic acidification.  The findings
from  this study and from other analyses clearly
point to  the importance of considering potential
effects of both long-term chronic and  short-term
episodic  acidification when considering the effec-
tiveness  of  an acid deposition  standard  or stan-
dards.

For most regions of North  America at risk from
acidic deposition, the effects from nitrogen deposi-
tion on aquatic systems are more likely to remain
primarily  episodic in nature, except when water-
sheds move toward nitrogen saturation, and nitro-
gen increasingly becomes a direct cause of both
episodic  and chronic acidification. Currently, data
available for most regions are inadequate to exten-
sively assess episodic  effects  related to  nitrogen
deposition or to assess the potential for and rate of
watershed nitrogen  saturation. Furthermore, while
available  data on  episodic  acidification  may in-
crease, because of  the  difficulty, expense,  and
often  the risk involved in collecting data during
episodic  events (e.g., intensive  spring sampling in
high-elevation snowmelt areas),  data bases for as-
sessing episodic effects are  not likely to  become
comprehensive.   Nevertheless,  existing evidence
clearly show that both  SO42~ and NO3~  deposition
can have major influences  on both long-term and
short-term  (episodic) surface  water  acidification
processes.  Further,  the  relative importance  be-
tween these two chemicals in producing acidifica-
tion effects often varies among regions and seasons
within regions. Evaluating the effectiveness of and
options for  acid deposition standards  should in-
clude simultaneous consideration of both  acidifi-
cation causes.

2.2.3  Cumulative Loading Effects
Deposited sulfur and nitrogen can be incorporated
into watersheds over the  long term through a vari-
ety of physical, chemical,  and biological proc-
esses. Current evidence suggests that the principal
dynamic   mechanism of concern with  regard to
watershed sulfur retention is physical/chemical in
nature (i.e., adsorption of inorganic sulfate). On
the other hand,  the principal  dynamic storage
mechanism  of nitrogen  retention  in watersheds
appears to be biological  in nature. Non-biological
removal  mechanisms also may  incorporate some
nitrogen into long-term storage within some water-
sheds, but this process is poorly understood.
                                                15

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
Overall, the  dynamics  of sulfur  adsorption  and
desorption and nitrogen retention over landscapes
of different scales (watersheds to regions) can vary
significantly.  Considerable  uncertainty  exists in
understanding these dynamics. Despite this uncer-
tainty, available  results strongly indicate that time
is critical in defining sensitivities  of  resources to
inputs of both sulfur and  nitrogen. Whereas most
watersheds can assimilate considerable quantities
of both chemicals without significant adverse ef-
fects, their assimilative capacities are finite. This is
especially  the case  in watersheds holding  acid-
sensitive aquatic resources,  a  primary focus of
concern in this report. Watershed  assimilative ca-
pacities vary  with how  rapidly deposited chemi-
cals are assimilated  and  the  time  over which  re-
peated deposition events impair these abilities. In
other words,  there  are  varying  deposition  fre-
quencies, rates, and durations when watershed as-
similative  capacities reach  saturation.  This  is
sometimes called steady state, the  point when the
output (loss) of a  substance from a watershed (e.g.,
sulfur leaving a watershed in the form of SO42~ in
stream flows)  equals its  input (e.g., as sulfur-con-
taining  compounds in deposition)  on  an annual
basis.

Although sulfur deposition over the long term can
lead to  equilibrium or steady state in watersheds,
similar steady-state conditions for nitrogen deposi-
tion are likely to be much less common. This is
because nitrogen  uptake dynamics  are  affected
much more by  biological changes within water-
sheds, such as forest cutting and regrowlh, fire, in-
sect infestation, disease,  as well as natural vegeta-
tive  succession.  Despite  these  factors, nitrogen
saturation does appear to occur in some systems,
as  discussed  above  in   Section  2.2.2.   Conse-
quently, assessing the history of  both sulfur and ni-
trogen  deposition   is  important  in   assessing
long-term regional effects attributable to cumula-
tive loadings by acidic deposition.  Similarly, a re-
source's or region's current  sensitivity to  acidic
deposition  also needs to be evaluated  with respect
to the historical deposition patterns and responses.

Many regions with ample buffering capacity and
remaining capabilities to retain inputs  of sulfur and
nitrogen may benefit little from future  decreases in
acidic deposition.  Other regions facing imminent
depletion of  their buffering or adsorption capaci-
ties, however, would likely be highly responsive to
decreasing deposition rates. The DDRP, discussed
in Section  2.5.2, provides a  useful beginning  for
understanding underlying  relationships  and  defin-
ing  remaining uncertainty about dynamics  of at-
mospheric sulfur deposition in  watersheds within
three regions of the eastern United States. The Ni-
trogen Bounding Study  (NBS),  discussed in Sec-
tion 2.5.3, provides additional useful results to im-
prove our understanding of the  influence of nitro-
gen  saturation on watershed processes affected by
combined sulfur and  nitrogen  deposition within
the three regions studied in the DDRP.

2.2.4 Recovery of Acidified  Ecosystems
Acidified  ecosystems can show signs of recovery
following  reductions  in  acidic  deposition  rates.
Benefits  have  been demonstrated  in (1) regions
where major local source emissions have been re-
duced, (2) experimental aquatic and terrestrial sys-
tems where applied doses of acids  have been re-
duced or  ended, (3) limed watersheds and surface
waters, and (4) model projections. Varying degrees
of  successful  recovery  in  communities of  mi-
crobes, algae,  higher plants,  invertebrates,  fish,
and  amphibians were  noted in the 1990 NAPAP
studies.30  Degrees  of  ecological  recovery varied
among the species, groups, and studies reviewed.

Recovery  rates depend primarily on three factors:
(1) rates of reduction in  emissions and deposition
of SO42"  and  NO3-;  (2) ongoing acid  retention
processes  in  terrestrial  environments,  including
sulfate adsorption and base cation weathering; and
(3) time lags caused  by delayed biological process
responses. In some  instances, significant  lags are
involved  or  irreversible changes have occurred.
The  influence  of  such lags accompanying  slow,
gradual chemical improvements are  well   illus-
trated in a series of studies of Ontario lakes follow-
ing  reduced emissions  from  the  Sudbury  area
smelters.31

Mitigation strategies that attempt to restore ecosys-
tems without reducing deposition  (e.g., liming) are
only partially successful  in  restoring water quality
and  recovering biological   populations.  In  fact,
rarely will  distressed  ecosystems return  to  their
predisturbance condition after the cause of the dis-
30 Baker, J.P., D.P. Bernard, S.W. Christensen, M.J. Sale, J. Freda,
   K. Heltcher, D. Marmorek, L. Rowe, P. Scanlon, G. Suter, W.
   Warren-Hicks, and P. Welbourn.  1990.  Biological Effects of
   Changes in Surface Water Acid-base Chemistry.  NAPAP Report
   13.  In:  Volume II, National Acid  Precipitation Assessment
   Program, Acidic Deposition: State of Science and Technology.
   Superintendent of Documents, Washington, DC.

31 Keller, W.,  J.R.  Piblado, and ).  Carbone 1992.  Chemical
   responses of acidic lakes in the Sudbury,  Ontario  area to
   reduced smelter emissions,  1981-1989.  Canadian Journal of
   Fisheries and Aquatic Sciences 49 (Suppl. 1 ):25-32.
                                                  16

-------
                                                                     CHAPTER 2:  ENVIRONMENTAL GOALS
turbance has been removed, because the complex
ecological interrelationships among predisturbance
species are rarely the same following disturbance
of the system and its recovery. When natural evo-
lutionary and successional regimes of predisturbed
systems are disrupted,  competition  for nutrients
and other habitat resources and predatory relation-
ships among  species  in  recovering  systems have
subtle to substantive differences from their predis-
turbance  relationships.  Thus,  restoration  to the
"predisturbance condition"  is  not always possible,
nor is it necessarily an appropriate goal.32 Alterna-
tively, rehabilitating or rejuvenating selected  at-
tributes or functions may be all that is required for
restoration to be deemed successful.  For example,
some  have argued  that  short-term  mitigation
measures such as liming a specific number of lakes
may be an appropriate, albeit  temporary, step.

A  recent work group  review  and assessment con-
cluded  that uncertainty  remains concerning the
definition  of appropriate measures of reversibility
and  recovery  for  acidified ecosystems.33  Differ-
ences exist particularly  between   setting   goals
based on  human-centered objectives  (e.g.,  fish
production  for human use)  versus more intangible
ecological  and conservation purposes. Further,  as-
sessment of ecosystem recovery following deposi-
tion  reductions can be obscured by other  envi-
ronmental  perturbations  such as climate change
and modified land-use practices.

Importantly, beyond  this uncertainty,  this  work
group also concluded  that changes in abiotic (i.e.,
chemical and physical) responses of ecosystems to
acidic deposition are reversible when  acidic depo-
sition is decreased. As abiotic environmental con-
ditions improve, many biological components of
the ecosystem  also will progress, sometimes rap-
idly, toward recovery.  But notable delays in this
biotic (i.e., biological) recovery can  occur,  often
depending on the rates at which emissions are  re-
duced and on watersheds factors  affecting internal
reposes to  the remaining deposition.    Further,
some irreversible changes may not recover.
32 Cairns, j.,  Jr.   1989.  Restoring damaged ecosystems-  Is
   predisturbance condition  a  viable  option?   Environmental
   Professional 11:152-159.

33 Dise, N, W. Ahlf, C. Brahmer, B.J. Cosby, J. Fott, M. Hauhs, I.
   Juttner, K.  Kreutzer,  C.C. Raddum,  and R.F. Wright  1994.
   Croup Report:   Are  Chemical  and  Biological  Changes
   Reversible? Pages 275-381 in C E.W. Steinberg and R.F. Wright
   (editors) Acidification of Freshwater Ecosystems: Implications
   for the Future.  J. Wiley and Sons, New York, NY.
2.3  CHARACTERIZING RESOURCES AT RISK FROM
     ACIDIC DEPOSITION
Relationships of resources to acidic deposition de-
pend on  two characteristics:  resource sensitivity
and  acidic  deposition exposure rates. Simultane-
ously  considering  regional  distributions  of both
characteristics allows assessments of risk potential
produced by acidic deposition over discrete geo-
graphic regions. This approach helps to define re-
gional need for and effectiveness of acid deposi-
tion  standards.

In this assessment, sensitivity is an inherent attrib-
ute of an individual resource that increases its sus-
ceptibility to likely adverse effects due to  acidic
deposition.  Exposure is determined  by the deposi-
tion  intensity,  frequency, duration, and  specific
times that acidic deposition falls into an area. Risk
is the probability that exposure to potentially haz-
ardous  environmental   conditions  produced  by
acidic deposition will exceed the  tolerance level
for a sensitive resource  and cause an  adverse ef-
fect.  For any sensitive resource to  be at high risk
from  any hazardous  substance  or  environmental
condition, it must have a high probability of being
sufficiently exposed to the substance or condition,
such that its inherent ability to tolerate the change
will be exceeded and harmful effects will likely re-
sult.34   Because  environmental resources  have
ranges of sensitivities and risks to potential effects
caused  by  acidic  deposition,  resources  having
equivalent sensitivities can  have different  risk po-
tential for effects depending on where they are lo-
cated.

The next two subsections review in  more detail the
concepts of resource sensitivity and  risk as they
apply to acidic deposition. Understanding these
concepts is essential for determining

   *  Locations containing sensitive  resources at
     risk,

   *  Which  sensitive resources may be the ap-
     propriate primary focus of protection,

   *  Appropriate environmental assessment  in-
     dicators, and

   *  The extent of protection afforded.
34 This discussion of sensitivity and risk is consistent with the
   concepts  presented  by the  Risk  Assessment Forum's
   Framework  for  Ecological Risk Assessment.   EPA/630/R-
   92/001 U.S. Environmental Protection Agency, Washington,
   DC
                                                  17

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
2.3.1   Defining Sensitive Resources
There are different types of sensitivities to acidic
deposition,  and  a resource  can be  insensitive to
one effect while being sensitive to others.  For ex-
ample, watersheds in a region with highly alkaline
surface waters may  begin to leak nitrogen into its
surface waters. This  could  lead to  no change in
surface water acidity, but could lead to significant
increases in  eutrophication  downstream.  Conse-
quently, the term sensitive can be imprecise and
confusing.   It can be  used  to  describe different
scales of resolution for different resource units
(e.g.,  extent of landscape areas, water chemistry
characteristics, or species groupings), different de-
grees of resource sensitivities,  and different time-
scales at which resources experience effects. Also,
each  use has different  scientific and policy impli-
cations. Such differences affect each potential cri-
terion used  in determining  acid deposition stan-
dards. Thus, when considering  potential resource
effects and  risks, it is important to  carefully define
the specific ecosystem components within the re-
gion and specific concerns regarding  the sensitivity
being addressed.

An early MAS  report  indicated  that  lakes and
streams with  alkalinity of  200 ueq/l  or less are
sensitive and subject to damage at moderate acidic
deposition  rates, whereas surface waters with  al-
kalinity of 40 ueq/l or less are critically sensitive to
such  effects.35  Although   alkalinity or ANC (as
noted  in Section 2.2,  these  terms are essentially
synonyms)  is an important response  indicator of
potential surface water sensitivity, it is not the only
relevant response indicator  of  sensitivity.  For ex-
ample,  the  presence or absence of  acid-sensitive
fish, invertebrates, algae, and higher plant species
are other relevant indicators of  potential sensitivity
and  acidification problems  in  lakes and streams.
Further, knowing the ANC of surface or  ground
waters provides little indication of the actual sensi-
tivity of neighboring terrestrial  resources.  For ex-
ample, injury to red spruce foliage attributable to
acidic deposition typically has little direct relation-
ship to the ANC of neighboring  soils or  waters.
Consequently, when there is a need to assess po-
tential effects of acidic deposition on terrestrial re-
sources or  ecosystems, assessments  often  need to
consider other parameters or indicators of sensitiv-
ity in addition to ANC.
Because numerous natural phenomena and proc-
esses influence  the sensitivity  and potential  risk
status of resources, interpretation and projection of
receptor responses to  acidification  are difficult.
Factors that should be carefully  evaluated when
assessing needs  for acid  deposition  standards  in-
clude naturally  occurring organic acidic systems,
annual  and seasonal variabilities in  precipitation,
and  related climatic variability.  Land management
and  resource use practices (e.g., changes in fishing
pressure, point  and nonpoint nutrient discharges,
mining runoff, and other watershed activities) also
potentially confound interpretation of acidification
sensitivity and effects. The types of effects caused
by  many  of  these  factors  are  summarized  in
Exhibit 2. The   list, although  incomplete, shows
that  a considerable  matrix of  factors interact to
determine  the  potential  sensitivity of  individual
surface  waters,  watersheds,  and   the   natural
resources they  contain. These  interactions cause
differences in sensitivity  and responses to acidic
deposition   among  resources  within  individual
watersheds and  among adjacent watersheds. Many
factors summarized in the exhibit are discussed in
greater detail in  subsequent sections.

Most factors presented in Exhibit 2 that can  in-
crease the sensitivity of  watersheds to acidifica-
tion, also usually correlate with increasing poten-
tial for watersheds to saturate with nitrogen, which
may lead to concurrent increases in surface water
acidification effects caused by nitrogen deposition.
For example, older forests are often less efficient in
using deposited  nitrogen and,  consequently,  can
leach greater quantities of potentially acidifying ni-
trogen compounds into surface drainage waters.36

Exhibit 2 also suggests  a focus on potential acidic
deposition  effects  linked  to  terrestrial soils  and
aquatic resources.  Such a focus  is not surprising,
because most acidic: deposition  eventually flows
through  soils and into aquatic systems. Responses
by these resources to acidic deposition are clear,
well understood, and  in many  cases  well docu-
mented. Therefore, most of the following discus-
sion  concentrates on  concerns  associated  with
environmental changes in soils,  lakes, and streams.
 35 National Academy of Sciences.   1983.   Acid  Deposition:
   Atmospheric Processes in Eastern  North  America.  National
   Academy Press.
36 Mitchell, M.J., N.W. Foster, J.P. Shepard, and I.K. Morrison.
   1992.  Nutrient cycling in Huntington Forest and Turkey
   Lakes  deciduous stands:  nitrogen  and  sulfur.   Canadian
   Journal of Forest Research 22:457-464.
                                                  18

-------
                                                                     CHAPTER 2:  ENVIRONMENTAL GOALS
                    EXHIBIT 2. PRINCIPAL WATERSHED AND SURFACE WATER CHARACTERISTICS
                         THAT INFLUENCE RESOURCE SENSITIVITY TO ACIDIFICATIONa
Category
Bedrock geology
Soils:
Buffering capacity
Depth
SO42~ adsorption
SO42~ reduction
NO3~ retention
Topography
Elevation
Watershed to surface water area ratio
Lake size
Lake flow characteristics
Flushing rate for drainage lakes
Watershed vegetation and land use:
Dominant vegetation
Cultural influence
Forest management
Water quality:
Alkalinity/ANC
SO42' reduction
Trophic status
Humic substances
Sphagnum moss
Climate/meteorology.
Precipitation
Snow accumulation
Growing season
Alkaline dusts
Increased Sensitivity
Resistant to weathering
(metamorphic, igneous)

Lower potential
Shallower
Lower potential
Lower potential
Lower potential
Steep-sloped
Higher
Lower
Smaller
Seepage
Higher
Coniferous
Forested
Reforestation
Lower (< 200 ueq/l)
Lower potential
Highly oligotrophic
Lowest concentrations
Present
Higher
Higher
Shorter
Lower
Decreased Sensitivity
Easily weathered
(sedimentary, calcite containing)

Higher potential
Deeper
Higher potential
Higher potential
Higher potential
Shallow-sloped
Lower
Higher
Larger
Drainage
Lower
Deciduous
Agriculture, municipal
Deforestation
Ffigher (> 200 ueq/l)
Higher potential
Less oligotrophic to eutrophic
Higher concentrations
Absent
Lower
Lower
Longer
Higher
      Modified from Marcus, M.D., B.R. Parkhurst, and F.E. Payne. 1983. An Assessment of the Relationship
      among Acidifying Depositions, Surface Water Acidification, and Fish Populations in North America.
      EA-3127, Volume 1, Final Report. Electric Power Research Institute, Palo Alto, CA.
Wherever possible, however, discussion  is punctu-
ated with summaries and highlights of potential re-
lationships of acidic deposition to other terrestrial
resources.

Our accumulated knowledge indicates that sensi-
tive resources can be defined over multiple ranges
of temporal,  geographic, geochemical, and  bio-
logical  categories  and scales.  Also, aquatic re-
sources can  be  sensitive  to episodic short-term
acidification, chronic  long-term acidification,  or
both.  Fundamental factors  and  attributes associ-
ated with differences in sensitivity include:

  * ANC of surface and ground waters;

  * Supply of base cations from bedrock and
    soil particle weathering;
                                                 19

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
  * Supply of base cations and buffering by soil
     solutions;

  * Physical  and chemical adsorption, desorp-
     tion, and reduction of sulfate;

  * Biological  assimilation  of nitrogen within
     watersheds; and

  * Ability of  sensitive  resident organisms  to
     modify physiological processes or behavior
     patterns or otherwise escape in response to
     habitat  changes  attributable  to  acidifica-
     tion.

Thus, although  considering levels  of  ANC in as-
sessing sensitive resources is  important, consider-
ing the presence and possible implications of other
indicators  of  potential sensitivity  (including  the
occurrence of threatened and endangered species)
is often equally important. Appropriate definitions
of sensitivity may therefore vary  with  precise pol-
icy questions being asked.  When the primary con-
cern is protecting water quality and most acid sen-
sitivity species  in sensitive surface waters,  ANC
often is  a useful response indicator. If concern is
broadened to include the potential sensitivity of all
natural resources (e.g., forest tree species), the ap-
proach used to classify sensitivities also should be
expanded.

In  Congressional discussions regarding the  man-
date for  an acid deposition  standard  study, Con-
gress followed  the  example of the  NAS  report
noted above in distinguishing between resources
that are  "sensitive" and those that are "critically
sensitive" to  the effects of acidic deposition.  Re-
source sensitivity occurs on  a continuum. Conse-
quently,  rather than refining assessments in the fol-
lowing sections  to distinguish among responses for
subcategories of sensitive  resources (i.e, sensitive
versus critically sensitive), it  is more  valuable to
assess the degree to which sensitive resources, in
general,  are exposed to different levels of risk from
acidic deposition across  different  geographic ar-
eas. As such,  the term critically sensitive resource
is  not used in this report. Much of the following
discussion does, however, focus  on  sensitivities
and responses for surface waters  projected by  EPA
model analyses  for lakes and streams  having ANC
of 50 ueq/l or less, a value that  approximates the
alkalinity value of 40 ueq/l considered by NAS and
Congress to distinguish between their two sensitiv-
ity  groupings of concern. Therefore, discussions in
the following section  regarding acidification  rela-
tionships  in  lakes  and  streams  with ANC of
50 ueq/l or less can generally be interpreted as ap-
plying to "critically sensitive" resources.

2.3.2  Identifying Resources at Risk
Sensitivity, as noted above, is only one determi-
nant of potential risk. For a resource to be at risk, it
must be sensitive to a potential stressor and must
have an  actual or reasonable possibility of expo-
sure to  the stressor in  a magnitude sufficient to
cause an  effect of concern.37 Sensitive resources
are at low risk when located where acidic deposi-
tion loads are currently below and are projected to
remain below thresholds likely to cause significant
effects. For  example, NAPAP studies reviewed in
Appendix A generally  indicate that many surface
waters in  western  North America are likely to be
more sensitive to acidic deposition than are similar
resources in eastern North America.  Because cur-
rent deposition  levels in the West are generally
below thresholds that produce  long-term surface
water acidification,  however, the  present risk to
these resources from chronic  acidification is  low.
If the  intensity  of western deposition  increases,
chronic acidification effects in the West might ex-
ceed those in the East. Consequently, the potential
for  high future  risk to  sensitive western resources
remains a concern.

Risk assessment, therefore, must  address not only
whether a location now receives sufficient acidic
deposition to produce significant effects, but also
the likelihood that the intensity and composition of
the deposition may change in the  future, thereby
changing future exposure and  potential risks to re-
ceptor resources. Consequently, it  is necessary to
determine what residual risks to sensitive resources
remain after implementing emissions controls re-
quired by the CAAA, where these risks may be lo-
cated,  and  their significance.  For example,  EPA
modeling analyses indicate the degree to which
sensitive resources will  be protected when current
Title IV requirements are fully implemented  (i.e.,
emissions reductions of 10 million tons  SO2 and
2 million  tons NOX). But, how might sensitive re-
sources benefit from  further decreases in deposi-
tion rates or implementation of a deposition stan-
dard? How might these benefits change with vary-
ing  additional  reductions  of  sulfur  or nitrogen
deposition?  How do possible  changes in sulfur or
nitrogen  retention within watersheds affect these
37 Risk Assessment Forum. 1992. Framework for Ecological Risk
   Assessment. EPA/630/R-92/001.  U.S. Environmental Protection
   Agency, Washington, DC.
                                                 20

-------
                                                                       CHAPTER 2:  ENVIRONMENTAL GOALS
possible relationships?  The  DDRP and NBS, as
discussed in Sections 2.5.2 and 2.5.3,  begin to an-
swer these questions.   Another  question,  outside
the scope  of this report, is whether a risk of im-
paired productivity exists for resources in  regions
not sensitive to harmful effects from  acidification
(such as forests managed for timber production
and agricultural  crops)  due to fertilization effects
of high sulfur and nitrogen deposition?

Resources  potentially sensitive to and at risk from
acidic deposition occupy vast expanses of water-
sheds and forests, multitudes of lakes,  and miles of
streams in  regions scattered across North America.
A complete census  of  these  resources would be
prohibitively time  consuming and costly. Instead,
assessments of locations  containing  sensitive  re-
sources must depend on statistically based surveys,
using proven methods to characterize populations
within an estimated  probability of error. Regional
and resource priorities for an acid deposition stan-
dard or  standards  should be based on such sur-
veys. Information  summarized  in Section 2.4  re-
views survey results  useful in targeting geographic
regions at risk and identifying sensitive resources.

2.4  IDENTIFICATION OF RESOURCE AND
     REGIONAL PRIORITIES

2.4.1  United States
Scientific information from the extensive research
efforts supported and reviewed by NAPAP directly
apply to setting needs and priorities for protecting
resources and regions sensitive to  acidic deposi-
tion. Appendix A summarizes the  major conclu-
sions from  10 of NAPAP's  State  of Science and
Technology  reports.  This section  synthesizes gen-
eral  findings from these  reports.38 Also  summa-
rized here  is additional  information from more re-
cent research regarding identification of  regions
sensitive to the effects of acidic  deposition  in the
United  States39  and  Canada,40  as well as  from
   This review is primarily drawn from conclusions presented by
   P.M. Irving (editor).  1991.  Acidic Deposition: State of the
   Science and Technology - Summary Report of the U.S. National
   Acid Precipitation Assessment  Program.   National  Acid
   Precipitation Assessment Program, Washington, DC.

39 The primary source for this additional summary information is
   NAPAP. 1992. Report to Congress. National Acid Precipitation
   Assessment Program, Washington, DC.

40 Brydges, T.G.   1991.   Critical  loads, reversibility  and
   irreversibility of damage to ecosystems.   Pages 245-260  in
   Electricity  and the Environment,  International Atomic Energy
   Agency, Vienna, Austria.
more recent research  assessing the  sensitivity of
individual resources.

The NAPAP studies, which provide  much of the
best information currently available for the United
States, contain  clear implications for identifying
resources most at risk from atmospheric deposition
of acidic compounds. Of all  effects to environ-
mental resources from acidic deposition, the scien-
tific community best understands changes in lakes,
streams,  rivers,  and soil chemistries.  The rate and
extent of acidic  deposition effects  on other  re-
sources are  less clear.

  1. FOREST SOIL CHEMISTRY EFFECTS: In the  east-
     ern United States, concentrations of  sulfur
     in forest soils generally follow trends in sul-
     fur  deposition.  In some  regions, soil  con-
     centrations of calcium and magnesium are
     inversely related to sulfur deposition loads
     resulting in soil nutrient depletion.  Further,
     a recent  review reports that most calcium
     and magnesium in the soil of the spruce-fir
     ecosystem in the Northeast was  lost 20-40
     years ago due to acidic  deposition,  when
     deposition rates were increasing rapidly.41
     This review  also  reports  that, while   ex-
     perimental   studies    quantitatively   link
     changes in  soil  chemistries to tree and
     other plant responses,  similar studies  link-
     ing  acidic deposition effects on soils to ac-
     tual plant  effects  in  nature  remain  incon-
     clusive. It concludes  further, that  within
     forested regions, acidic deposition  primar-
     ily  exerts  its stress on nutrient cycling and
     aluminum  mobilization.  No systems pres-
     ently appear to be deficient in  necessary
     nutrients as a result of deposition and it  is
     unknown  if  they ever will  become  defi-
     cient,  even  with   accelerated  leaching
     rates.42 Many  important  studies are  con-
     tinuing, however. The most apparent influ-
     ence of soil chemistry responses attribut-
41  Brandt,  C.J.   1994   Acidic Deposition  and forest  Soils:
   Potential Changes  in  Nutrient  Cycles and Effects on Tree
   Growth.     Report  to   Watershed  Response   Program,
   Environmental  Research   Laboratory,  U.S.  Environmental
   Protection Agency, Corvallis, OR

42  Johnson,  D.W.   1992. Relationships among atmospheric
   deposition, forest nutrient status, and forest decline.  Pages
   577-580  in  D.W. Johnson  and  S.E. Linberg  (editors).
   Atmospheric  deposition and forest nutrient  cycling:  A
   synthesis of the Integrated Forest Study. Ecological Studies
   Volume 91. Springer-Verlag, New York.
                                                   21

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
     able to acidic deposition are seen in effects
     on surface waters (see Item 3, below).

  2.  FOREST AND AGRICULTURE CROP EFFECTS: De-
     veloping evidence  indicates  that  acidic
     cloud water, in combination with  other
     stresses, likely increases winter injury, and
     reduces tree vigor and growth, and causes
     crown damage and death to high-elevation
     red spruce forests in the United States, par-
     ticularly in the northern Appalachians and
     high-elevation regions of  the   Northeast.
     Evidence of acidic deposition involvement
     in the decline of red spruce in the southern
     Appalachians is less  substantial. Involve-
     ment of acidic deposition in the  decline of
     sugar  maples in parts of the northeastern
     United States and eastern Canada  has  not
     been demonstrated but cannot be ruled out
     on the basis of available information.  Re-
     cent information, in fact, indicates  an  ap-
     parent  improvement  in sugar maple tree
     health  since 1988,  with the exception of
     observed decline in health of the  Ontario
     maple.43 The vast majority of forests in the
     United States  and Canada  have not  de-
     clined. Some evidence suggests that lichen
     communities and chemistries may be use-
     ful early indicators of forest health effects.
     Ambient acidic deposition levels have  not
     been shown to  be responsible for  agricul-
     tural crop yield reductions.

  3.  SURFACE WATER  ACIDIFICATION:   Numerous
     lines of  evidence support  the  fact  that
     acidic deposition can  acidify surface wa-
     ters and  that acidification attributable to
     acidic deposition has  occurred in sensitive
     aquatic systems during this century  (see
     Appendix A).  Most  sensitive  lakes  and
     streams in the  United States—especially
     those that have current ANC of 50 ^ieq/l or
     more—probably have not experienced re-
     cent chronic declines in pH or ANC asso-
     ciated with acidic deposition.

  4.  REGIONS  CONTAINING    ACIDIFICATION  AF-
     FECTED  SURFACE WATERS: The National Sur-
     face Water Survey (NSWS) conducted  un-
     der the auspices of  NAPAP in 1934-85,
     identified  six "high-interest areas" contain-
     ing most  of  the surface waters surveyed
43 U.S. EPA. 1994.  U.S. Canada Air Quality Agreement Progress
   Report.
   (95 percent of the lakes and 84 percent of
   the stream  reaches) that were chronically
   acidified  as indicated by concentrations of
   inorganic  anions,   predominately  SO42',
   NO3-, and CK  These  areas include  the
   southwest  Adirondack  Mountains,  New
   England, mid-Appalachian Region, Atlantic
   Coastal Plain, northern  Florida  Highlands,
   and low-silica lakes in  the eastern  Upper
   Midwest. Historical evidence supports the
   premise that acidic deposition undoubtedly
   is related to surface water acidification in
   the  Adirondacks,  the  Pocono/Catskill
   subregion, mid-Appalachians, eastern por-
   tion of the Upper Midwest, the New Jersey
   Pine  Barrens, and,  to a  lesser extent,  the
   Florida panhandle.  (Other areas of the Mid-
   Atlantic Coastal Plain appear to be affected
   more by organic acidification and land-use
   activities such  as  acid mine  drainage.)
   Chronic acidification of western lakes from
   acidic deposition appears not to  have oc-
   curred. The following  subsections further
   describe several  of  the  regions  containing
   sensitive surface waters.

5.  CHARACTERISTICS  OF  WATERSHEDS   CON-
   TAINING SENSITIVE SURFACE WATERS: Surface
   waters  most sensitive to acidic  deposition
   are often located   in  watersheds having
   shallow acidic  soils with  rapid, shallow
   subsurface flows. Acidic  lakes and streams
   tend to occur in smaller watersheds and, in
   regions where significant elevation gradi-
   ents exist,  at the  higher elevations  (e.g.,
   watersheds less than 30 km2 and elevations
   greater than 300 m in the mid-Appalachian
   region and  the Pocono/Catskill subregion).
   It must be noted, however, that these rela-
   tionships are  derived from studies empha-
   sizing watershed responses to sulfate depo-
   sition. Other, primarily biological, relation-
   ships exist  where deposition  of acidifying
   nitrogen  compounds are a  significant or
   predominant concern.

6.  RESPONSES BY SENSITIVE AQUATIC SPECIES AND
   ECOSYSTEMS: Acid-sensitive species occur in
   all major groups of aquatic organisms, but
   most is known about responses by fish  and
   aquatic invertebrates.  In general,  sensitive
   aquatic  species  inhabiting surface waters
   that  have   low  calcium  concentrations
   (<100-150 ueq/l) begin  to be affected by
   acidification  processes  as pH  decreases
   below about 6.0-6.5 (Exhibit 3) and as inor-
                                                22

-------
                                                                       CHAPTER 2: ENVIRONMENTAL GOALS
                      EXHIBIT 3. CRITICAL pH FOR SELECTED TAXA IN LAKES AND STREAMS'* b
                                      6.5
                                             Critical pH Levels for Selected Aquatic Organisms
                                                  6.0          5.5           5.0
                                                                                       4.5
                                                                                                   4.0
        Yellow Perch

        Brook Trout

        Lake Trout

        Smallmouth Bass

        Rainbow Trout

        Common Shiner

        American Toad'

        Wood Frog'

        Leopard Frog'

        Spotted Salamander'

        Crayfish"

        Mayfly"

        Clam"

        Snail"
     From National Acid Precipitation Assessment Program. 1991.  1990 Integrated Assessment Report  NAPAP Of-
     fice of the Director, Washington, DC.
     Solid symbols for each type of organism are placed in favorable pH ranges; shaded symbols are placed in less
     favorable ranges. No symbol is placed in pH ranges  that generally do not support populations of a particular
     type of organism.
     Embryonic life stages.
     Selected species
44
     ganic  monomeric  aluminum  concentra-
     tions increase above 30-50 ^jg/l, especially
     at concentrations of dissolved calcium  less
     than 2 to  3 mg/l.   These NAPAP  conclu-
     sions  generally  followed  those   from  a
     slightly  earlier review of more than 300
     peer-reviewed publications  that  reported
     results from field and laboratory studies of
     acidic effects.44   That review concluded,
     across all aquatic taxa and systems  studied,
     the overall number and severity of reported
     adverse effects tend to increase as pH  de-
     creases below about 6.5 to 6.0 and as total
    Marcus, M.D., B.R. Parkhurst, J.P. Baker, C.S. Creager, T.S.
   Fannin, C.G. Ingersoll, D.R. Mount, and F ). Rahel  1986.
   An Evaluation and Compilation of the Reported Effects of
   Acidification on Aquatic Biota. Volume!: Compiled Data.
   EPRI EA-4825.  Final Report. Electric Power Research
   Institute, Palo Alto, CA.
aluminum concentrations  increase above
about 10 ug/l.  Further, adverse effects tend
to be  reduced  at  given adverse pH  and
aluminum levels as dissolved calcium con-
centrations increase above 1  to  2 mg/l; for
increases  above 6  to 8 mg/l, calcium ap-
pears to have little  additional benefit.  Both
reviews suggested  that changes  in  water
quality produced by  increased acidity tend
to affect aquatic species first  by decreasing
their ability to survive, reproduce, or com-
pete in  acidic  surface  waters. Such re-
sponses can  eliminate affected species and
reduce species richness (i.e., the number of
species living within  a surface water).  Such
changes typically occur first in affected sur-
face waters  during episodic  runoff events
(i.e., when storm water or snowmelt runoff
causes short-term  flushes of acutely toxic
water  chemistries  to enter receiving  wa-
ters). System-level  processes  such as com-
                                                   23

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
     position, nutrient  cycling, oxygen usage,
     and  photosynthetic  rate are fairly robust
     and are affected only at relatively high lev-
     els of acidity (e.g.,  chronic  pH  less than
     5.0-5.5).45

These findings lead to several working conclusions
regarding sensitive resources and regions at poten-
tially greatest  risk  from  acidification, and the
maximum degree of protection that may be neces-
sary. It is  on these resources, regions, and protec-
tion goals that considerations should focus regard-
ing potential necessity and benefits of acid deposi-
tion standards:

  1.  SENSITIVE RESOURCES  OF PRIMARY CONCERN:
     Considering that the  natural resources most
     sensitive to acidic deposition would exhibit
     the  strongest responses  and  provide the
     most conclusive evidence of effects, it is
     reasonable to conclude that the natural  re-
     sources most sensitive  to acidic deposition
     are aquatic systems and high-elevation red
     spruce  forests.  Therefore,  possible  future
     acid  deposition standards should likely first
     focus on  recovery and protection of sensi-
     tive aquatic  resources in  eastern  North
     America (including the region from Florida
     north and east into  southern Ontario).   A
     second focus may include red spruce for-
     ests in the northern Appalachians and  high-
     elevation regions  of  the Northeast.  These
     regions  hold  most of the natural  resources
     showing the greatest magnitude of past and
     ongoing impacts and have the highest on-
     going rates of acidic deposition.   Both fac-
     tors indicate the potential need for possible
     additional deposition control.   Protection
     of sensitive aquatic  resources should par-
     ticularly focus on lakes and streams located
     where  watersheds are smaller, have  shal-
     low acidic soils with rapid, shallow subsur-
     face flows, and are at higher elevations.

  2.  REGIONAL PRIORITIES FOR PROTECTION:  In the
     eastern  United States,  the 1990 CAAA and
     any  future acidic  deposition  controls are
     most likely to reduce the threats of  acidic
     deposition  to surface  water resources  in
     these regions: Adirondack Mountains, Po-
     cono     and     Catskill     Mountains,
45 Schindler, D.W.  1987.  Detecting Ecosystem Responses to
   Anthropogenic Stress.  Canadian journal  of  Fisheries and
   Aquatic Sciences 44(Suppl.):6-25.
   mid-Appalachian   Region,  the  Southern
   Blue Ridge Province, New Jersey Pine Bar-
   rens, northern Wisconsin and Michigan's
   Upper  Peninsula,  and,  possibly, northern
   Florida. The  first three regions  apparently
   are  now at continuing risk from acidifica-
   tion effects.  Additional  monitoring and  as-
   sessment to  evaluate whether  continuing
   acidic  deposition will  affect  sensitive  re-
   sources would be necessary  at all areas
   listed above,  as well as in parts of  Maine,
   New Hampshire,  Vermont,  Massachusetts,
   Connecticut,  and  Rhode  Island; northern
   Minnesota; parts of the Ozark Mountains,
   Ouachitas Mountains,  the Carolina Pied-
   mont, and the Atlantic  Coastal  Plain; and
   parts of the Rocky Mountains,  Sierra Ne-
   vada Mountains, and Cascade Mountains.

3.  PROTECTION GOALS FOR AQUATIC SPECIES: The
   biological effects of  inorganic  monomeric
   aluminum associated with  acidic deposi-
   tion  are minimized as the level of acidic
   deposition is  decreased and pH and ANC
   levels in sensitive waters are kept relatively
   high. Based on studies of  sensitive aquatic
   species reported   by NAPAP   and  other
   sources cited  above,  to  protect aquatic  re-
   sources in sensitive watersheds from the ef-
   fects of long-term, chronic acidification, a
   general goal is to maintain the pH of sensi-
   tive lakes above pH  6.0-6.5 and inorganic
   monomeric aluminum  below  30-50 ug/l.
   To protect these resources from  the poten-
   tial  effects of  episodic, acute acidification,
   surface water ANC should  be  maintained
   at or above 50 ueq/l. No single water qual-
   ity goal,  however, addresses all needs to
   protect  sensitive    watershed  resources.
   Goals to protect   resources may also ad-
   dress site-specific needs to maintain sensi-
   tive  species, species  of  special concern
   (e.g., listed threatened or  endangered spe-
   cies), and species richness  in these sensi-
   tive  watersheds. When establishing  protec-
   tion  goals and  objectives  for sensitive
   aquatic resources, this effort certainly must
   include recognition  and  allowances that
   pH  levels less than 6.0  and  ANC less than
   50 ueq/l  occur in some  naturally acidic
   (e.g.,  organically  acidic) surface  waters,
   and that levels of pH less than 6.0 can oc-
   cur naturally  in some  locations accompa-
   nying periods of  episodic stormwater and
   snowmelt runoff from relatively  unpolluted
                                                 24

-------
                                                                    CHAPTER 2: ENVIRONMENTAL GOALS
     deposition.  That is, the specific  environ-
     mental  objectives of any  acid deposition
     standard should accommodate the natural
     ranges of chemical  qualities occurring  in
     waters  in  the  environment.  Furthermore,
     they  may be  designed  to protect  those
     special  biological  communities evolved  to
     inhabit naturally acidic surface waters.

2.4.2   Qualitative Assessment of Sensitive
        Aquatic Resources in  Three  Regions
        of the United States
Most of the  rest of this chapter  discusses results
from EPA model analyses for three  case study re-
gions:  the Northeast (including  the  Adirondack
Mountains), the mid-Appalachian Region,  and the
Southern  Blue   Ridge  Province. Similar  model
analyses were not performed for other sensitive re-
gions of North America due to time and primarily
data limitations.  Instead, and in  addition to the re-
sults of the NAPAP National Surface Water Survey,
EPA supported  a recent review that assessed the
responses of aquatic resources to acidic deposition
in four other regions of North  America identified
in previous studies as holding sensitive  aquatic re-
sources: the  mountainous  western United States,
upper midwestern United States,  northern  Florida,
and eastern Canada.46  That  review addressed ef-
fects from acidic deposition, specifically sensitive
aquatic resources in each region. The approach
incorporated  key results  available  from past re-
search and  assessment efforts  in  North America
and Europe. The major conclusions derived during
this  review that  specifically  related to  the three
regions assessed in the United States are presented
below.  Conclusions  from  this  study  regarding
sensitive Canadian resources are  included in the
next section.  (Some  conclusions from this review
regarding  general deposition and response rela-
tionships duplicate the findings of other studies re-
ported above and are not repeated in this section.)

Western  United States
   * Most low-ANC lakes in the West  are con-
     fined primarily to  glaciated, higher eleva-
     tion, mountainous  regions.  These water-
     bodies  can  be generally consolidated into
     five lake populations, based on their  loca-
46 Sullivan, T.J., and ).M. Eilers.  1994.  Assessment of Deposition
   Levels of Sulfur and Nitrogen Required to Protect Aquatic
   Resources in Selected Sensitive Regions of North America.
   Final  Report.  Environmental Research Laboratory-Corvallis,
   U.S. Environmental Protection Agency, Corvallis, OR.
   tions  within  similar  geomorphic  units:
   (Dthe Sierra Nevada in California; (2) the
   Cascade Mountains in California, Oregon,
   and Washington; (3) the Idaho Batholith in
   Idaho and  Montana; (4) the  mountain
   ranges  of  northwestern  Wyoming;  and
   (5) the Rocky Mountains in Colorado.

»  Results of the NSWS indicate that although
   no  lakes  in  this  region  are chronically
   acidic, many lakes were found to have very
   low  ANC  (i.e.,  17  percent  have  ANC<
   50 ueq/l),  and should  be  considered sus-
   ceptible  to acidification,  particularly epi-
   sodic acidification,   should   atmospheric
   deposition loadings increase.

*  Watersheds in the alpine areas of these five
   regions generally include  broad  expanses
   of exposed bedrock, which is often highly
   resistant  to weathering, and  contain  little
   soil or vegetative cover to  neutralize acidic
   inputs. Consequently, these regions include
   a significant portion of the region's aquatic
   resources  that are  the most  sensitive  to
   acidic deposition.

*  Natural characteristics of these watersheds
   particularly predispose the surface  waters
   they contain  to  episodic  acidification ef-
   fects: (1) low-water  retention capacities of
   most watersheds; (2) ultra-low concentra-
   tions of base cations; (3) low ANC  in sur-
   face  waters throughout the year; (4) large
   snowpack  accumulations  and substantial
   base  cation   dilution   during   runoff;
   (5) frequent,  periodic,  heavy  rain  storms
   with high runoff events; and (6) short water
   retention times and  high flushing rates for
   most lake basins.

*  Nitrate  concentrations in  the majority  of
   the western lakes are virtually undetectable
   during the fall. The NSWS samples, how-
   ever, revealed that a relatively large num-
   ber of lakes in northwestern Wyoming, Si-
   erra  Nevada, and Colorado Rockies  con-
   tained high concentrations of NO3-. These
   concentrations were sufficiently high to in-
   dicate that many watersheds  in these re-
   gions may have little remaining capacity to
   assimilate excess NO3~ deposition.

*  No  extant data suggest that  lakes  in  the
   West have experienced chronic acidifica-
   tion. It is likely, however,  that episodic ef-
   fects have occurred  in some  lakes under
                                                 25

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
     some  current deposition regimes and that
     deposition  concentrations  of NO3; espe-
     cially  have caused small, chronic losses of
     ANC  in  some  high-elevation watersheds.
     As previously described, both nitrogen and
     sulfur  have  the  potential  to contribute to
     episodic acidification.

  * Fish population  data,  although very lim-
     ited, do not indicate a significant eflect due
     to acidification. However, trout species na-
     tive to Western lakes such as rainbow trout
     and cutthroat trout  are considered more
     sensitive to low pH and elevated aluminum
     than brook trout native to eastern lakes and
     streams.

  * Although  potentially  difficult   based  on
     wide  ranges in  precipitation volume, an
     event-based deposition  standard may be  a
     consideration to  adequately address the po-
     tential  effects  of episodic  acidification in
     the  West.  Another  consideration  which
     combines concerns for both total mass and
     maximum concentration of pollutants  de-
     posited,  is  an  acid  deposition  standard
     which  establishes  limits  based  on  the
     maximum allowable annual-weighted pol-
     lutant concentrations or based on total an-
     nually deposited chemical mass loading.

Upper Midwestern  United States
  * The Upper  Midwest is characterized by
     numerous lakes created by repeated glacia-
     tion, little topographic  relief, deep glacial
     overburden, and rarely exposed bedrock.
     In this region the primary aquatic resources
     sensitive  to  acidic deposition are  seepage
     lakes with  low base cation  concentrations
     that receive nearly all of their water inputs
     as precipitation  directly onto the lake sur-
     face. These lakes have generally long water
     retention times,  which  provide opportuni-
     ties for in-lake SO42' reduction and NO3~
     assimilation processes to neutralize most
     acidic inputs and to prevent the concentra-
     tion of SO42~ through evaporation.

  • The Upper Midwest has a large population
     of low-ANC lakes, but relatively few acidic
     (ANC<0 ueq/l)   lakes.  Paleolimnological
     evidence suggests that some of  these lakes,
     particularly in Michigan's Upper Peninsula,
     have  developed  low  ANC or  become
     acidic, consistent  with historic trends for
     sulfur deposition since  preindustrial times.
     It should be noted that land use changes
     and other human disturbances in the wa-
     tersheds have exerted greater influences on
     the acid-base chemistry in more sensitive
     lakes of this region than has acidic deposi-
     tion.47

  *  The NSWS indicated that 19 percent of the
     lakes in  this region have ANC<50 ueq/l
     (only 3 percent of that figure is ANC<0).
     Historical data are too limited to  determine
     the degree to which acidic deposition has
     impacted  fish  populations in this  region
     such  as yellow  perch, bass,  and  others.
     However,  lakes with  low pH in  northeast-
     ern Wisconsin and upper Michigan support
     fewer fish species than expected for their
     size and lake type.

  »  Concentrations  of  inorganic  nitrogen are
     uniformly  low in surface waters throughout
     this region. Most nitrogen is efficiently re-
     tained by terrestrial and aquatic organisms.
     Snowmelt has not been shown to provide
     any significant influx of NO3~ to these lakes
     because most  snowmelt water  percolates
     through the soil prior  to  entering surface
     waters, allowing terrestrial organisms to as-
     similate the deposited nitrogen.  Therefore,
     the key concern for this region is chronic,
     sulfur-driven acidification. If recent trends
     of decreasing sulfur deposition in the Up-
     per  Midwest were to  reverse, lakes with
     ANC near zero in this region may acidify.

Northern Florida
  *  Florida lakes are located on marine sands
     overlying   carbonate   bedrock;   where
     groundwater  interacts   with   the  deeper
     aquifer, surface water can be  highly alka-
     line.  Lakes that receive input waters only
     from shallow aquifers in highly weathered
     sands,  however, can be quite acidic and
     sensitive   to  acidic  deposition.  In fact,
     northern Florida contains one of the largest
     populations of acidic lakes in the  United
     States.  Seventy-five percent of the Panhan-
     dle  lakes  are acidic, as are 26 percent of
     the lakes in the northern peninsula.
47 Sullivan, T.J. 1990. Historical Changes in Surface Water Acid-
   Base Chemistry in Response to Acidic Deposition. SOS/T 11,
   National Acid Precipitation Assessment Program, Washington,
   DC. 212pp.
                                                 26

-------
                                                                     CHAPTER 2:  ENVIRONMENTAL GOALS
   * The  NSWS determined that approximately
     60 percent of the  acidic lakes in Florida,
     primarily in the northcentral peninsula, are
     acidic due to acidic deposition. Subsequent
     scientific study  however, suggest  that the
     role  of  the natural sulfate  bearing ground
     water, significant land use  changes  in the
     region,   and marine  sources  likely  have
     greater   effects  on  the   acidic  water
     chemistries of these systems than was pre-
     viously  estimated. Therefore, the extent of
     possible  water   quality  changes  due  to
     acidic deposition alone in  Florida cannot
     now be quantified, but is likely lower than
     the NSWS estimate. The best evidence that
     acidic deposition effects have altered the
     surface  water chemistry exists for the Trail
     Ridge   region   in   northeastern   Florida.
     Available data currently indicate that there
     has been no widespread biological damage
     due  to  acidic deposition within  the  sensi-
     tive regions studies in northern Florida.

2.4.3  Qualitative Assessment of Sensitive
       Aquatic Resources in Canada
The area  of  Canada  considered  to be  at greatest
risk from  acidification  (i.e.,  the  region  having
minimal ability to neutralize incoming acids  and
receiving  elevated deposition  of potentially acid-
forming chemicals), includes the  region  east of the
Manitoba-Ontario border and roughly south of 52
N latitude (near the southern limit of James Bay).48
(Implications to Canadian forests ,  as summarized
by  NAPAP, were reviewed earlier in this chapter.)
This area  contains more than 700,000 lakes cover-
ing  about   160,000km2   (excluding  the  Great
Lakes).  Extrapolation  of survey  information  indi-
cates that 14,000 lakes are presently acidic. Mod-
eling projections for  eastern Canada indicate that
at  least  an  additional  10,000 to 40,000 lakes
would become chronically acidic at 1985 deposi-
tion levels,  as  watershed  input-output  budgets
reach equilibrium over time with concentrations of
atmospherically deposited acid-forming ions.
Four  important relationships  primarily  influence
the surface water chemistry  of these  Canadian
lakes and their potential sensitivity to acidification:

  1. Because  of the predominance of  silicate
     bedrock,  thin overburden,  and  generally
     high precipitation volumes (approximately
     100cm/yr or greater),  nearly  all lakes in
     eastern   Canada  can  be   hydrologically
     characterized as  drainage  lakes.49  (The
     acidification  processes that  dominate seep-
     age lakes in the  upper Midwest and  Florida
     lack importance in most Canadian lakes.)

  2. Most glacially deposited soil covering east-
     ern Canada has  essentially  no  capacity for
     SO42'adsorption. Because there appears to
     be  no  significant geological  sources of
     SO42" in  this  region, limited adsorption ca-
     pacity  indicates that existing  SO42" levels
     are principally controlled by  atmospheric
     inputs. Note  that  most  glacially deposited
     soil covering eastern Canada  is  similar to
     the soils  in the northeast United States (i.e.,
     Adirondacks) in that they have  very  limited
     SO42" adsorption  capacity,  and  acidifica-
     tion is primarily controlled by  atmospheric
     inputs in both regions. 50

  3. Base cation production  of principally Ca2+
     and Mg2+ by primary weathering  or cation
     exchange in the  surrounding terrestrial  wa-
     tershed provide most of the ANC of  all Ca-
     nadian lakes. Thus, the variable mineralogy
     in glacial overburden surrounding the lakes
     dominates control  of subregional variability
     in sensitivity to acidification.

  4. Organic  acid anions appear to be  impor-
     tant in  many waters, particularly in  the ex-
     tensive  wetland  areas  of  Nova  Scotia,
     Newfoundland,  Labrador, and northwest-
     ern Ontario;  the occurrence of these  ani-
     ons, however,  are generally not the  pri-
     mary cause of acidity in all lakes with ANC
     of 0 ueq/l or less and pH less than 6.0.
48 Information presented in this section regarding sensitive aquatic
   resources in Canada, unless otherwise cited, comes from the
   summary of Jefferies, D.S.  1991.  Southeastern Canada: An
   Overview of the Effects of Acidic  Deposition on Aquatic
   Resources.  Pages 273-286 in  D.F. Charles (editor).  Acidic
   Deposition and  Aquatic  Ecosystems - Regional  Case Studies.
   Springer-Verlag, New York, NY.
49 Drainage lakes are lakes with permanent surface water inlets
   and,  usually, outlets.   Seepage lakes are  lakes with  no
   permanent surface water inlets or outlets.

50 Sullivan, T.). 1990. Historical Changes in Surface Water Acid-
   Base Chemistry in Response to  Acidic Deposition, SOS/T 11,
   National Acid Precipitation Assessment Program, Washington,
   DC. 212 pp.
                                                  27

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
Almost all known acidification related  losses of
lake trout (Salvelinus fontinalis), smallmouth bass
(Micropterus dolomieui), and walleye (Stizostedion
vitreum) in Ontario surface waters have occurred
in the Sudbury area and are related to emissions
from the Sudbury area smelters; complete or bio-
logically significant reversal of acidic conditions in
these waters,  however, may depend on continued
reductions in  emissions over  a wider  region.51
Concern also  continues  regarding probable epi-
sodic  influence of acidic  deposition on  Atlantic
salmon in tributary streams  along the  Atlantic
coast from Maine northward.52

2.5  ASSESSING PROTECTION  NEEDS AND
     RESOURCE RESPONSES IN THE CONTROL OF
     ACIDIC DEPOSITION
Determining the potential for  future benefits from
additional control of acidic deposition involves as-
sessing changes over broad geographical and eco-
logical scales. Projecting potential future  relation-
ships is best done using models to simulate possi-
ble future scenarios.  Simulation  models  are, in
fact, conceptualizations of the way things "work."
Due to the very simple character of their control-
ling process,  some models can be very precise and
can be verified repeatedly by comparisons of pro-
jections to subsequent observations (e.g.,  projec-
tions of solar  eclipses). In contrast, models of wa-
tersheds and  of surface water responses  to envi-
ronmental  perturbations  (e.g.,  acidic  deposition)
are much more difficult to design and test.

This section briefly reviews the use and constraints
of simulation models.  It presents  detailed results
from two major EPA modeling studies and an  ex-
51 Keller, W.   1992.   Introduction  and  overview  to aquatic
   acidification studies  in the Sudbury, Ontario, Canada, area.
   Canadian Journal of  Fisheries and Aquatic Sciences 49(Suppl.
   1V.3-7.

52 Lacroix, G.L., and  D.R.  Townsend.  1987.   Responses of
   juvenile Atlantic salmon (Salmon salar) to episodic increases in
   acidity in Nova Scotia rivers. Canadian Journal of Fisheries and
   Aquatic Sciences 44:1475-1484.

   Lacroix, G.L. 1989.  Ecological and physiological responses of
   Atlantic salmon in acidic organic rivers of Nova Scotia, Canada.
   Water, Air, and Soil Pollution 46:375-386.

   Lacroix, G.L.  1989. Production of juvenile Atlantic Salmon
   (Salmon salar) in two acidic rivers of Nova Scotia.  Canadian
   Journal of Fisheries and Aquatic Sciences 46:2003-2018.

   Lacroix, G.L., D.J. Hood, and J.A. Smith.  1995. Stability and
   microhabitat use by  brook trout and juvenile Atlantic salmon
   after stream acidification.   Transactions of the  American
   Fisheries Society 124:588-598.
tensive literature review  aimed  at  increasing the
understanding  of  how  acid-sensitive  soil  and
aquatic resources are affected by both sulfur and
nitrogen deposition. International and state effects-
based efforts to regulate acidic deposition are also
described.   Special concerns regarding spatial and
temporal issues of importance in developing a po-
tential deposition standard or standards are also in-
troduced.

2.5.1   Model Selection and Application
A major goal established for the Acid Deposition
Standard  Feasibility  Study  is  to assess  how
changes in  acidic deposition loading could affect
the chemistry of surface waters:

   * In a dynamic fashion;

   * Over regional scales;

   * For combinations of  sulfur and  nitrogen
     deposition;

   » In areas containing  potentially vulnerable
     lakes and streams; and

   * Where high levels of both sulfur and nitro-
     gen deposition exist.

The geographic region  selected by the Agency for
this study was the  eastern  United States, an area
containing lakes and streams with  low ANC and
receiving the highest combined levels of sulfur and
nitrogen deposition in the United States.

To address this goal, the assessment must be com-
pleted over the scale of regions.  Modeling and as-
sessment of possible effects from acidic deposition
on waters draining from  a single watershed (or a
few watersheds) has very limited  usefulness, be-
cause  extrapolation from  such  model  projections
to regional-scale implications is not reliable. That
is,  knowing what such a modeling result would
mean for the distribution of surface waters over a
region is not possible.  Thus, a regional modeling
approach is required.

Active work on modeling  simulations for the Acid
Deposition  Standard  Feasibility Study  began  in
early-mid 1992.  At that time there  had been pub-
lished only  a  single watershed-scale  model that
combined  above-ground forest processes with be-
low-ground  biogeochemistry—the   Nutrient  Cy-
cling Model (NuCM).53 The model includes both
53 Liu, S., R. Munson, D.W. Johnson, S. Gherini, K. Summers,
   R. Hudson, K. Wilkinson  and  L.F. Pitelka. 1992.   The
                                       (continued)
                                                  28

-------
                                                                         CHAPTER 2:  ENVIRONMENTAL GOALS
sulfur and nitrogen cycling and was developed as
an extension of the ILWAS model and as a part of
the activities of the Integrated Forest Study.54   At
that time, published reports of model applications
were limited to a few intensively studied sites and
then to estimating effects on leaching rates of base
cations of deposition of sulfur.

As indicated above, the Acid  Deposition Standard
Feasibility Study calls for regional  assessments on
surface water chemistry of the leaching of nitrate.
To complete such  an assessment requires both a
model  capable of  such  simulation  and the data
necessary to run the model.  The NuCM model re-
quires a variety of specific data inputs to simulate
the complex cycling of forest  nutrients (e.g.,  nitro-
gen).  Needed inputs  include information, for  ex-
ample, on vegetation  growth  rates, rates of  foliar
leaching  and exudation,  rates  of  organic matter
decay,  and   rates  of  nitrification.    In  early-mid
1992, high-quality data  on such rates  existed at
only a very few intensively studied  watershed sites
within the United States.  Sufficient necessary data
did not exist to run the NuCM model in a reliable
regional application. Therefore,  at that time,  the
goal of running a process-based watershed acidifi-
cation model that could  account for the effects of
both  nitrogen and  sulfur inputs and cycling and
that could be credibly used at regional  scales for
estimating  future   effects  on  lake   and  stream
chemistry was  not  attainable.  The possibility re-
mained,  however,  to  investigate  the   potential
bounds of nitrogen  deposition  effects on  surface
water  chemistry  in   combination  with  process
modeling of effects of sulfur deposition.  This was
accomplished  through   the   Nitrogen  Bounding
Study.55

The Nitrogen Bounding Study (NBS) was built on
model approaches and field sampling information
compiled during the Direct/Delayed Response Pro-
ject (DORP),56-57  a regional study that used  three
   Nutrient Cycling Model (NuCM): overview and application.
   Pages 583-609 in D.W. Johnson and S.E. Lindberg (editors).
   Atmospheric  Deposition  and  Forest  Nutrient  Cycling,
   Ecological Studies 91, Springer-Verlag, New York. 707 pp.

54 Johnson,  D.W., and S.E. Lindberg (editors).  Atmospheric
   Deposition and Forest Nutrient Cycling, Ecological Studies
   91, Springer-Verlag, New York. 707 pp.

55 Van Sickle, J., and M.R. Church.  1995. Methods for Estimating
   the  Relative Effects  of Sulfur and Nitrogen Deposition on
   Surface  Water Chemistry.  U.S.  Environmental  Research
   Laboratory, Corvallis, OR.

56 Church, M.R.,  K.W. Thornton, P.W. Shaffer, D.L. Stevens,
   B.P. Rochelle,  G.R. Holdren, M.G. Johnson, J.J. Lee,  R.S.
                                        (continued)
watershed  models to assess effects  produced by
sulfur deposition on lake and stream chemistry in
the  eastern United States.58   In  the DDRP,  the
Agency gathered watershed and soil data at hun-
dreds of  eastern U.S. watersheds to run multiple
models of watershed acidification  on the future ef-
fects of sulfur  deposition only in the  Northeast,
mid-Appalachians, and Southern Blue Ridge Prov-
ince of the United States.  Nitrogen deposition ef-
fects could not  be modeled at that time.   Water-
sheds modeled during the DDRP were selected as
a statistical subset of those surveyed by the  Na-
tional Surface Water Survey.59  Because  of this sta-
tistical basis,  each watershed  modeled during  the
DDRP held a statistical sampling weight represent-
ing  a  definable  proportion  of  the watersheds
within  each of the three study regions.    In  turn,
this  allows results from this modeling  to  be  ex-
trapolated over regional scales to project responses
by the target  population of interest (low-ANC  wa-
tersheds).  Not all of the three models used during
the  DDRP  could be successfully calibrated for all
of the sensitive watersheds in the  sample.  By far,
the  greatest  number  of  successfully calibrations
occurred   for  the  Model  of  Acidification  of
Groundwater in Catchments (MAGIC).60  Signifi-
cantly fewer  successful  calibrations  occurred  for
   Turner,  D.L. Cassell, D.A. Lammers, W.C. Campbell, C.I.
   Liff, C.C. Brandt, L.H. Liegel, G.D. Bishop, D.C. Mortenson,
   S.M. Pierson, and D.D. Schmoyer.   1989.  Direct/Delayed
   Response  Project:   Future Effects  of  Long-term  Sulfur
   Deposition on Surface Water Chemistry in the Northeast and
   Southern Blue Ridge Province.  EPA/600/3-89/026a-d. U.S.
   Environmental Protection Agency, Washington, DC. 887 pp.

57 Church, M.R., P.W. Shaffer, K.W. Thornton, D.L. Cassell, C.I.
   Liff, M.G. Johnson, D.A. Lammers,  J.J. Lee, G.R. Holdren, J.S.
   Kern, L.H. Liegel, S.M. Pierson, D.L. Stevens, B.P. Rochelle, and
   R.S. Turner.  1992. Direct/Delayed Response Project:  Future
   Effects of Long-term Sulfur Deposition on Stream Chemistry  in
   the Mid-Appalachian  Region of the  Eastern  United States.
   EPA/600/R-92/186.  U.S.  Environmental  Protection  Agency,
   Washington, DC.  384 pp.

58 Section 2.5.2 presents provides additional detailed discussion
   and reference to publications resulting from the DDRP.

59 Linthurst, R.A., D.H. Landers, J.M. Eilers, D.F. Brakke, W.S.
   Overton, E.P. Meier, and R.E. Crowe. 1986. Characteristics
   of lakes in the eastern United States, volume I: Population
   description  and physio-chemical relationships.  EPA/600/4-
   86/007a.    U.S.  Environmental  Protection  Agency,
   Washington, DC.  136 pages.

60 Cosby,  B.J., G.M. Hornberger,  J.N. Galloway, and  R.F.
   Wright.  1985.  Modeling the  effects of  acid  deposition:
   Assessment of a lumped parameter  model of soil water and
   streamwater chemistry.     Water  Resources Research
   21:51-63.
                                                    29

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
the Enhanced Trickle Down Model61 and the Inte-
grated  Lake  Watershed   Acidification   Study
(ILWAS)   Model62  (predecessor of  the  NuCM
model).

For lake watersheds in  the  Northeast  that were
successfully calibrated by all three  models during
the DDRP,  the results showed that  each model
produced  similar regional-scale watershed projec-
tions for the target population of future effects from
sulfur deposition on  lake chemistry (see  footnote
56).  That is, the three  models projected similar
trends of regional-scale change in the distribution
of ANC values in the target  population  of lakes
over the various  modeled changes  in sulfur load-
ing.  For  the Southern Blue Ridge Province only
two of the watershed models (MAGIC and ILWAS)
were  used.   Again, significantly more v/atersheds
were successfully calibrated for the  MAGIC model
than the ILWAS model.  In simulations of stream
watersheds  in the Southern Blue Ridge Province,
MAGIC projected greater adverse effects of sulfur
deposition than did ILWAS.

A second  phase of the DDRP repeated the process
of modeling future effects of sulfur deposition for
stream watersheds in the mid-Appalachian region.
In this latter case, for purposes of time and cost ef-
ficiency, modeling was performed  using only the
MAGIC model.  Comparison of model projections
of ANC by  MAGIC  to those made  by the ILWAS
model for two  of the same mid-Appalachian wa-
tersheds, indicated only slight differences between
these projections at  20 or 50  years (see  footnote
57).

In addition to providing the primary analytical ba-
sis for the DDRP and the NBS, as described in the
next two  subsections, an increasing  diversity of
other studies have effectively applied MAGIC  to
assess many watershed processes associated with
acidic deposition (Exhibit 4).   MAGIC has  been
tested  more than  any other  acidic deposition ef-
fects  model. Results from  these tests  (including
some still underway) indicate that MAGIC cor-
rectly projects  the direction of change of water-
shed responses and accurately projects the magni-
tudes of rates of change for surface water ANC and
pH. MAGIC reasonably represents sulfur retention
61 Nikolaidis,  N.P.,  H.  Rajaram,  J.L. Schnoor, and K.P.
   Georgakakos.  1988. A generalized soft water acidification
   model. Water Resources Research 24:1983-1996.

62 Gherini, S.A., L. Mok, R.J. Hudson, G.F. Davis, C.W. Chen,
   and R.A. Goldstein. 1985. The ILWAS model: Formulation
   and application. Water, Air, and Soil Pollution 26:425-459.
within watersheds and the generation and leaching
of cations from watersheds, two functions gener-
ally acknowledged to be the most important of the
modeled processes.

In recognizing that all  models have strengths and
weaknesses, it is obviously unreasonable to expect
that MAGIC (or any other watershed acidification
model)  will  predict accurately  exact  values  of
ANC or pH for any individual lake or stream in the
distant future (e.g., 50 years or more) under condi-
tions  of  significant dynamic change. Rather, the
appropriate use of MAGIC and other such models
is to project the direction and magnitude of possi-
ble chemical changes and to compare the relative
potential effects of  different scenarios of  acidic
deposition. MAGIC appears to be reasonably well
suited for such tasks. It is important to recognize
also, as discussed in Appendix B, that all models
often  are difficult  to test.  That is, most models
may remain largely "unverified," "unconfirmed," or
"unvalidated."  In fact, it can be strongly argued
that a model  can never  be confirmed to be true, it
can only be falsified by  failing to project accu-
rately some outcome. Further, when a model does
accurately predict an outcome,  its validity  is not
proven, because the "right" result may have been
projected for the wrong reason.   Therefore, in  re-
viewing model projections from the DDRP and the
NBS studies on potential effects  attributable to  fu-
ture sulfur and nitrogen deposition, it remains im-
portant to keep in mind the associated  uncertain-
ties that are highlighted  in the following sections.

2.5.2  Direct/Delayed Response Project
As introduced in Section 2.2.1, a 1984 NAS panel
identified two geochemical processes as the domi-
nant  watershed  factors  affecting  or  mediating
long-term surface water acidification: (1) the rates
at which  a  watershed  supplies  base   cations
through  neutralization  and buffering processes in
response to its assimilation of deposited  acids, and
(2) the capacity of a watershed to retain deposited
sulfur-containing compounds. (This NAS report  in-
cluded  minimum concern regarding effects from
nitrogen deposition).  Depending on rates  of  in-
puts,  the capabilities  of watersheds to  perform
these processes  can decrease over time.  Conse-
quently,  based largely  on the  NAS  conclusions,
defining needs to protect various aquatic  and ter-
restrial resources from acidic deposition depended
on  whether  acidification is  immediately propor-
tional  to the  intensity of the  deposition  (i.e.,
"direct")  or  lags in time  (i.e.,  "delayed")  through
such watershed processes.
                                                 30

-------
                                                                      CHAPTER 2:  ENVIRONMENTAL GOALS
                                           EXHIBIT 4. MAGIC
  MAGIC is a lumped parameter model of intermediate complexity that was originally developed to project
  long-term effects (i.e., decades to centuries) caused by acidic deposition on surface water chemistry.  The
  model uses a minimum number of critical chemical ana hydrological processes in watersheds to simulate soil
  solution and surface water chemistry, and to project average monthly or annual concentrations of acid-base
  chemistry in surface water. MAGIC was introduced in a pair of 1985 articles by B.J. Cosby, G.M. Hornberger,
  J.N.  Galloway, and R.F. Wright: (1) Time scales of catchment acidification: A  quantitative  model  for
  estimating freshwater acidification. Environmental Science and Technology 19:1144-1149; and (2) Modeling
  the effects of acidic deposition: Assessment of a lumped parameter model of soil water and stream water
  chemistry. Water Resources Research 21:51-63.

  Church et al. (see footnote 56) summarize various studies using MAGIC.  Recent modifications of the model
  are summarized by T.J. Sullivan, B.J. Cosby,  C.T. Driscoll, H.F. Hemond, D.F. Charles, S.A. Norton, and J.M.
  Eilers (1993.'The influence of naturally occurring organic acids on model estimates of lake water acidification
  using the Model of Acidification of Groundwater in Catchments (MAGIC). Report DOE/ER/30196-3. U.S.  De-
  partment of Energy, Washington, DC). MAGIC has been tested more than any other acidic deposition effects
  model. Those tests indicate that its projections are reasonably reliable:  (1) individual process formulations in
  the model have been tested against laboratory experiments with  soils; (2) model hindcasts (i.e., backward
  predictions)  of historical lake chemistries in the Adirondacks have been made and compared with values
  inferred from lake sediment records; and (3) predictions of the effects  from whole-watershed manipulations
  have been compared to observed effects.

  Two very recent studies examined varying  formulations and calibration  approaches for the application of
  MAGIC to Adirondack lakes:  (1) T.J. Sullivan and B.J. Cosby.  In press. Testing, improvement, and confirma-
  tion of a watershed model of acid-base  chemistry. Water, Air, and Soil Pullution; and (2) T.J. Sullivan, B.J.
  Cosby, C.T. Driscoll, D.F. Charles, and H.F. Hemond. Influence of organic acids on model projections of lake
  acidification. Water, Air, and Soil Pollution. The authors found that increasing  the calibration  value of the
  background (i.e., pre-industrial) levels of  lake sulfate concentration from sea salt contributions to 13% of cur-
  rent levels together with calibrating Adirondack lakes  using soil sulfate adsorption isotherms from only Adi-
  rondack soils (rather than from all Northeast  soils) increased the percentage of target population of lakes pro-
  jected to have ANC<0 u,eq/l at 50 years under a 30% deposition decrease scenario from 6% (using MAGIC as
  formulated and calibrated within the DDRP) to 14%. Changes in either organic acid representation or formu-
  lation of the aluminum algorithm had no  further effect on projections for ANC<0 ueq/l. None of the changes
  had any effect on  model projections for ANC<50 u.eq/1. The model calibration and application tests  resulted
  in somewhat more striking changes  in the estimates of percentages of target populations below various pH
  thresholds or above aluminum concentration thresholds, thus  indicating the sensitivity of MAGIC to the se-
  lection of calibration  approaches for these  variables. This sensitivity was  one reason that  the Nitrogen
  Bounding Study (Section 2.5.3) chose to  compute pH empirically (rather than internal to MAGIC) and chose
  to forego the use of output chemistry (i.e., pH and aluminum) in any predictive models of fish response.

  MAGIC illustrates, as do all models, problems associated with uncertainty, parameterization, and  validation.
  For example, as discussed in the text, MAGIC currently does not explicitly represent detailed cycling or proc-
  esses  affecting the rate of nitrogen  uptake and release. In fact,  processes (and their governing factors) that
  control the transition of a watershed to a  state of nitrogen saturation leading to surface water acidification are
  poorly known. Better nitrogen models to  address the questions are being developed. The Nitrogen Bounding
  Study developed for this report used a series of four scenarios to "bound" the possibilities for time-to-nitrogen-
  saturation.

  In analyses completed for the DDRP and the NBS, it is impossible  to know precisely what deposition levels
  will be over the next 50 years or more. Therefore, the NBS approach assessed a range of deposition levels to
  evaluate potential effects of possible sulfur and  nitrogen deposition combinations. This approach indicates
  why model runs are more correctly termed "projections" rather than "predictions." The latter implies an exact
  knowledge of model inputs and system  dynamics. The NBS projects  watershed acidification  responses  for
  possible alternative  acidic deposition rates in the year 2040.  The  purpose of the NBS was to  evaluate the
  bounds of effects of possible scenarios of  deposition and time to watershed nitrogen saturation.
The DDRP was designed to  begin assessing  the
state and influence of these processes to  support
analyses conducted by the National Acid  Precipi-
tation  Assessment  Program  (NAPAP).   The  man-
date of the DDRP was to make comparative re-
gional projections  of the future effects of sulfur
deposition on long-term surface water chemistry in
the eastern United States, based on the best avail-
able data and most widely accepted hypotheses of
the acidification  process  related  to atmospheric
deposition.  Its data and model calibrations formed
the basis of subsequent NAPAP modeling using a
large number  of  alternative scenarios of atmos-
pheric sulfur deposition.
                                                  31

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
Two principal  project  reports produced by  the
DDRP assessed potential long-term effects of sulfur
deposition on  lake and  stream water chemistry in
the eastern United States:
   * Church, M.R., K.W. Thornton, P.W. Shaffer,
    D.L. Stevens, B.P. Rochelle,  G.R.  Holdren,
    M.G. Johnson, J.J.  Lee, R.S. Turner,  D.L.
    Cassell,  D.A.  Lammers,  W.G. Campbell,
    C.I.  Liff,  C.C.  Brandt, LH.  Liegel, G.D.
    Bishop, D.C. Mortenson, S.M.  Pierson,  and
    D.D. Schmoyer.  1989. Direct/Delayed Re-
    sponse Project:  Future Effects of Long-Term
    Sulfur  Deposition  on   Surface  Water
    Chemistry in the Northeast and  Southern
    Blue Ridge Province.   EPA/600/3-89/026a-
    d. U.S. Environmental Protection Agency,
    Washington, DC. 887 pp.
   * Church, M.R., P.W. Shaffer, K.W. Thornton,
    D.L. Cassell, C.I. Liff, M.G.  Johnson, D.A.
    Lammers, J.J. Lee, G.R. Holdren, J.S. Kern,
    L.H. Liegel, S.M. Pierson,  D.L. Stevens, B.P.
    Rochelle,  and  R.S.  Turner.    1992.  Di-
    rect/Delayed Response Project: Future Ef-
    fects of Long-Term Sulfur Deposition on
    Stream  Chemistry  in the mid-Appalachian
    Region   of  the  Eastern   United  States.
    EPA/600/R-92/186.     U.S.   Environmental
    Protection Agency, Washington, DC.   384
    pp.
The first report focused on  analysis of lake re-
sources in the  Northeast and  stream resources in
the Southern Blue Ridge Province (SBRP). The se-
cond  report  addressed potential stream chemistry
effects in the mid-Appalachian Region, and sum-
marized and integrated conclusions from I he three
regional  analyses.  Additional documenlation of
the DDRP and subsequent  NAPAP modeling ac-
tivities based upon  this  project is  included  in the
following publications:
   * Models Planned for Use in the NAPAP Inte-
    grated Assessment. 1989.  National Acid
    Precipitation Assessment Program,  Wash-
    ington, DC.
   * Methods  for Projecting Future Changes in
    Surface Water Acid-Base Chemistry. 1990.
    Acidic  Deposition: State  of Science  and
    Technology: Report  14.   National Acid
    Precipitation Assessment  Program., Wash-
    ington, DC.
   *  7990 Integrated Assessment Report. 1990.
    National Acid Precipitation Assessment Pro-
    gram, Washington,  DC.
   »Turner, R.S., P.F  Ryan,  D.R.  Marmorek,
    K.W. Thornton,  T.J. Sullivan, J.P.  Baker,
    S.W.  Christensen,  and  M.J.  Sale.  1992.
    Sensitivity to change for low-ANC eastern
    US  lakes  and  streams  and  brook  trout
    populations under alternative sulfate depo-
    sition scenarios.   Environmental Pollution
    77:269-277.
Exhibit 5  shows the locations of three study re-
gions  used during the DDRP.  General characteris-
tics and sizes of target surface water populations
for  all regions  included  as  part of  the  NSWS,
DDRP, and  NBS studies  are presented for com-
parison in Exhibit 6.  Specific characteristics of the
three  DDRP study areas  and their surface  waters
are summarized in the following.

Northeast63
This region includes lakes potentially sensitive to
acidic deposition over the near-  to long-term and
covers an area extending  from northeast Pennsyl-
vania  and northern  New Jersey through the entire
State  of  Maine  (Exhibit 5).  Bedrock  and surface
physiographic  characteristics in  these subregions
help to limit supplies of base cations draining from
these  glaciated and predominately forested water-
sheds. Seepage  lakes are  uncommon,  representing
only 7 percent of the lakes classified by hydrologic
type,  but seepage  lakes  also generally  had  the
lowest values of ANC and pH of any lake type in
this region.  Although the NSWS included lakes
with areas only between 4 and 2,000 ha (see Ex-
hibit 6), there may  be from one to four  times as
many lakes with areas less than 4 ha  in the North-
east.  Because  of their smaller  sizes and  higher
rates of water turnover, such lakes are likely to be
more  highly susceptible  to  acidic deposition ef-
fects.   Concentrations  of nutrients   (i.e.,  NO3',
NH4+, and PO43+) were low for most of lakes sam-
pled in this region.
This region includes the highest dissolved concen-
tration of SO42' observed during the NSWS. Acidic
lakes  were also characterized by high concentra-
tions of extractable aluminum.
The Adirondack subregion, including Adirondack
State  Park, which was emphasized during the NBS
(see Section 2.5.3), has the highest  number and
percentage of acidic (ANC<0 ueq/l)  lakes (14 per-
cent)  found for any NSWS subregion, except Flor-
ida.  Both pH and ANC tended to decrease as the
63 This summary is primarily drawn from Linthurst, R.A.,  D.H.
   Landers, J.M. Eilers, D.F. Brakke, W.S. Overton, E.P. Meier, and
   R.E. Crowe.  1986.  Characteristics of Lakes in the Eastern
   United States, Volume I: Population Descriptions and Physico-
   chemical Relationships.  EPA/600/4-86/007a.  U.S. Environ-
   mental Protection Agency, Las Vegas, NV.
                                                32

-------
                                    CHAPTER 2: ENVIRONMENTAL GOALS
EXHIBIT 5. STUDY REGIONS INCLUDED IN THE DIRECT/DELAYED
 RESPONSE PROJECT AND THE NITROGEN BOUNDING STUDY
                       Northeast
            Adi rondacks
                                            Mi d-Appal aclti an
                                                     i  on
                                      Sout hern  Bl ue  Ri
                                            Provi nee
                     33

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
             EXHIBIT 6. TARGET POPULATIONS INCLUDED IN THE NSWS, DDRP, AND NBS STUDIES
   Target population refers to the number of systems for which
   model projections can be extrapolated. Studied target popula-
   tions of surface waters generally became refined and smaller in
   each of these successive studies, allowing acidic deposition re-
   search efforts  to  focus  increasingly on  relationships  in more
   sensitive surface waters (see figure—not drawn to scale). As a
   consequence  of  narrowing  research efforts,  proportions  of
   sensitive surface waters and the magnitude of the potential re-
   sponse to acidic deposition by these respective target popula-
   tions tend to increase through  subsequent studies.  General
   characteristics of these target populations are presented below.
   As part of the  NSWS, the Eastern Lake Survey (ELS) includes lakes between 4 ha (10 acres, the lower
   limit of resolution of the 1:250,000-scale maps used to design this study) and 2,000 ha (5,000 acres)
   in size that have ANC<400 ueq/l, excluding non-freshwater  lakes (i.e., small ocean bays and estuar-
   ies); broad waters  with apparent flows (reservoirs were included, however); marshes or swamps; and
   waterbodies surrounded by urban, industrial, or agricultural  activities (i.e., lakes with extensive cul-
   tural disturbance in their watersheds). The target lake population in the Northeast was estimated to in-
   clude 326 lakes (4.6%) with ANC<0 ueq/l; 1,364 lakes (19%) with ANC<50 ueq/l; 4,258  lakes (59%)
   with ANC<200 ueq/l; 240 lakes (3.4%) with pH<5.0; and 916 (13%)  lakes with pH<6.0.  In turn, the
   National Stream Survey (NSS) included target stream reaches in sensitive regions not sampled during
   the  National Lake Survey (NLS) that had drainage areas <155 km2 (<60 mi2) and showed as "blue
   line" streams  on  1:250,000-scale U.S.  Geological Survey  topographic maps. Such streams were
   judged to be large enough to  be important for fish habitat, yet small  enough to be susceptible to po-
   tential effects of acidic deposition. At least 50 percent  of the  stream reach had to be within the desig-
   nated region to be included. Among the stream reaches excluded from the target population were, for
   example, reaches  affected by gross pollution (e.g.,  mine or oil-field  drainage), highly urbanized de-
   velopment, or tidal influence.  Of the approximate 72,000 km (44,600 miles) of streams included  in
   the target population in the mid-Appalachian  Region, acidic streams (ANC<0 ueq/l) accounted for
   2,330 km (1,450 miles) (i.e., 3 percenl of the target streams'  length); 18 percent of the upstream ends
   and  7 percent of the  downstream ends had ANC of 50 (j.eq/1  or less.  The 9,036 km (5,615 mi) of tar-
   get streams in  the Southern Blue Ridge Province (SBRP) included some of the lowest concentration  of
   dissolved solids of any  region sampled in the United States (median conductivity of less than  40 u.
   S/cm) and had among the highest deposition  rates for H+, SO42-,  and NCy.  Although no acidic
   streams were found  in the SBRP during  the NSS, acidic streams are known to occur in this region;
   7.8 percent of the target streams had ANC of 50 u.eq/1 or less.
   DDRP target lakes in the Northeast included a subset of NLS target lakes by excluding lakes <1.5 m
   deep. DDRP target stream reaches included a subset of the NSS target reaches by excluding streams
   with ANC>200 ueq/l in the mid-Appalachian region  (Southern Blue Ridge  Province streams had no
   additional restriction on ANC); drainage areas >3,000 ha (>7,400 acres); and all watersheds in the
   Southern Piedmont Regions and Coastal Plain and north of the maximum extent of glaciation.  During
   the  DDRP, 123 lakes in the Northeast, 29 stream reaches in  the mid-Appalachians, and 30 stream
   reaches in the Southern Blue Ridge were modeled.
   The NBS target lakes included a subset of about 700 lakes from the DDRP target lakes within Adiron-
   dack Subregion 1A having ANC<200 ueq/l (approximately  45 percent of the region's lakes meeting
   the  ELS sampling requirements).  The  NBS target stream reaches included the same subsets of target
   reaches for both the  M-APP and SBRP study areas as used during the DDRP studies.  For the mid-Ap-
   palachians, this included about 4,300 stream reaches  (approximately 17  percent of the total included
   in the NSS); for the SBRP, it included about 1,300 stream reaches (approximately 65 percent of those
   included in NSS).
                                                34

-------
                                                                     CHAPTER 2: ENVIRONMENTAL GOALS
lake elevation increased in the Adirondacks,  a  re-
lationship not observed in other NSWS subregions
of the Northeast. Drainage lakes  were the  most
common  type of lake (77 percent of the target
population).  A higher proportion of lakes with  ar-
eas of less than 4 ha in the Adirondacks  tended to
be more boglike and more strongly influenced  by
organic acidity, compared to the  larger lakes in
the Adirondacks.  Approximately half of the Adi-
rondack  lakes having pH 5.0 were organic acidic
dark-water lakes, while the remainder were  clear
water acidic lakes. Inorganic  ions,  including  min-
eral acids, were likely the primary cause of acidity
in these low pH clear water lakes.

Mid-Appalachian Region64
This region  included most stream  reaches poten-
tially sensitive to acidic deposition  within the area
from  central  and  eastern  Pennsylvania  through
western  Maryland  and Virginia  and into eastern
West Virginia (Exhibit 5).  This DDRP study region
included the mountainous physiographic provinces
of the Mid-Atlantic Appalachian  Mountains,  in-
cluding the northern Blue Ridge Mountains, Valley
and  Ridge  Province,  and Appalachian  Plateau.
This area includes Shenandoah  National  Park in
Virginia.  Much of the area extends over bedrock
that  is relatively resistant to weathering.  Rates of
sulfur  deposition in  the  mid-Appalachians  are
much greater than in the Northeast  or SBRP.

Based on  the   National   Stream  Survey  (NSS)
streams  acidified by  acidic  deposition  (ANC<0
ueq/l) accounted for 4 percent of the target stream
length in this region;  18  percent of the upstream
ends  and 7 percent of the downstream ends had
ANC<50 ueq/l. These  estimates excluded  streams
acidified  by mine drainage  (e.g.,  coal  mining).
Mine drainage was responsible for acidifying four
times  as many  downstream  reaches as  acidic
deposition.

More than 99 percent of the acidic target streams
(ANC<0  ueq/l) within  the  mid-Appalachians were
located in watersheds  with at least 85 percent for-
est cover. Many more  streams  with very low ANC
(<50 ueq/l) are found in these forested areas, com-
pared to those in mixed forest or open areas. This
situation probably is not due to the fact that forests
control ANC  and acidic levels;  rather,  most  re-
maining forested areas that  were never  clear cut
lie in the less-weatherable, less-fertile uplands un-
suitable for agriculture.  Areas where forests were
historically  cleared  for agriculture  predominate
along the more-weatherable, more-fertile  valleys.

Southern Blue Ridge Province
This region includes potentially  sensitive stream
reaches in the extreme western portions  of North
Carolina, South Carolina, eastern Tennessee, and
northern Georgia (Exhibit 5).65 The SBRP includes
a steep mountainous region characterized by high
rainfall,  highly weathered   base-poor soils,  and
relatively unreactive bedrock. Target  surface wa-
ters of this region contain some of the lowest con-
centrations of dissolved  solids of  any  region sam-
pled in the  United States, and among the highest
deposition rates for H+, SO42-, and NO3'.  This area
includes  the  Great  Smoky   Mountains  National
Park. Although  no acidic streams (ANC<0 u.eq/1)
were found during the NSS, statistical analysis of
the results from this study indicated that a small
number  representing less than 1  percent of the
streams in the region may be acidic. Also, a sepa-
rate  non-random survey during 1982-1984 found
3 percent of the small streams in  the region to  be
acidic; no larger acidic streams were reported in
this study.66

Watershed retention of SO42- and NO3~ is the ma-
jor  process generating  ANC in  surface  waters
within this region, exceeding base cation mobili-
zation in importance. Sulfur and nitrogen  retention
capacities are generally similar and provide  rela-
tively consistent sources of ANC across the region
with the exception of certain high-elevation forests
affected  by insects.  The primary cause  of ANC
differences in these streams appears to be different
rates  of  acidic  cation  mobilization from the  re-
gion's watersheds. Dissolved organic carbon  con-
centrations are typically low and do not  appear to
provide significant contributions to stream acidity.
64 This summary is primarily drawn from  Herlihy,  A.T., P.R.
   Kaufmann, M.R. Church, P.). Wigington, Jr., J.R. Webb, and M.J.
   Sale.  1993. The effects of acidic deposition on Streams in the
   Appalachian  Mountain and Piedmont Regions of the  Mid-
   Atlantic United States. Water Resources Research 29(8):2687-
   2703.
65 This summary for the Southern Blue Ridge Province is primarily
   drawn from Elwood, J.W., M.J.  Sale, P.R. Kaufman, and C.F.
   Cada.  1991. The Southern Blue Ridge Province. Pages 319-
   364 in D. F. Charles (editor). Acidic Deposition and Aquatic
   Ecosystems Regional Case Studies. Springer-Verlag, New York,
   NY.
66 Winger, P.V., P.J. Lasier,  M. Hudy, D.L. Fowler,  and M.J. Van
   Den Avyle, 1987.  Sensitiviiy of high-elevation streams in the
   Southern Blue Ridge  Province to  acidic deposition. Water
   Resources Bulletin 23:379-386.
                                                  35

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
Concentrations of SO42- have increased at an an-
nual  rate of  approximately 1 ueq/l  during the
10 years prior to 1993 in selected streams draining
both high- and low-elevation watersheds. Over the
same period both ANC and base cation concentra-
tions declined,  indicating that base cation mobili-
zation is not keeping pace with acidic deposition.
Despite this, no significant biological effects from
acidic deposition on streams and lakes have been
conclusively demonstrated for the SBRP.

Overview of DDRP Results
The  DDRP  projected  changes  in target  surface
water  chemistry for several sulfur deposition sce-
narios, while holding nitrogen deposition  and re-
tention constant,  using  up to  three  watershed
models. Model  projections were  compared among
the three DDRP regions.

Results for the Northeast indicated that these target
lakes would  likely respond relatively  rapidly  to
changes in sulfur  deposition,  because  Northeast
watersheds appear, on average,  to be near sulfur
steady state. That is, annual loads of atmospheric
sulfur  deposited into  most watersheds  approxi-
mately equal loads discharged with waters drain-
ing from the watersheds. Remaining sulfur reten-
tion capacities of Northeast soils  appear to be gen-
erally limited. In contrast, DDRP projected that at
either current or increased  sulfur deposition  load-
ings, it might take  150-200 years, on average, be-
fore  SBRP watersheds attain sulfur steady  state.
That is, sulfur  retention potential  appears much
greater in the SBRP than in the Northeast.

Watersheds  in  the  mid-Appalachians  have sulfur
retention characteristics similar to some Northeast
and some SBRP watersheds. As such, they appear
to represent a transition region  where some sys-
tems  will  likely  respond  relatively   rapidly  to
changes in sulfur deposition rates, whereas  other
systems may  respond  more slowly. Projections  of
times necessary to reach sulfur steady  state  aver-
aged about 50 years for the mid-Appalachians wa-
tersheds. Further, in contrast to the  lack of cur-
rently acidic SBRP streams, about 4 percent of the
DDRP target  population  stream reaches in the
mid-Appalachians   are  now chronically  acidic.
Acidic sulfur deposition appears to be the  most
likely   cause  of  surface  water   acidity   in
mid-Appalachian  streams;  sulfate  from  atmos-
pheric sources dominates  the strong acid anion
component in these streams.

Results of the DDRP model projections for the tar-
get populations in  each of the three regions stud-
ied and for  the sulfur deposition scenarios  mod-
eled for purposes of the NAPAP  1990 Integrated
Assessment Report can be summarized as follows.
For lakes in the Northeast, the 1990 CAAA was
projected to reduce the loss of habitat for sensitive
fish species between 1990 and 2030 by 16 percent
to 18 percent from  that which was projected to
occur without the CAAA.67 The results  also pro-
jected  that a  decrease  in  sulfur  deposition of
30 percent  from  1985 levels would  lead  to in-
creases in lake ANC and decreases in the numbers
of chronically and episodically acidic lakes in the
Northeast through the  last  year  of the  DDRP
model projection, 2030.

The  DDRP model projections for the SBRP indi-
cated that continued sulfur deposition at  1985 lev-
els for 50 years would increase stream sulfur con-
centrations  and decrease stream ANC, with the re-
sult that  a  small  percentage  of the DDRP target
population  stream reaches might  become  chroni-
cally acidic in  50 years. There might also be a
slight increase in the number of stream reaches
susceptible  to acidic episodes.

In the mid-Appalachians at the 1985 rate of sulfur
deposition,  the Magic model  projected that in 50
years the proportion of target acidic (ANC<0 ueq/l)
stream reaches would increase from 3 percent and
11 percent. This same deposition  scenario  also is
projected   to   double,   from   25 percent  to
54 percent, the target population of stream reaches
that have an ANC less than 50 ueq/l. This  would
also  double   the number   of  stream  reaches
potentially  susceptible to acidic episodes. Models
of a deposition  scenario involving a 50 percent
decrease  from   1985  sulfur  deposition   rates
projected that increases in stream  ANC  would
occur  in  50 years   across  the   DDRP  target
population   (although   statistically   significant
changes  in the  number of  acidic  reaches or
reaches with ANC less than 50 ueq/l were not ex-
pected) (see footnote 67).

2.5.3  Nitrogen Bounding Study
Modeling Approach
When the Nitrogen Bounding Study began in mid-
1992 to  assess the  combined effects of nitrogen
and  sulfur  deposition over  regional scales,  no
combination of dynamic watershed model and sta-
tistically  based  regional watershed  data  existed
sufficient for the  regional modeling of effects of ni-
67 National Acid Precipitation  Assessment Program.    1990.
   Integrated Assessment Report. Washington, DC.
                                                36

-------
                                                                    CHAPTER 2:  ENVIRONMENTAL GOALS
trogen deposition on watershed nitrogen retention
and surface water chemistry.  Furthermore,  it was
clear that such a goal, desirable as it is, could not
be achieved during the time frame required for this
assessment. Therefore, the  NBS focused on com-
bining existing modeling approaches for dynamic
sulfur effects on watersheds  at regional scales with
estimations of potential bounds  of nitrogen  ef-
fects.68  This modeling expanded upon MAGIC
model runs for sulfur deposition effects made dur-
ing the DDRP.

When the NBS started in 1992 the MAGIC  model
did not have a full representation (e.g., set of equa-
tions) for nitrogen dynamics.  It did, however, con-
tain a representation  of watershed nitrogen input
and output (the latter as nitrate in  surface water).
That is, based upon observed surface water  chem-
istry and  estimated levels  of deposition, MAGIC
contained an estimation of net  annual nitrogen  re-
tention within each modeled watershed. The man-
ner in which this retention might change (i.e., the
magnitude and direction of  change) over different
loadings  modeled provides the assessment basis
used during the NBS.

From   a  functional view,  the  effects of  nitrate
leaching  on  base cation depletion  and surface
water  chemistry is analogous  to those of  sulfate
leaching.   Both ions  act as mobile  anions.   In-
creased leaching of sulfate, nitrate, or both these
anions would  lead to  increases in base cation and
hydrogen  ion leaching on a  net annual basis. Un-
less leached base cations equaled or exceeded the
leached anions, increased leaching of either  nitrate
or sulfate would lead  to decreased surface  water
ANC.

A representation within the MAGIC model of such
effects allows direct computation of effects from
changes in nitrate leaching.  The NBS used alter-
native scenarios of change  in  net annual  water-
shed nitrogen  retention together with potential  fu-
ture alternative loading rates of atmospheric nitro-
gen and sulfur to simulate effects on surface water
chemistry.  These simulations  projected potential
effects in target populations of sensitive surface
waters as  watershed  retention of nitrogen de-
creased from  their  estimated  current  states  to
where only 5 percent or less  of the atmospheric
68 Van Sickle, J., and M.R. Church.  1995.  Methods for Estimating
   the Relative  Effects of Sulfur and Nitrogen  Deposition on
   Surface  Water  Chemistry.  U.S.  Environmental  Research
   Laboratory, Corvallis, OR.
nitrogen deposited was retained in the watershed,
on a net annual basis.  (Retention of nitrogen pres-
ently is  fairly  high for most target  watersheds
studied.)  The shift in the percent of nitrogen  re-
tained by the watersheds was assumed to have the
shape of inverse logistic curves (roughly, a left-
tilted S-shaped curve,  with flattened bottom and
top portions).  These shifts, from  nearly no water-
shed  nitrogen  leaching to  95  percent  or  more
leaching, were separately modeled to occur over
periods of 50 years, 100 years, and 250 years. The
NBS also simulated  the effects of no changes in
net annual retention for watershed nitrogen.

Thus, the NBS did not directly simulate how nitro-
gen deposition  might alter watershed retention of
nitrogen; no combination of model and  regional
data available in early-mid 1992 could do that at
the regional scales required for this report.  Rather,
the NBS  illustrated what would  be  the  result  on
surface water chemistry at regional scales //certain
scenarios of changes in watershed nitrogen reten-
tion were to come to pass.  In so doing, the NBS
effectively bounded all reasonable possibilities of
such effects.  To mimic important biological  ef-
fects on watershed nitrogen cycling, NBS included
curves of declining watershed  nitrogen retention.
That is, at  the  start, modeled retention was high
(e.g.,  because of biological  uptake), then the net
annual retention was modeled to decline over time
(e.g.,  due to biological uptake and sequestration)
until the  net annual  retention was very low  (e.g.,
nitrogen no  longer limited productivity in the wa-
tershed).   This  approach mimics, in  a simplified
sense, watershed systems moving towards nitrogen
saturation (i.e.,  principally  biological saturation)
through the processes described in  various pub-
lished research reports discussed in Section 2.2.1.

A  continuing research  need remains  to  develop
both (1) dynamic watershed process models that
can be used to model nitrogen deposition effects at
regional scales and (2) the necessary regional data
sets to run  such models.  Since 1992, EPA has
funded two parallel projects (one at the University
of New Hampshire and the second at  the Univer-
sity of Virginia) to develop the necessary dynamic
watershed models of combined sulfur and nitrogen
cycling and  effects.

EPA designed the NBS to begin providing a quanti-
tative estimation of potential effects attributable to
nitrogen deposition during surface water acidifica-
tion. This study examined the combined effects on
surface water chemistry due to  potential  changes
in the deposition rates of total sulfur and total  ni-
                                                 37

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
trogen, and due to possible alternative rates of ni-
trogen saturation  within watersheds.  Projected
surface water chemistry for two target years (2015
and 2040) were assessed, with the assumption that
emissions reductions mandated by the 1990 CAAA
(the 10 million tons of SO2 and the 2 million tons
of NOX) were fully implemented.

The study evaluated target populations of surface
waters in three regions: lakes in the Adirondack
Region and stream reaches in the mid-Appalachi-
ans and the Southern Blue Ridge. Target popula-
tions of waters modeled are described in Exhibit 6.
The NBS model projections were completed using
a  modification of  the MAGIC model (see  Ex-
hibit 4). The  primary input  data came from  the
NSWS, DDRP studies, and updated deposition in-
formation from the  EPA atmospheric  modeling
studies discussed in  Chapters.  The NBS  results
represent responses for proportions of NBS  target
surface waters; they do not represent responses for
either all surface waters or for all NSWS sampled
surface waters in the modeled regions.

The NBS study has important implications  regard-
ing the feasibility of aquatics-based acid  deposi-
tion standards in the United  States. Therefore, it is
valuable to understand clearly the nature of the re-
sults produced by this study.  Exhibits 7 to 9 present
3 of  over 60 similar appearing  sets of four plots
presenting NBS model results. These sets of plots
show modeled responses for percentages  of the
target population  of Adirondack  Region lakes pro-
jected to meet the criterion  of ANC of 0 ueq/l or
less in the year 2040 and the percentages of target
populations of mid-Appalachians and SBRP stream
reaches projected to meet the criterion of ANC of
50 ueq/l  or  less in year 2040.  Exhibit 10   intro-
duces guidance for interpretation of the NBS plots
presented in  Exhibits 7 to 9 and  in Appendix B.
More detailed interpretative  guidance is presented
also in Appendix B.

Times to  watershed  nitrogen saturation  in  these
systems remain a major uncertainty. Some model-
ing and empirical analyses (for  example, at Hub-
bard  Brook  in New Hampshire  and the Harvard
Forest in Massachusetts) indicate rather long times
to nitrogen saturation, whereas other results from
the  experimentally  manipulated  watersheds  of
Bear Brook in Maine and Fernow in West Virginia
indicate shorter response times to increased nitro-
gen additions. Also, surface  water nitrate concen-
trations have noticeably varied recently at regional
scales in the Catskills and Adirondacks, confound-
ing the question as to whether some watersheds in
these regions are moving toward or have reached
nitrogen saturation.  Consequently, evaluating cur-
rent  trends for nitrogen saturation on  a  regional
basis remains very difficult. To accommodate this
uncertainty, the NBS  model projections assumed
constant  rates  of  nitrogen assimilation (i.e., no
change from present) and included scenarios of
time to watershed  nitrogen saturation of 50,  100,
and 250 years. Additional considerations regarding
possible regional  difference in  times to  nitrogen
saturation  are presented in Exhibit 11.

For these  plots, deposition rates of sulfur  and ni-
trogen were assumed  to be those projected to ac-
company implementation of the 1990 CAAA to the
year 2010. At  that time, different sulfur and  nitro-
gen  deposition scenarios were defined and  mod-
eled. Some  modeled scenarios  maintained the
2010 deposition rates, while other scenarios mod-
eled  deposition rates  that decreased to  back-
ground deposition rates over the period from 2010
to 2020. Rates for  still other scenarios  reduced to
levels between these extremes from  2010 to 2020.
(Background deposition rates include  only  those
due  to airborne natural, agricultural fertilizer, and
domestic  livestock sources.) Each modeled depo-
sition rate was then assumed to remain  constant at
the  specific  modeled 2020  rate  until the  year
2040, the  end of the model projection period.

Sets  of plots similar to those shown in Exhibits 7
through 9 have been  produced  through this study
projecting  year 2040 proportions  of NBS  target
surface waters  within each of the three modeled
regions  meeting five evaluation criteria:  ANC<
0 ueq/l, ANC<50 ueq/l, pH<5.0, pH<5.5, and pH<
6.0.  Similar plots for  all four water chemistry cri-
teria  were also produced  for  the year 2015. All
other NBS plots showing results  of both ANC and
pH projections are presented in Appendix B.

Because of uncertainties associated  with the com-
plex  chemical  relationships modeled in MAGIC's
derivations for pH, ANC projections are  consid-
ered to be more reliable than those for pH. As pre-
viously described,  ANC is an important indicator
of both  chemical  and  biological  sensitivity to
acidification (see Section 2.3.1).  Projected  water
quality changes are likely to be highly transient in
nature for the year 2015, largely because potential
benefits from  implementation  of 1990 CAAA re-
quirements  will still  be  accruing  at  that  time.
Therefore,  this section focuses  primarily  on pro-
jected ANC changes  in  the year 2040. For those
                                                38

-------
                                                                        CHAPTER 2: ENVIRONMENTAL GOALS
                            EXHIBIT 7. NBS MODEL PROJECTIONS FOR YEAR 2040 PERCENTAGE
                           OF TARGET POPULATION ADIRONDACK LAKES WITH ANC<0 U.EQ/L
      12  -
      10  -
 *•  8H
       6 -


       4 -


       2 -


       0 -
&
o
8°
              6%
                                i
                                6
                                      8
             Total Sulfur Deposition (kg S/ha/yr)
            (Assumes nitrogen saturation @ 50 yr)
                                                     I
 o
Q
 
-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                  EXHIBIT 8. NBS MODEL PROJECTIONS FOR YEAR 2040 PERCENTAGE OF TAR-
                   GET POPULATION MID-APPALACHIAN STREAMS WITH ANC< 50 JIEQ/L
     14  -
f   ,2-

Q
c
I
10 -

 8 -

 6 -

 4 -

 2 -

 0 -
             5.3%
0     2     4     6     8     10

  Total Sulfur Deposition (kg S/ha/yr)
 (Assumes nitrogen saturation @ 50 yr)
                                                        14  -
                                                    1?  12  -i
                                                    .c
                                                    z
                                                    »  10  -
                                                    i    an
                                                    o
                                                    Q
                                                    ID
                                         z
                                         I    2"
                                               o -
                                                                4.8%
                                                              i
                                                              0
                                                                                      i
                                                                                      8
                                                                                       10
                                                                Total Sulfur Deposition (kg S/ha/yr)
                                                              (Assumes nitrogen saturation @ 100 yr)
8
Q.
O
I
    14 -

    12 -

    10 -

     8 -

     6 -

     4 -

     2 -

     0 -
             4.6%
                             T
                             6
                             8
                                        10
            Total Sulfur Deposition (kg S/ha/yr)
          (Assumes nitrogen saturation @ 250 yr)
                                                        14  -
                                                    •£  12  H
                                                    .c
                                                    o
                                                    4)
                                                    Q
                                                   10  -

                                                    8  -

                                                    6  -

                                               2    4  -
                                               Z
                                               S    2  -
                                               \-
                                                    0  -
                                                                     T
                                                                      6
                                                                                            10
                                                               Total Sulfur Deposition (kg S/ha/yr)
                                                               (Assumes nitrogen uptake constant)
                                             40

-------
                                                                     CHAPTER 2: ENVIRONMENTAL GOALS
                       EXHIBIT 9. NBS MODEL PROJECTIONS FOR YEAR 2040 PERCENTAGE OF TAR-
                       GET POPULATION SOUTHERN BLUE RIDGE STREAMS WITH ANC<50
 g
  .
Z
"3
     12  -
     10  -
 *   8
 6  -

 4  -

 2  -

 0  -
             4.4%
                        T
                        4
                          I
                          6
8     10
             Total Sulfur Deposition (kg S/ha/yr)
           (Assumes nitrogen saturation @ 50 yr)
                                                    I
                                               f
                                               .2
                       8-
                      Q
                       0>
                                                    12  -

                                                    10  -
                     6 -

                     4 -

                     2 -

                     0 -
                                   3.8%
                                i
                                0
                                                    8
 I
10
                                                           Total Sulfur Deposition (kg S/ha/yr)
                                                         (Assumes nitrogen saturation @ 100 yr)
I
¥
§
8"
12 -

10 -

 8 -

 6 -

 4 -

 2 -

 0 -
             3.4%
          I
          0
                   i
                   4
1
6
i       i
8     10
            Total Sulfur Deposition (kg S/ha/yr)
          (Assumes nitrogen saturation @ 250 yr)

                                                         12 -
                                                         10 -
                      r   8-
                      _o
                      I    R  _,
                      o_   6  -<
                           4  -
                                                    i
                                                    8
 i
10
                                                           Total Sulfur Deposition (kg S/ha/yr)
                                                           (Assumes nitrogen uptake constant)
                                                  41

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                                 EXHIBIT 10. INTERPRETING NBS PLOTS
                  To illustrate interpretation of the NBS plots, the four individual plots
                  in  Exhibit 7 show projected percentages of NBS target population
                  lakes  in the Adirondack Mountains with ANC  of  0 ueq/l or less,
                  where watershed  nitrogen saturation is assumed to occur at 50, 100,
                  and 250 years, as well as where watershed assimilation rates  for ni-
                  trogen remain constant at recently estimated  rates  (i.e.,  watershed
                  nitrogen saturation  will never occur).  Response contours for each
                  plot show  how percentages of target waters are projected to vary
                  with changes in total  sulfur and nitrogen deposition throughout the
                  modeled ranges of deposition. These ranges begin at projected back-
                  ground deposition rates for sulfur and nitrogen (1  kg  S/ha/yr and 4 kg
                  N/ha/yr, as shown by  the respective axes) and  extend to their maxi-
                  mum  modeled concentrations for 2040 (7.5 kg S/ha/yr and 11.3 kg
                  N/ha/yr),  i.e., the rates projected to accompany  fully implementing
                  the 1990 CAAA (see Chapter 3).  Thus, for the upper right plot of Ex-
                  hibit 7, which shows projections based on an assumed 100 years to
                  nitrogen   saturation,   the  model   projects  that   approximately
                  26 percent of the target Adirondack  lakes  may be acidic (ANC<
                  0 ueq/l)  at year 2040 for modeled sulfur and nitrogen deposition
                  rates projected to accompany implementation of the  1990 CAAA.
                  With only background deposition of sulfur and nitrogen, 3.4 percent
                  of these target lakes are projected to be acidic at 2040.  (Appendix B
                  provides additional guidance on interpreting these plots.)

                  Several general observations apply to these plots:

                    » The  slopes of contour lines in each plot  reflect the relative
                      importance of sulfur and nitrogen in causing  the projected re-
                      sponse relationships. Nearly vertically plotted response con-
                      tours indicate that the projected ANC response (i.e., change)
                      is attributable primarily to sulfur deposition. Nearly horizon-
                      tal plotted response contours indicate  the  plotted ANC re-
                      sponse is attributable primarily  to nitrogen  deposition. A
                      forty-five degree  diagonal contour  indicates equal contribu-
                      tions by both sulfur and nitrogen deposition.

                    « Changes in the spacing between individual response contours
                      within  each  plot appears to  be dependent on patterns in
                      sample weighting during model projections, rather than  due
                      to some intrinsic character of the  deposition-response rela-
                      tionships.

                    » The density of  response contours across the  modeled deposi-
                      tion  ranges for each plot directly relates to the potential aver-
                      age   responsiveness  by  target  waterbodies   to  potential
                      changes in  sulfur  and  nitrogen deposition  rates  on  the
                      specified water quality classification variable modeled (e.g.,
                      ANC<0 ueq/l).  Therefore, plots with a high density of contour
                       lines depict  a high level of  responsiveness to changes in
                      deposition.
                                                 42

-------
                                                                   CHAPTER 2:  ENVIRONMENTAL GOALS
                         EXHIBIT 11. TIME TO WATERSHED NITROGEN SATURATION
  Available information suggests that some individual watersheds in diverse regions of North America
  show symptoms of various states of nitrogen saturation.  Yet, for  each of the three regions modeled
  during the  NBS, existing scientific knowledge  does not allow  us to quantify the  likely times to
  nitrogen  saturation  for  specific proportions  of the target  watershed populations.   Further, for
  watersheds  within any of these regions or in any region, no scientific consensus exists regarding the
  probable times to nitrogen saturation.  Indeed, NBS investigators and its technical reviewers deem  it
  scientifically premature to define specific times to saturation for any region.

  Suggesting that potential  times to nitrogen saturation would tend to vary among regions for those wa-
  tersheds where that process would  most likely occur is, nevertheless, reasonable.  The basis for such
  variations would include differences in seasonal temperatures, moisture, soil fertility, primary produc-
  tion rates, decomposition rates, forest age, and the accumulation  of plant biomass, different histories
  of nitrogen  deposition among the regions, and other factors shown in Exhibit 2.  In addition, given
  historic and current nitrogen deposition rates in  these three regions, we can reasonably assume that
  various of the more sensitive forested watersheds included in the target population for each region
  would eventually reach  nitrogen saturation  (barring major disturbances in these watersheds such as
  extensive logging, fires, blow downs, and insect infestations).

  As a speculative example, watersheds in the Adirondacks have cooler annual temperatures,  shorter
  growing seasons, lower inherent productivity potential, restrictive logging practices and fire  control
  policies, very mature old-growth forest stands, and long histories of elevated deposition rates of sulfur
  and nitrogen. Consequently, barring other severe watershed disturbances, watersheds in these areas
  are likely to include those having the shortest remaining times to nitrogen saturation.  Some research-
  ers suggest that saturation times  in this region may average less than 50 or 100 years.  Recalling that
  various Adirondack watersheds now show initial symptoms of nitrogen saturation, some suggest that
  it may even be as short as 25 years; other recent data suggest the nitrogen trends in  the earlier data
  may be  reversing, so a time of  100 to 250 years may  be more  appropriate.  In comparison, more
  southerly watersheds in  the  mid-Appalachians have generally  warmer annual  temperatures, longer
  growing seasons, less restrictive  forestry practices, and greater inherent productivity potential.  These
  watersheds  also have greater recent nitrogen deposition rates, although historic rates  may have been
  lower.  Such circumstances suggest that target watersheds in this region could have longer  remaining
  durations to nitrogen saturation.

  Watersheds even farther south in the Southern Blue  Ridge Province have  even warmer annual tem-
  peratures, longer average growing seasons, the greatest productivity potential, the fastest decomposi-
  tion rates, historically lower nitrogen deposition  rates, higher recent deposition rates, smaller pools of
  soil  nitrogen, and generally very low stream nitrogen concentrations.  Here, the remaining time to
  watershed nitrogen saturation may  likely be greater still.  Yet, in  recognizing that watersheds across
  any single region generally include  a widely diverse continuum of times to potential watershed nitro-
  gen saturation, some watersheds in the Great Smoky Mountains  National Park contain mature old-
  growth forests; times to nitrogen  saturation would likely be shorter for these areas and, as cited in the
  text of this report, some of these watersheds have been  reported to show symptoms of advancing ni-
  trogen saturation.  Thus,  considerable uncertainty exists for durations remaining for modeled propor-
  tions of target watersheds within  each region to reach advanced stages of nitrogen saturation.
also interested in  projected  changes in pH,  the      ANC<0 ueq/l and ANC<50 ueq/l approximate pH<
ANC  changes discussed  can be related to corre-      5.3 and pH<6.5, respectively. Also, NBS plot pro-
sponding pH changes using the empirical relation-      jection for pH changes using  empirical pH/ANC
ships  between these variables  presented  in  Ex-      relationships are presented in Appendix B.
hibit 1: on average, across the three NBS regions,
                                                43

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
Peer Review of the Nitrogen Bounding Study
The Nitrogen  bounding Study (NBS)  report was
primarily  a methods report (see footnote 68).
Thus, peer review of the NBS focused on methods
used in the study.   Peer reviewers were asked to
comment  on  any part  of the  report that they
wished  and  to answer the  following  specific
questions.  Questions pertaining to the regional
modeling approach included:

  * Is the regional  approach  to the modeling
    appropriate for the purposes  of the work?

  » Is there a  better regional approach?

  * Is the target population appropriate?

  » If there are alternative suggestions for a re-
    gional approach, what data are available
    for implementation?

Questions pertaining to the sulfur modeling in-
cluded:

  * With regard to  sulfur and associated base
    cation  dynamics,  is  the  model  selected
    (MAGIC)  appropriate for the  analyses?

  * Is there a better model?  If so, identify it
    and describe  how  it might be preferable
    and how  it would be applied in a regional
    context.

Questions pertaining to the nitrogen modeling in-
cluded:

  * Is the approach pursued for  mimicking the
    potential  interactions of N appropriate? If
    not, what other approach would be prefer-
    able (be as specific as possible)?

  * Is the shape of the  transition of watershed
    retention  of N-to-N saturation reasonable?
    If not, what other shape would be prefer-
    able and  how mathematically would it be
    represented and parameterized?

  * Are the starting points for watershed nitro-
    gen retention  reasonable?   If not, what
    method  of computation  and what  data
    would be preferable? Do the scenarios of
    time to N saturation  span the possibilities
    in a useful way? Of the scenarios (50,  100,
    250 years,  never), which do you  think is
    most appropriate for each of the regions?

NBS reviewers were asked whether results  of the
work were presented in a manner useful to  policy
makers and whether other graphical presentation
formats or additional information would be prefer-
able.  Finally, reviewers were asked if there were
other ways in which the study and reporting might
be improved.

Peer review comments were very positive.  For ex-
ample, given the  existing limitations on dynamic
modeling of  nitrogen cycling in watersheds, all re-
viewers endorsed the general regional modeling
approach, the specific use of the MAGIC model for
sulfur dynamics, and the representation of  scenar-
ios of potential  changes in watershed nitrogen re-
tention used. There were no strongly negative re-
view comments on the  study. One  reviewer rec-
ommended that the effects of declining base cation
deposition be evaluated, and such  an analysis was
added  to the study.  The first version of the  NBS
analyses included  a model analysis of potential ef-
fects of surface water chemistry changes  on fish
populations.  Upon further consideration of uncer-
tainties associated with these analyses (especially
regarding model projections of pH and aluminum
species), this modeling was  dropped.  This deci-
sion was endorsed in  subsequent  reviewing.   All
reviewers  recommended publication of the  NBS
report as a peer-reviewed EPA Research Report.

Summary of NBS Results
Exhibit 12 summarizes the observed and modeled
percentages of surface waters in each  NBS region
target population  for both ANC criteria. The ob-
served values  were those  measured during  the
1984  NSWS studies in  the Adirondacks and the
1985  studies in the other two regions. For exam-
ple, 19 percent  of the target  lakes  in the Adirond-
acks used during  the  NBS  were  observed to be
acidic  (ANC<0 ueq/l)  during  the 1984   NSWS.
Note again, however, that the target population of
the NBS modeling included generally more sensi-
tive subsets  of  target population  surface  waters
than were included in the NSWS (see Exhibit 6).

Exhibit 12 shows proportions of surface waters in
the two ANC categories projected by the NBS for
the year 2040  under  the assumed times  of 50
years,  100 years,  250 years,  and never for water-
shed  nitrogen  saturation for  each  region.  This
range brackets  the modeled times for watershed
nitrogen saturation occurring across the three NBS
regions for proportions of waters within each ANC
category.  The percentages  presented encompass
the range of NBS results for modeled minimum
(background) and modeled  maximum  deposition
rates for both total sulfur and nitrogen. For exam-
ple, with  an assumed time to watershed nitrogen
                                               44

-------
                                                                    CHAPTER 2:  ENVIRONMENTAL GOALS
           EXHIBIT 12. SUMMARY OF NBS RESULTS: RANGE OF MINIMUM (BACKGROUND DEPOSITION) TO
          MAXIMUM (IMPLEMENTATION OF CAAA) PERCENTAGES OF ACIDIC AND SENSITIVE TARGET WATERS


Acidic (ANC<0 peq/l)
Observed3
Projected 2040, nitrogen saturation in 50 yr
Projected 2040, nitrogen saturation in 100 yr
Projected 2040, nitrogen saturation in 250 yr
Projected 2040, nitrogen saturation never
Sensitive (ANC<50 peq/l)
Observed3
Projected 2040, nitrogen saturation in 50 yr
Projected 2040, nitrogen saturation in 100 yr
Projected 2040, nitrogen saturation in 250 yr
Projected 2040, nitrogen saturation never
Percentage of Target Waters

Adirondacks
19
6^3
3-26
0-15
0-11
55
53-67
51-57
44-54
44-54

Mid- Appalachians
4
0-9
0-5
0-4
0-0
27
5-41
5-37
5-28
4-23
Southern
Blue Ridge
0
0-4
0-0
0-0
0-0
6
4-16
4-16
3-14
2-11
     a Observed 1984 for Adirondack lakes and 1985 for mid-Appalachian and Southern Blue Ridge streams
saturation of 100 years, the  lower left plot of Ex-
hibit  7 shows that background total sulfur and ni-
trogen deposition in the Adirondacks  is projected
to result in  3.4 percent of the target lakes having
ANC  of 0 ueq/l or less in the year 2040. Similarly,
the maximum modeled deposition  rates for both
anions likely under implementation of the  1990
CAAA are projected to result in  about 26 percent
of the target lakes  in  this water quality class,  as
shown in the upper right corner of that plot. Also,
across this  same  range of deposition  scenarios,
when the time to watershed nitrogen saturation is
assumed to  equal 250 years, the  lower left plot in
Exhibit  7 shows that the model  projects that  be-
tween 0 percent and 15 percent of these same tar-
get Adirondack lakes will have ANC of 0 ueq/l or
less. Exhibit 12 shows these two  ranges and sum-
marizes all other similar NBS projections for ANC
by the year 2040 for all three modeled regions.

The numerical  ranges  in  the model  projections
presented in Exhibit 12 provide   one indication of
the extent of uncertainty associated with each set
of model projections for each region. For example,
with  the  modeled rates  of  sulfur and nitrogen
deposition expected  to accompany implementa-
tion of the  1990 CAAA, the percentage of target
lakes  in the Adirondacks  with ANC of  0 ueq/l  or
less would likely range from  about 15 percent to
43 percent, depending on  whether the true time to
watershed nitrogen  saturation  is  nearer 250 or
50 years,  respectively. As  discussed  previously,
many sources of variability and  uncertainty affect
the overall uncertainty of these model projections.
If these sources were included in an overall evalu-
ation of uncertainty,  the  associated uncertainty
could  be greater, with  projections of future re-
sponses by target waterbody populations potential-
ly falling beyond either end of all modeled ranges
presented in Exhibit 12.

The uncertainty in  any computation or modeling
analysis  is the estimation  of the potential differ-
ence between the calculated value (under a set of
conditions) and the "true" value.   It is impossible to
quantify  levels of total uncertainty in model  pro-
jections such  as those made within the  Nitrogen
Bounding Study (or for any watershed acidification
model used in a predictive sense) in any absolute
objective manner.69 This is because there  exist im-
portant categories of uncertainty that defy quanti-
fication.   For regional watershed modeling exer-
cises such as those performed  in the DDRP  or
NBS, certain  components of  model  uncertainty
69  Thornton, K.W., D.R. Marmorek, and P.P. Ryan.  1990.
   Methods for forecasting future changes in surface water acid-
   base chemistry, NAPAP Report 14. Acid Deposition: State of
   the Science and Technology,  National Acid  Precipitation
   Assessment Program, Washington, DC.

   Turner, R.S., P.F Ryan, D.R. Marmorek, K.W. Thornton, T.J.
   Sullivan, J.P. Baker, S.W. Christensen, and M.J. Sale. 1992.
   Sensitivity to change for  low-ANC  eastern US lakes and
   streams and brook trout populations under alternative sulfate
   deposition scenarios. Environmental Pollution 77:269-277.
                                                 45

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
(e.g., sample uncertainty,  input  uncertainty) may
be estimated quantitatively. Previous studies have
addressed, for example, the quantitative uncertain-
ties  associated  with  sampling  and  input  uncer-
tainty in earlier modeling of effects of sulfur depo-
sition and have found that,  given  the scope and in-
tent  of  the  analyses (regional scenario testing for
policy  analysis),  these uncertainties appear rea-
sonable.70 the  difficulty in estimating total uncer-
tainty, however, is  that other components of  un-
certainty (e.g., aggregation uncertainty, structural
uncertainty)  likely  overwhelm  the  former and
these latter components cannot be estimated quan-
titatively.71

A question remains regarding the reliability of the
modeling results from the NBS relative to their in-
tended  use.  The purpose  of the  model runs per-
formed  for the Acid Deposition Standard Feasibil-
ity Study was to test,  at regional  scales,  the sensi-
tivity of potential watershed  responses to varying
scenarios of (1)  nitrogen deposition, and (2) water-
shed transition to nitrogen saturation,  in relation to
projected  effects  of   sulfur  deposition.   Further,
rather than to focus on  explicit  quantitative esti-
mates of percentages of target  populations,  the
utility of the watershed simulations is to examine
direction  and  magnitude of  projected  relative
changes.  Prior  modeling assessments and model
evaluations have established  the credibility of the
MAGIC model and its basic  structure for estimat-
ing the  direction and magnitude of future effects of
sulfur deposition on  surface  water  chemistry in
that  model  projections are consistent with theory
and  observational  data,72 including recent water-
shed manipulations.73  In  that  it allows variable
(increased) nitrate leaching within the model (with
   For detailed discussion of these studies, see the two DDRP
   principal project reports by Church and others published in
   1989 and 1992, as cited in Section 2.4.2; see also the two
   studies by Thorton, Turner and others cited in foolnote 68

   For additional consideration of these issues see the reports
   by Thornton, Turner, and others as cited in footnote 68.

72 This has been shown by the utility of the DDRP results.
   Other  examples are discussed by Church  and others  the
   1992 DDRP principal project report. See also, Wright, R.F.,
   B.J.  Cosby, R.C. Ferrier,  A. Jenkins,  A.J.  Bulger, and  R.
   Harriman.   1994.  Changes in acidification of lochs  in
   Galloway,  southwestern Scotland, 1979-1988: The MAGIC
   model used to evaluate the role of afforestation, calculate
   critical loads, and predict fish status.  Journal of Hydrology
   161:257-285

73 Cosby, B.J., S.A. Norton, and J.S. Kahl. In  press.  Using a
   paired-watershed manipulation experiment to evaluate a
   catchment-scale biogeochemical  model.   Science of  the
   Total Environment.
its consequent effects on base cation exchange and
equilibrium chemistry), the combined sulfur and
nitrogen effects modeling exercised  in this study
makes  no  changes to  the  basic  structure  of
MAGIC.

The purpose of the watershed modeling performed
in support of this Feasibility Study was to examine
at regional scales how  important might be the ef-
fects of nitrogen deposition as compared to those
of sulfur deposition.  Results  indicate clearly, that
under some scenarios, the effects might be compa-
rable in direction and magnitude.

As  noted in  Exhibit 10,  the density  of  contours
across the modeled deposition ranges in NBS plots
for ANC, including those in Exhibits 7-9,  appears
to relate to the potential average responsiveness of
target waterbodies across regional scales to poten-
tial changes in deposition rates. (Vertical and hori-
zontal  contours indicate a strong role of sulfur or
nitrogen, respectively.)  Based on this relationship,
all  regional  plots for alternative projected times to
watershed nitrogen saturation were  categorized
into one of three generalized levels of  projected
response sensitivities. These  categories provide  a
basis for evaluating  the  relative confidence that
reducing sulfur or nitrogen deposition below levels
projected to accompany the  1990 CAAA would
produce detectable  improvements  in ANC within
the NBS target surface  waters across the analyzed
regions.  Exhibit 13 presents the  results of  the sur-
face water  responsiveness  categorization for the
three modeled regions. The following summary of
regional relationships to acidic deposition rates  is
drawn  from Exhibits 12 and 13 and from the indi-
vidual plots for all three NBS study regions.

Because the strongest scientific  data collected on
acidity in the eastern lakes and streams come from
the  1984 NSWS, the  water quality  conditions
found in that survey serve as an illustrative model
for protective goals used in this report.

Regional Summaries74
Adirondack Region
ROLES OF SULFUR,  NITROGEN AND  NATURAL  ACIDITY:
For the  NBS target population of Adirondack lakes,
after implementation  of the  1990 CAAA,  sulfur
deposition appears to continue to  be the primary
cause  of the present  chronically  acidic  surface
74 Most of this discussion was developed from evaluations of
   results from the National Surface Water Survey and model
   projections from the Nitrogen Bounding Study.
                                                   46

-------
                                                                   CHAPTER 2:  ENVIRONMENTAL GOALS
                  EXHIBIT 13. SURFACE WATER RESPONSIVENESS TO REDUCTIONS IN DEPOSITION
                BEYOND THE CAAA: DETECTIBLE IMPROVEMENTS IN LONG-TERM ANC BY 2040a'b
Region
ADIR
M-APP
SBRP
Deposition
Parameter
Reduced
Sulfur
Sulfur
Nitrogen
Nitrogen
Sulfur
Sulfur
Nitrogen
Nitrogen
Sulfur
Sulfur
Nitrogen
Nitrogen
Criterion
(ANC)
<0u.eq/l
<50 u.eq/1
<0|ieq/l
<50u,eq/l
250 years) are assumed.  Nitro-
gen and sulfur deposition  are projected to share
relatively equal future roles in affecting modeled
ANC when watershed nitrogen saturations are as-
sumed  to occur  within  100 years.  And,  when
50 years is assumed as the time to nitrogen satura-
tion, the future importance of nitrogen deposition
as a direct cause of surface water acidification is
projected to be greater.

Proportions of ANC<50 ueq/l  lakes in this region
are projected as likely to show very small changes
due to  deposition reductions by the year  2040.
This relatively  small potential for change reflects
the fact that this  region  has  a high proportion of
lakes which naturally have ANC of 50 ueq/l or less
without the influence of acidic deposition.  These
lakes will continue to have  low ANC levels regard-
less of acidic deposition rates. This condition is not
detrimental in  itself, but makes these waters highly
sensitive to episodic acidic events.

CHRONIC ACIDIFICATION:  Under an assumed time to
watershed nitrogen saturation of 50 years and un-
der the deposition reductions projected from 1990
CAAA implementation,  the  proportion of chroni-
cally acidic (ANC<0 ueq/l) Adirondack target lakes
is  projected to increase  by  about 50 percent in
2015  and may double by 2040,  relative to 1984
proportions. Assuming 100 years to nitrogen satu-
ration, NBS modeling  projects  that  with  imple-
mentation of the 1990 CAAA, proportions of ANC
<0 ueq/l lakes in  the NBS target population  may
increase from  19 percent in  1984 to 26 percent in
2040.  These  increased  proportions  appear due
largely to the increased effects of nitrogen deposi-
tion. If, however, the time to nitrogen saturation
                                                47

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
equals or exceeds 250 years, the model projects a
reduction in the proportion of acidic lakes in 2040
with the implementation of the 1990 CAAA (i.e.,
from  19 percent  to  15 percent  or  less).  The
uncertainty regarding time to  watershed nitrogen
saturation remains the overriding consideration.

EPISODIC  ACIDIFICATION:75 Because episodes  are
driven principally by deposition acidity, reductions
in acidic deposition rates for  either sulfur,  nitro-
gen, or both can be expected to significantly re-
duce the occurrence of acidic episodes in the tar-
get population of Adirondack lakes. This would be
expected to occur at a more rapid rate than the re-
duction in  proportions of chronically acidic lakes
because deposition reductions are likely to have
the greatest immediate influence in  reducing the
mass of acids and acid anions deposited by  major
storms.

RESPONSIVENESS TO DEPOSITION REDUCTIONS: Model-
ing results  for the NBS target population of Adi-
rondack  lakes  indicate  a reasonable expectation
that additional  reductions  in sulfur  deposition
rates, beyond those projected to accompany  the
1990 CAAA, would likely  produce  detectable
long-term improvements  in ANC,  regardless  of the
time to nitrogen saturation for the NBS target wa-
tersheds. It  is also  reasonable to  expect that re-
duced nitrogen deposition would produce detect-
able ANC changes  in these lakes, but primarily if
times to nitrogen saturation for these watersheds
average 100 years or less.

Although considerable uncertainty regarding time
to watershed nitrogen saturation exists,  if the aver-
age time for Adirondack watersheds to reach ni-
trogen saturation  is close to 100 years or less,  the
model predicts that maintaining the proportion of
chronically acidic (ANC<0 ueq/l) target population
Adirondack lakes near their  1984 proportions  in
2040 may  require  reducing anthropogenic  sulfur
and nitrogen deposition by 40-50 percent or more
below the  reductions projected to accompany the
1990 CAAA. The model projects that reductions in
sulfur  and  nitrogen deposition of about  4.5  kg-
S/ha/yr  and 7.5 kg-N/ha/yr, may be necessary to
maintain proportions of  sensitive  lakes IANC<50
ueq/l) near their 1984 levels (i.e, 55 percemt) /Ythe
time to watershed nitrogen saturation approaches
50 years or less. If the time to saturation actually is
100 years or longer, the model projects that  depo-
sition reductions accompanying the 1990 CAAA
75 Also see Section 2.2.2 on episodic acidification.
will  allow proportions of Adirondack  lakes with
ANC<50 ueq/l to maintain their approximate 1984
levels to the end  of the projection interval at the
year 2040.

Mid-Appalachian Region
ROLES OF SULFUR AND NITROGEN: For the NBS target
population of mid-Appalachians stream  reaches
assessed,  model  projections  indicate  that sulfur
and  nitrogen deposition appear about equally im-
portant in potential future surface water acidifica-
tion  for this region (Exhibit 8).

CHRONIC ACIDIFICATION: As  progressively shorter
times to  watershed  nitrogen saturation  are  as-
sumed, effects associated with nitrogen deposition
are projected to increase,  essentially offsetting re-
duced proportions of  affected  streams  resulting
from implementation  of the  1990  CAAA sulfur
reductions in  mid-Appalachians target  streams.
Under assumptions of 250 years or less as the time
to watershed nitrogen saturation, no net change in
the proportion of  acidic (ANC<0 ueq/l) streams in
the  NBS  target  population   is  projected   to
accompany implementation of the 1990 CAAA.

EPISODIC  ACIDIFICATION: Reducing deposition  of
sulfur, nitrogen, or both would be expected to re-
duce the number of episodically acidic  stream
reaches in the mid-Appalachians target population
faster than the rate of reduction for  chronically
acidic reaches, for reasons similar to  those con-
cluded for Adirondack lakes, above.

RESPONSIVENESS TO DEPOSITION REDUCTIONS: Poten-
tial benefits from additional  deposition reduction
beyond the 1990 CAAA is projected to  have bene-
fits for the target  population  of mid-Appalachians
stream reaches having ANC of 50 ueq/l or less, re-
gardless of the time to nitrogen saturation. The na-
ture of  these  benefits  should be viewed  not  so
much as potentially reducing chronically acidic
conditions in these target streams (although this is
likely), but as  potentially reducing the  susceptibil-
ity of sensitive streams  to episodic acute acidifica-
tion effects (i.e., decreasing the proportion  of
stream segments with ANC less than 50 ueq/l).

The 1985 Eastern Stream Survey found 27 percent
of the NBS target streams in the mid-Appalachians
had ANC of 50 ueq/l or less. NBS projections indi-
cate that if the average time to watershed nitrogen
saturation approximates 250  years or greater,  im-
plementation of the 1990 CAAA would likely re-
sult  in  target stream  reaches  maintaining  their
1985 proportions of  chronically acidic  (ANC<0
                                                48

-------
                                                                   CHAPTER 2:  ENVIRONMENTAL GOALS
ueq/l) as well as sensitive (ANC<50 ueq/l)  stream
reaches in the year 2040.  If, instead, average time
to  watershed  nitrogen  saturation  approximates
100 years or less, the model projects that reducing
either  sulfur  or nitrogen deposition  by   about
25 percent below projected CAAA reductions,  or
some lesser  combined deposition reduction  for
both chemicals, could be necessary to maintain
proportions  of  target  stream  reaches  in the
year 2040 near their 1985 conditions. That is, the
deposition rates  may have to be reduced by about
3.5 kg-S/ha/yr or 3 kg-N/ha/yr, or  some combina-
tion leading to reduced deposition for both acidify-
ing chemicals, if the time to nitrogen  saturation
approximates 100 years to maintain  1985 propor-
tions.

Southern Blue  Ridge Province
ROLES OF SULFUR AND NITROGEN:  In the SBRP, like
the mid-Appalachians, projected deposition  effects
after implementation  of the  1990 CAAA appear
about equally attributable to  sulfur  and nitrogen
deposition. This is probably due to the relatively
high remaining  potential for SO42~ adsorption  in
soils and NO3~ retention  in  watersheds of the
SBRP. NBS  modeling  results  project a relatively
minor response  to deposition  changes during the
period  modeled; more  discernible water quality
changes related to either sulfur or nitrogen deposi-
tion may occur after the year 2040.

CHRONIC ACIDIFICATION: Generally,  the NBS  target
population  of  stream reaches  in this  region pres-
ently appears to be little affected by chronic acid-
ity. Six percent of the NSWS stream reaches in this
region  had ANC of 50 ueq/l  or less. Under the
1990 CAAA and under all four assumptions of time
to watershed  nitrogen saturation, the acid chemis-
try  in most target stream reaches are projected  to
change generally little by  2040 from 1985  condi-
tions. A marked exception to  this  pattern is that,
even with CAAA implementation,  the proportion
of target  stream  reaches with ANC of 50 ueq/l  or
less are projected to approximately double,  reach-
ing 11 percent to 16 percent by 2040,  under the
modeled times to nitrogen saturation of 250 years
or greater. In turn, if nitrogen  saturation occurs  at
about 50 years, about 4 percent of stream reaches
might become acidic (ANC<0 ueq/l) where none
previously had been acidic.

EPISODIC ACIDIFICATION:  As the number of stream
reaches in the SBRP target population with chronic
ANC of 50 ueq/l or less increases, the possibility of
episodically acidic  conditions increases substan-
tially and can be reasonably expected to occur in
these target streams prior to (and at greater per-
centages  than)  the  occurrence  of chronically
acidic conditions.

RESPONSIVENESS TO DEPOSITION  REDUCTIONS: Model
projections indicate that deposition reductions ac-
companying the CAAA would  likely prevent long-
term acidification (decreasing  ANC) of sensitive
streams at least until 2040, if the time to nitrogen
saturation is 100 years or longer. Modeling projec-
tions also indicate that  additional  reductions in
sulfur or nitrogen deposition in the SBRP, beyond
those expected to accompany the CAAA,  could
reduce  the  proportion  of  target  stream  reaches
with ANC of 50 ueq/l or less. If the time to water-
shed nitrogen saturation  in  this  region  is  near
250 years or longer,  the NBS  modeling projects
that sulfur deposition would need to be reduced by
about an additional  25 percent to  maintain  the
ANC in the  target stream reaches  near their 1985
conditions. Whereas,  if the time to watershed ni-
trogen  saturation is nearer 100 years  or less,  a
65 percent total decrease in both sulfur  and nitro-
gen deposition beyond the CAAA is projected as
necessary to maintain ANC in these target stream
reaches  near  their 1985  conditions in the year
2040. NBS projections for this potential time to ni-
trogen saturation  (100 years or less) indicate that
deposition may have  to be reduced below  4 kg-
S/ha/yr and  below 5  kg-N/ha/yr to maintain pro-
portions  of  stream reaches in  the SBRP  target
population with ANC<50 ueq/l at 1985 values.

Implications for an Acid Deposition Standard
The  potential sensitivities  of  target aquatic  re-
sources,  their potential  responses, and  response
times to changes in acidic deposition rates as well
as the relative current and potential roles of  sulfur
and  nitrogen  clearly  differ  among  regions.  This
strongly  supports  the  development of  a  site-
specific  deposition standard  or  target load  as
opposed to a single national standard.

The  acidifying  effects   of nitrogen  deposition
should be considered when evaluating options and
potential need for acid deposition  standards. Spe-
cifically,  NBS modeling indicates  that nitrogen
deposition appears to produce  important conse-
quences for the future acidification rates of surface
waters. Additionally, for many watersheds the ef-
fect of nitrogen deposition could be a greater con-
cern than are the effects of sulfur deposition alone,
given the reductions taking place under Title  IV.
                                                49

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
Exhibit 14 presents NBS model projections of per-
centages of  acidic  and  sensitive  surface waters
(ANC<0 ueq/l and ANC<50 ueq/l) in three regions
of the eastern United States in the year 2040 under
three scenarios:  (1) without the 1990 CAAA (sulfur
deposition  held constant at  1993  levels and a 1
percent annual  increase in  nitrogen  deposition
after the year 2000); (2) without the 1990 CAAA
sulfur reductions (sulfur deposition held constant at
1993  levels) but  with the nitrogen  controls
(nitrogen deposition held constant at 1985 levels);
and  (3) with  implementation  of  the CAAA sulfur
and  nitrogen controls  (nitrogen deposition  held
constant at  1985 levels). Results  of these  three
scenarios illustrate the proportion of surface waters
that  would likely have  been  acidic  or sensitive to
becoming acidic had there been no CAAA. These
modeling   projections   are  subject   to    the
uncertainties  described previously.  As  such,  they
indicate  approximate  proportions  of the surface
water target populations projected to have ANC<0
or <50 fieq/l for  the indicated deposition scenarios.
The  scenario depicting  no sulfur reductions and a
1 percent annual increase in nitrogen deposition is
intended to be representative of a situation without
CAAA reductions (no sulfur reductions  and a  con-
tinuing  increase in  nitrogen deposition). The ex-
hibit shows that the reduction in sulfur deposition
levels resulting from the 1990 CAAA are projected
to  provide  benefits   in  improving  ANC  and
reducing acidic  stress in the lakes  and streams of
the three regions that hold a major proportion of
sensitive aquatic resources  in the  eastern United
States.

The  wide-ranging projections of possible benefits
associated with  implementation of the 1990 CAAA
for each of these regions illustrate the need to im-
prove the  ability  to quantify watershed nitrogen
saturation rates. Until watershed nitrogen satura-
tion  is better understood, significant  uncertainty
will  continue to accompany surface water benefits
analyses of potential reductions  in  sulfur and ni-
trogen deposition.  Despite the  uncertainty,  how-
ever, it is useful to  recognize that any reductions
in nitrogen deposition would not only reduce total
acidic deposition rates, but also  tend to lengthen
the actual  times to  watershed nitrogen saturation
in some watersheds sensitive regions. This process
is similar to  the ongoing process whereby reduc-
tions in sulfur deposition due to the 1990 CAAA
are  likely  extending times for   sulfur saturation
within watersheds.
2.5.4  Overview of International and State
       Acidic Deposition Criteria and
       Standards
International  consideration  of  ecologically based
standards to address air  pollution problems origi-
nated in the  mid 1960s.  Driven primarily by  the
acid rain debate over the next  30 years, the origi-
nal concept of using concentration-based  criteria
for precipitation  gave  way  to  using  uniform
maximum allowable  mass  deposition rates, with
20 kg-wet SO42Vha/yr (6.7 kg-S/ha/yr)  being  the
first  widely   recognized   interim  target  load.
Subsequently,  site-specific  critical   loads  were
increasingly  emphasized.  Their development is
generally attributed to Swedish  research efforts in
the late 1960s.76

Critical loads are estimates of  the maximum  pol-
lutant loadings that environmental resources can
absorb on a sustained basis without experiencing
measurable degradation.  Only  inherent ecological
properties are included in site-specific critical load
determinations. Steps involved  in defining and  im-
plementing    critical   loads   usually    include
(1) resource   identification  and  characterization,
(2) identification of regions or functional  subre-
gions,  (3) characterization  of  deposition  within
subregions, (4) definition  of assessment endpoint(s)
(see below), (5) selection  and application of mod-
els, and  (6) mapping  projected  environmental re-
sponses
77
Target loads differ from critical  loads in that their
definitions incorporate  social, policy,  economic,
and  related considerations along  with scientific
findings. An example of a target load would be an
acidic deposition level adequate to maintain  pro-
portions of ANC<50 ueq/l  waters at or below the
proportions found during the 1984-85  NSWS for
one or more of the surveyed  regions. (This exam-
ple is illustrated  in  Section 3.6.)  Other possible
target loads could include, for example, a deposi-
tion level to produce a specified percentage reduc-
tion  in  the 1984-85  proportions of ANC<50 ueq/l
waters.
76 Nilsson, J. and P. Grennfelt (editors).  1988. Critical Loads for
   Sulphur and  Nitrogen  Report  from  a  Workshop  Held  at
   Skokoster, Sweden,  19-24 March 1988,  UN/ECE and Nordic
   Council of Ministers.

77 Strickland, T.C., C.R. Holdren, Jr., P.L. Ringold, D. Bernard, K.
   Smythe, and  W. Fallen.  1993.  A National Critical Loads
   Framework for Atmospheric Deposition Effects Assessment:  I.
   Method Summary. Environmental Management 1 7:329-324.
                                                 50

-------
                                                                    CHAPTER 2: ENVIRONMENTAL GOALS
     EXHIBIT 14. IMPACT OF CAAA ON SENSITIVE SURFACE WATERS: NBS MODEL PROJECTIONS FOR YEAR 2040

               PROPORTIONS OF TARGET POPULATION SURFACE WATERS IN TWO ANC CATEGORIES
                    FOR THREE DEPOSITION SCENARIOS (SEE TEXT FOR ADDITIONAL DETAILS)
Region
ADIR
M-APP
SBRP
Criterion
ANQSO |jeq/l
ANC<50 ueq/l
ANC<0 ueq/l
ANC<50 ueo/l
ANC<0 ueq/l
ANC<50 peq/l
Deposition Scenario
w/o CAAAb
Constant S and Nc
CAAAd
w/o CAAAb
Constant S and Nc
CAAAd
w/o CAAAb
Constant S and Nc
CAAAd
w/o CAAAb
Constant S and Nc
CAAAd
w/o CAAAb
Constant S and Nc
CAAAd
w/o CAAAb
Constant S and Nc
CAAAd
Observed
Proportion3
19
55
4
27
0
6
Proportions (Percentages) at Modeled
Times to Watershed Nitrogen Saturation
50 years
52
50
43
77
74
67
42
33
9
76
67
41
14
13
4
31
20
16
1 00 years
39
36
26
59
58
57
28
23
5
66
54
37
7
2
0
22
17
16
250_years
23
23
15
55
55
54
23
21
4
65
48
28
2
1
0
17
15
14
Never
24
25
11
55
55
54
21
8
0
49
38
23
0
0
0
15
15
11
3 Observed in 1984 in Adirondacks and 1985 in Mid-Appalachians and Southern Blue Ridge.
b Sulfur deposition held constant at 1993 levels; nitrogen deposition increases 1% per year after 2000.
c Sulfur deposition held constant at 1993 levels; nitrogen deposition held constant at 1985 levels.
d Reflects decreases in sulfur deposition from implementation of Title IV; nitrogen deposition held constant at
  1985 levels.
Assessment endpoints are formal expressions of the
environmental value(s) to be protected.  They can
include  thresholds  for  "deleterious  conditions"
(commonly some ecological condition of concern)
that a standard would attempt to prevent. Assess-
ment  endpoints should  be biologically relevant,
operationally definable,  accessible  to prediction
and measurement, and sensitive to the pollutant(s)
of concern. From a policy perspective, assessment
endpoints also should be socially relevant; that is,
they should be environmental  characteristics mu-
tually understood and valued by the  public and by
decision makers (e.g.,  populations of crops,  trees,
fish, birds, or mammals). When the most appropri-
ate sensitive species or other endpoint used  is not
socially valued, then their link to valued species or
other  valued environmental  attributes  should be
explicitly demonstrated to simplify  understanding
of why using such  an endpoint is  useful.  Using
endpoints that have social relevancy helps to unify
scientific and social concerns in commonly shared
objectives.

The first and still dominant ecological assessment
endpoint used for critical and  target load estima-
tion  is  freshwater aquatic responses, most  com-
monly manifested as  changes  in pH or  ANC. Of
particular  interest here is that pH or ANC changes
themselves are often  a relatively minor  concern,
but the influence of  such changes on biological
species is  of considerable importance. Therein is a
defining attribute of  how  the  concept of critical
loads has  developed  in  its international  use, i.e.,
critical  loads of chemicals (e.g., SO42' and NO3~)
are surrogates  for biological  concerns.   The key
biological concern most often focused upon is fish
viability.

A  critical  load value  can be viewed as  a single,
especially  important   point along  a continuous
                                                 51

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
range of values representing an ecological  loss or
damage function. Viewing a critical load as part of
an ecological loss function is especially important
because  that view  has shaped much of the Euro-
pean debate over the appropriate and scientifically
defensible uses of critical loads in acid rain control
policy development. Further, these loss functions
are particularly important when it is recognized (as
it has  been  in  Europe, Canada,  and the  United
States) that significant numbers of  highly sensitive
receptor locations  have  associated critical loads
that likely cannot be met, even with widespread
and high levels of acidic deposition reduction em-
ploying the limits of technological  feasibility. Loss
functions provide a mechanism to aggregate eco-
logical  damage  across regions  and  integrate  a
quantitative understanding of acidic deposition to
ecological damage relationships.  Without these
functions,  more qualitative,  often highly subjec-
tive, aggregation  approaches  are  used to  place
critical  load  concepts into  policy-relevant  con-
texts.  The results of such approaches  most often
are termed target loads, as described above.

Most  countries  of  western Europe have adopted
the system for estimating critical loads developed
by the Coordination Center for Effects (CCE) of the
United Nations  Economic Commission for  Europe
(UNECE)  under the auspices of the UNIECE Con-
vention on Long-Range Transboundary Air Pollu-
tion (LRTAP)  (Exhibit  15). A  recently  published
manual presents improved methods that are being
widely applied  across Europe for  mapping  both
critical  loads to protect sensitive  resources and
critical levels of allowable atmospheric concentra-
tions of acidic pollutants.78 In  this  system,  critical
loads are  developed  for  individual cells  of the
mapping grid based on the potential sensitivity to
acidification  of forest soils  and   surface  waters.
Critical loads  for sulfur, nitrogen, and total  acidity
have been mapped across Europe.  Although a va-
riety of models were used, nearly all countries that
participated in the  European mapping effort em-
ployed   the  simple   mass  balance  steady-state
method  as the  underlying approach  to estimate
critical loads. Several  countries also used dynamic
models and other methods. In the CCE approach,
the indicator used to estimate critical loads for for-
est  soils (using  the simple mass  balance steady-
state model)  is the concentration  of aluminum in
the soil solution required  to  maintain pH above
4.0. Some countries varied the basis for their map-
ping procedures based on the availability of the
data collected by the individual countries, and the
regional and national concerns regarding the sensi-
tivity of specific sensitive resources. The majority
of critical load values in Europe  reflect the sensi-
tivity  of  forest soils. Critical  loads in Finland,
Norway, Sweden, Switzerland, and several of the
newly  Independent States  reflect forests and sur-
face waters.79

The  single   most  important  technical  attribute
around which European activities on acid deposi-
tion standards have revolved is associated with de-
fining the spatial resolution used. Interestingly, the
early decision to use a 150 km by 150 km square
grid as the fundamental spatial assessment unit for
acidic  deposition control  strategies had no direct
connection  to spatial levels of resolution deemed
appropriate  for critical load estimation. In fact, the
grid  was in  place  well before the critical loads
concept achieved common usage. This relatively
coarse grid size, however,  often allows for signifi-
cant spatial  variation in environmental  types and
designated  critical  load alternatives within  indi-
vidual  cells. This leads to difficult questions re-
garding spatial estimation of specific critical loads
appropriate  for  supporting deposition-based  con-
trol  policy  and  measurement  of  maintenance
and/or exceedance  levels.

To provide  a reasonable  level of  protection for
more sensitive  ecological  resources within each
grid cell, the European approach  uses cumulative
distributions of critical  load  values  and  selects
from this distribution a non-exceedance level for
each cell. Under this approach two loadings are
calculated:  one that  would protect 95 percent of
sensitive ecological resources  within the grid  (i.e.,
the 5-percentile load), and one that would  protect
99 percent of the resources (i.e.,  the  1-percentile
load). This procedure reconciles some of the basic
problems that arise when point estimates are  used
to represent regional concerns.  But the approach
still holds difficulties  related largely to the process
of selecting appropriate critical  load  values  from
the resulting distribution functions. Specifically,  it
is sometimes difficult to determine the rationale by
78 Task Force on Mapping.  1993.  Manual on Mapping Critical
   Levels/Loads. Coordination Center for Effects, U.N. Economic
   Commission for Europe. Berlin, Germany.
79 Coordination  Center for  Effects, National Institute of Public
   Health and Environmental Protection. 1991. Mapping Critical
   Loads  for  Europe.  CCE Technical Report  No.  1.   U.N.
   Economic Commission for Europe, Bilthoven, Netherlands.
                                                 52

-------
                                                                    CHAPTER 2:  ENVIRONMENTAL GOALS
                EXHIBIT 15. LRTAP
   In  1981, the United States became party to
   the  UNECE  Convention  on  Long-Range
   Transboundary Air Pollution (LRTAP). Signa-
   tory  countries include most western Euro-
   pean countries, several newly  Independent
   States,  Canada, and the  United States.  The
   Convention symbolizes a formal recognition
   on the part of  signatory countries  that the
   transboundary flow of air pollution is an im-
   portant issue that  merits formal  international
   cooperation. In 1985, the first Sulfur Proto-
   col under the  Convention committed those
   countries who signed to a 30 percent reduc-
   tion in emissions of sulfur dioxide from 1980
   levels. (The United States did not sign based
   on significant sulfur dioxide emission reduc-
   tion efforts already undertaken in the 1970s.)
   Soon thereafter, emissions reductions based
   on achievement of critical loads became the
   focus of efforts under LRTAP. In 1994, most
   countries signed the Second Sulfur Protocol
   requiring mapping of critical loads for all af-
   fected countries.  It is  the expectation that
   this effort will lead to emission reductions
   based on the critical  loads. (The United
   States  did   not   sign  the  Second Sulfur
   Protocol because  it lacks statutory  authority
   to  reduce emissions to meet critical loads.)
   A  Nitrogen Oxide  Protocol was signed in
   1988  by most   countries,   including  the
   United  States. The NOX  Protocol  outlined
   steps  to  reduce  national  annual  NOX
   emissions.  It also  initiated research  and
   cooperative  efforts  on  critical loads  for
   nitrogen.
which individual critical load values are  selected
among the different cells. Qualitative considera-
tions, which often stem from political agreements,
also have a role in this process.80Despite the dif-
ferent  sensitivities  of  various ecosystems,  most
critical loads  developed in Europe  are very low
when compared to present  deposition.  This has
given some countries the impetus to seek greater
emissions reductions than  were already planned.
Thus, most of the  reductions of sulfur emissions
under the second Sulfur Protocol fall in the range
of 50-80 percent. The European community as a
whole is projecting emissions decreases of over 60
percent by the year 2000 compared to 1980 levels.
The four European countries that signed the sec-
ond Sulfur Protocol and their commitments to  re-
ductions (relative to 1980) are:81

  .  France:          74% by 2000; 78% by 2010

  .  Germany:        83% by 2000; 87% by 2005

  .  Italy:            65% by 2000; 73% by 2005

  .  United Kingdom:  50% by 2000; 80% by 2010

The French  reductions translate into an emissions
level of approximately 825,000 tons of sulfur diox-
ide; emissions reductions in the other three coun-
tries are  around  1.1  million  tons each. Taken
together,  these   four   European   industrialized
countries represent a population very close to that
of the  United States.  By 2010, their emissions of
sulfur dioxide will be  less than  5 million  tons,
while  the United States  is  projected  to  have
emissions of around 15  million  tons.  Canada
committed to reducing its emissions by 46 percent
within a Sulfur Oxide Management Area (SOMA),
which represents a targeted approach to the acidif-
ication problem in Eastern Canada. Canada, with a
population of about 10 percent that of the  United
States,  is committed to a national cap of 3.2 mil-
lion metric tons (about 3.5 million  tons) in the
year 2000.

Canada adopted 6.7 kg-S/ha/yr  (wet deposition) in
the early 1980s as what would now be termed a
target load. This value was based on available data
indicating that loss of sport  fish  would occur at pH
less than 5.3, and this loss  would produce signifi-
cant economic and social impacts. This target load
was not, however, intended  to protect extremely
sensitive areas. Canadian policy makers concluded
that additional  research  was necessary to deter-
mine appropriate loading limits to completely pro-
tect  all  sensitive Canadian ecological  resources.
The target load was used as a  goal in developing
the Canadian  acid rain  control  program  and in
discussions with the United States on transbound-
ary air pollution. As a result of current U.S. and
Canadian acid rain control programs,  most areas
80 Henriksen, A., and D.F. Brakke. 1988.  Sulfate deposition to
   surface  waters.    Environmental  Science  and Technology
   22(1):8-14.
81  United Nations  Economic Commission for Europe.  1994.
   Protocol to the 1979 Convention on Long-Range Transboundary
   Air Pollution on Further Reduction of Sulphur Emissions.
   ECE/EB,AIR/40.  Geneva.
                                                53

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
of Canada are expected to reach the 6.7 kg-S/ha/yr
target.

Subsequently,  in  1990,  critical  loads for  water-
sheds  in  eastern  Canada  were  calculated  using
steady-state water chemistry models that projected
sulfur deposition necessary to maintain 95 percent
of the lakes at pH 6.0 or greater.82 Included in this
approach were allowances for maintaining cond-
itions in naturally acidic surface  waters. Resulting
critical  load estimates ranged from less than 2.7 to
more than 6.7 kg-wet S/ha/yr, with the variation
primarily dependent on geological  characteristics.
The eastern region of Canada, including the Atlan-
tic provinces, Labrador, and eastern Quebec, was
determined to require the lowest critical loads (less
than  2.7 kg-S/ha/yr),  which   are close to back-
ground deposition levels.  Critical loads estimated
for Quebec  ranged from  3 to more than 6.7 kg-
S/ha/yr, and for Ontario they ranged  from 2.7 to
more than 6.7  kg-S/ha/yr.  In a separate analysis, a
simple  mass balance approach was used to deter-
mine that a  loading  of  5 to  6.7 kg-S/ha/yr wet
deposition  would  maintain  surface   water  pH
greater  than 5.3 on  an annual basis in  watersheds
that  have lakes with ANC of  200 ueq/l or greater
in regions of low runoff.

Using  the initial Canadian effort as an example,
the New England states  and New York adopted
Canada's first  target  load of 6.7 kg-S/ha/yr  (wet
deposition) as  a level adequate to  protect moder-
ately sensitive ecological resources from additional
damage caused by  acidic  deposition.83 This  level
was not viewed, however, as adequate to protect
the most sensitive resources within these regions.

Maryland developed  critical  loads  based on the
sensitivity of individual streams to acidification.84
This effort included as its overall goal an  assess-
82 Federal/Provincial  Research  and  Monitoring  Coordination
   Committee. 1990.  The 1990 Canadian Long-Range Transport
   of Air Pollutants and Acid Deposition Assessment  Report.  8
   parts.   Research and Monitoring  Coordination Committee,
   Canada.

83 New England Governor's Conference.  1985.  History and the
   Development of the New England Position on Acid Rain.  New
   England Governor's Conference, Inc.

   New York State Department of Environmental  Conservation
   (NYSDEC). 1985. A Policy for New York State to Reduce Sulfur
   Dioxide Emissions:  The Sulfur Deposition Control Program.
   Final Environmental Impact Statement. NYSDEC, Albany, NY.

84 Sverdrup,  H., P. Warfvinge, M.  Rabenhorst, A. Janicki,  R.
   Morgan, and M. Bowman. 1992.  Critical Loads and Steady-
   State Chemistry for Streams  in  Maryland.  Environmental
   Pollution 77:195-203.
ment of the extent to which the state could meet
or surpass its ecological  objectives  to  minimize
potential  acidic  deposition  effects. Calculated
critical  loads for areas within Maryland ranged by
region from less than 8 to more than 64 kg-S/ha/yr.
These loads were developed using  (1) two models
(PROFILE and MAGIC), (2) pH limits required to
protect  the most sensitive life  stages  of biological
indicator  species,  and (3) a complex of  specific
physical,  chemical,  and  biological  factors that
potentially affect soil  and water chemistries. Acid
sensitivities for  three  indicator fish species were
used  across the different regions  assessed: blue-
back  herring (pH=6.2), smallmouth bass  (pH=5.8),
and  brook  trout  (pH=5.75). The   assessment  re-
vealed that critical loads at several sensitive recep-
tor locations could not be met for any plausible
emissions control scenario. These  locations were
thus deemed possible candidates  for site-specific
mitigation measures, principally stream liming.

Wisconsin has a precipitation pH goal of 4.7, cor-
responding to a wet sulfate deposition rate of ap-
proximately  11  kg/ha/year.85   Wet deposition is
used as the evaluation base rather than total depo-
sition because  of existing  uncertainties  regarding
dry deposition  rates and the effects of dry inputs
on  biological systems.  A primary intent  of  this
goal  is  to assist decision  makers   in determining
whether adequate environmental protection is oc-
curring.   When properly developed, such goals
can help  provide adequate protection of sensitive
biological systems.  Studies in Wisconsin indicate
the potential for acute acidification of sensitive
aquatic ecosystems by atmospheric inputs of sulfur
have  been  reduced by state  mandated emission
reductions intended to address this goal.

Minnesota is the only state with  an established
deposition standard for sensitive areas.86 Sensitive
areas are defined  based on  lake  ANC, with the
state's deposition standard of  3.7 kg-S/ha/yr (wet
deposition)  established to  protect  lakes  whose
ANC is less than 40 ueq/l. The standard was de-
rived using regression techniques to relate deposi-
tion SO42~ concentrations and acidity to the ability
85 Based  on a March 31, 1995,  letter from Donald Theiler,
   Director, Bureau of Air Management, State of Wisconsin, to
   Rona Birnbaum, Acid Rain Division,  U.S. Environmental
   Protection Agency.
86 Minnesota Pollution Control Agency. 1985. Statement of Need
   and Reasonableness: Proposed Acid Deposition Standard and
   Control Plan. State of Minnesota Pollution Control Agency, St.
   Paul, MN.
                                                   54

-------
                                                                     CHAPTER 2:  ENVIRONMENTAL GOALS
of sensitive Minnesota lakes (ANC<40 (jeq/l) to as-
similate  additional acid loadings. This deposition
standard may also be viewed as equivalent to the
critical load for this region because maps of sens-
itive soils show that the majority of areas with po-
tentially  the  most  sensitive  soils correspond  to
those regions with sensitive lakes.

In contrast to  regions east of the Mississippi  River,
deposition of  nitrogen has long been viewed as a
more significant problem than deposition of sulfur
in much of the western United States. For exam-
ple,  NO3'approximately equals SO42~ deposition in
some areas of California.  Consequently, critical
loads for nitrogen deposition  have been estimated
for California,87 but similar  loads for sulfur have
not.  Studies  show that loadings of  10 to  20kg-
N/ha/yr  would likely  protect  California forests.
Critical loadings recommended to protect sensitive
California  resources  ranged  from  5 to  45kg-
N/ha/yr,  depending  on  the region. Other state ef-
forts are  also underway.

2.5.5  Spatial and Temporal Issues in
       Development of a Standard
Information discussed in Sections 2.3,  2.4, 2.5.2,
and  2.5.3 clearly  demonstrate  that  regions  of
North America differ in both their potential  sensi-
tivity and risk  to adverse effects caused by  sulfur
and nitrogen deposition. These differences provide
a strong  scientific justification for setting different
standards that  recognize variations among and
within sensitive regions. Regions covered by any
individual  standard  would  be  larger  than  most
states and undoubtedly smaller than the nation as
a whole. It is also  clear that appropriate and scien-
tifically justifiable environmental goals could differ
across areas within individual  states.

Sensitive resources tend to cluster within relatively
easily defined  geographic areas often associated
with specific mountain ranges and other areas hav-
ing relatively unique geologic attributes.  Potential
protection requirements for ecological resources in
sensitive regions  can be identified,  categorized,
and  aggregated across several levels of organiza-
tion. These include  regions,  landscapes,  ecosys-
tems, communities,  populations, and individual
site-specific measures (e.g.,  critical stream habitat
87 Takemoto,  B.K., M.  Bergen, N. Motallebi,  M.  Mueller, H,
   Margolis, and S. Prasad.  1992.   The Atmospheric Acidity
   Protection  Program:  Annual Report to  the Governor and
   Legislature.  Draft report. State of  California Air  Resources
   Board, Research  Division, Sacramento, CA.
for a listed endangered fish species). In general,
the smaller the area of concern, the greater the
precision  required  in  establishing  the basis for
standards  and  in  determining  the boundaries
where standards would  apply.  Under  a critical
loads approach, appropriate ecological  rationales
would need to be developed for whatever scale is
targeted for protection by a standard. Furthermore,
there is significant variation in spatial scale of ex-
posure (i.e., wet and dry deposition) on a regional
as well as site-specific level. Deposition and effec-
ts monitoring (further described in  the following
chapter) is an essential  component in the standard
setting and implementing process.

Beyond these considerations regarding differences
in the spatial scale are important considerations
regarding the temporal scale. For example, the po-
tential for exposure to, and risk from, acidic cond-
itions is often highest during the spring due to the
mobilization  of the winter accumulations  of de-
posited  acids  and the activation of seasonal  bio-
logical  growth  process.  Also,  the  sensitivity  of
many resources changes over time. For example,
the most sensitive life stages of many fish species
are hatching  eggs  and newly hatched fry. The pe-
riods of greatest sensitivity for many species are
spring and fall when most fish species hatch. Simi-
larly, spring  budding  periods  for  flowers  and
leaves and initial root growth by seedlings are par-
ticularly   sensitive  periods for  many  terrestrial
plants. In contrast, many resources have low sensi-
tivity during the winter, when their biological ac-
tivity is low.  Patterns in  weather variations  may
change from  year to year and would also have to
be considered  in a  standard-setting process.  Op-
tions that  could be evaluated for appropriate aver-
aging periods which accommodate temporal issues
include single-event  loadings, seasonal loadings,
total  annual  loadings,  average  annual  loadings,
10-year average loadings  and  50-year average
loadings. The  timing of a standard, therefore,  ulti-
mately has significant  implications  for develop-
ment and  implementation of an  acid deposition
standard.

Determining  appropriate  averaging  periods must
also include  consideration of possible temporally
delayed  effects.  For example,  the  chemistry  of
spring meltwaters  may  better  reflect  accumulated
winter deposition than springtime deposition. Such
processes  could indicate the need in some regions
for  more  stringent  winter deposition standards.
Such standards might aim to minimize over-winter
accumulations of  strong-acid  anions  in  snow
packs, thereby minimizing the potential  acidity of
                                                 55

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
spring  meltwaters and their potential adverse eff-
ects on receiving waters. Likewise, in those areas
where  episodic pulses of nitrogen are the primary
concern, temporal considerations may best be fo-
cused  on a particular pollutant. Consideration  of
effects that are  cumulative in nature and  poten-
tially significantly delayed in time is a complex yet
critical component in the development of an acid
deposition standard.

2.6  CONTROLLING SULFUR AND NITROGEN TO
     REDUCE SURFACE WATER ACIDIFICATION
Atmospheric deposition  of sulfur- and nitrogen-
containing compounds, primarily SO42' and NO3~,
contributes to the  acidic deposition  problem,  as
reviewed in Section 2.2.1.  The relative contribu-
tion of sulfur and nitrogen to this problem differs
among  regions,  depending not only on external
differences in the deposition rates of  these chemi-
cals, but also on differences among the capacity of
receptor watersheds to retain sulfur and nitrogen,
as discussed in  Sections 2.2.1,  2.5.2,  and 2.5.3.
Such differences have led some  authorities (e.g.,
those in  California) to date to focus solely on the
need  to  control  nitrogen  deposition   (see Sec-
tion 2.5.4) while others have focused primarily on
sulfur.

For most regions of North America affected by and
at risk from chronic  effects of  acidic deposition,
the principal present  concern is sulfur deposition.
As more fully discussed  in Sections 2.2.1, 2.5.2,
and 2.5.3,  sulfur deposition appears to be the pri-
mary  cause of long-term chronic acidification  in
all affected sensitive areas. The NBS illustrates that
for the near term sulfur  deposition is likely to re-
main the overriding acidification problem in most
sensitive  areas  of  eastern  North America.  This
likely will remain the case until annual retention
of nitrogen decreases  sufficiently and the full po-
tential  acidifying influence  of nitrogen deposition
commences. Model  projections  indicate  that  at
such times and under  the deposition  scenarios
tested, sulfur and nitrogen  are  projected to have
approximately   equal  roles   in  surface   water
acidification. Thus, for most areas, where current
or near-term needs for additional acidic deposition
control  are projected,  and  where  watershed
nitrogen  saturation  is  not likely imminent,  the
greatest  potential  benefits will  come primarily
from control of sulfur emissions and deposition.

A significant and growing body of  scientific re-
search indicates, however, that nitrogen deposition
is a major and important contributor to  the acidic
deposition problem.  First, many areas of the West
are more affected by nitrogen deposition than by
sulfur deposition.  Second, as briefly reviewed  in
Section 2.2.1, nitrogen (in the form of nitrate an-
ion) frequently has been found to be a significant
contributor to episodic events in streams and lakes
in some parts of the  Northeast.88 In these areas,  as
effects accompanying chronic acidification due  to
sulfur deposition are reduced,  overall effects due
to episodic acidification would  likely continue  to
impair the water quality in many  of these surface
waters, but the extent of these effects would  likely
be  reduced  because reducing  the chronic  sulfur
effects also decreases potential episodic effects  as
well.  Third,  some watersheds  of the  Northeast
(e.g.,  in the Catskill  Mountains of New York) and
the mid-Appalachians may be  moving toward ni-
trogen  saturation.   For  these   regions,  nitrogen
deposition is now or would likely become a more
direct  cause of chronically acidic  conditions  in
sensitive waters, with the potential effects of acidic
sulfur and nitrogen deposition becoming approxi-
mately equal  and directly additive. In fact,  addi-
tional  limits  on nitrogen deposition would  likely
produce a two-fold  potential benefit by both re-
ducing acidic deposition rates and lengthening av-
erage  times  to  watershed nitrogen   saturation.
These  benefits would  effectively  allow  a greater
mass of NO3~ to be  deposited over longer periods
without  significantly  increasing   surface   water
acidification processes.

Scientific  uncertainties  regarding regional  rates
and differences in processes affecting watershed
assimilation  of  acid-forming sulfur and nitrogen
compounds  preclude defining either  national  or
regional protection levels below which deposition
of either chemical would produce  no significant
impact. Available  information does indicate, how-
ever, that additional  deposition reduction through-
out the range of potential reductions  in  sulfur
and/or nitrogen deposition down to background
deposition loads would likely reduce regional pro-
portions of chronically acidic surface waters  (ANC
<0 ueq/l) or proportions of surface waters potentia-
lly most sensitive to episodic effects  (ANC<50
ueq/l)  or proportions of both groups. The magni-
tude of these potential benefits to each group  of
surface waters varies considerably by region. NBS
88 This does not imply that sulfur deposition is not often a key
   component of episodic acidification,  because, as discussed
   Section 2.2.2, sulfur has often been found to be the  primary
   cause of episodic acidification in areas both within and outside
   the Northeast.
                                                 56

-------
                                                                   CHAPTER 2:  ENVIRONMENTAL GOALS
projections  indicate  for  some  regional  surface
water  groupings  that  potential   benefits   may
amount to a few percentage points shift in propor-
tions of acidic or sensitive surface waters benefit-
ing, while for other groupings in other regions po-
tential  benefits from deposition  reductions could
benefit 20 percent or more of the acidic or sens-
itive waters.  Note,  however,  that  even a  few
percentage points may mean many lakes or miles
of stream  reaches. Now, however, even a sound
qualitative ranking of  these differences  awaits
resolution  of key scientific unknowns, exemplified
by the marked uncertainty associated with quanti-
fying regional differences in their remaining times
to watershed nitrogen saturation.
                                                57

-------
                                          CHAPTER 3

                 SOURCE-RECEPTOR RELATIONSHIPS AND DEPOSITION
                 REDUCTIONS UNDER VARIOUS EMISSIONS SCENARIOS
3.1 INTRODUCTION
Acidic deposition results from a complex series of
interactions among chemical compounds present
in the atmosphere. The two most important groups
of chemicals are sulfur-  and nitrogen-containing
compounds. These  chemicals are emitted to the
atmosphere  from  natural  and  anthropogenic
sources, with anthropogenic sources dominating.
Natural sources include vegetative emissions, geo-
thermal activities, forest fires, lightning, soil, and
salt-water organisms.  Anthropogenic sources in-
clude  point sources such  as  utilities, industrial
boilers and  other  industrial  processes,  mobile
sources including  automobiles,  trucks, and off-
highway vehicles,  and  area sources such  as resi-
dential boilers which are  too small and numerous
to track individually.

Sulfur and nitrogen  compounds in the atmosphere
can be transported hundreds to thousands of kilo-
meters by meteorological forces. During transport
sulfur dioxide (SO2)  and nitrogen oxides  (NOX), the
primary emissions of these species, are oxidized in
the air or in cloud-water to form other,  secondary
compounds. The great majority of these  com-
pounds, particularly sulfate and nitrate, are acidic.
The oxidizers, such as the hydroxyl radical,  hydro-
gen peroxide, and ozone are produced by reac-
tions of volatile organic  compounds (VOC) and
NOX. The sulfur and nitrogen pollutants  are depos-
ited to the  earth by either dry or wet deposition.
Dry deposition occurs when particles settle out of
the air onto the earth or when gases or fine parti-
cles directly impact  land, plants, or water, or when
plant stomata take up gases, such as SO2.  In wet
deposition,  pollutants are removed  from  the at-
mosphere by rain or snow. Fine particles or secon-
dary aerosols formed by these same processes scat-
ter or absorb visible light and thus impair visibility.
When  inhaled these secondary aerosols and their
gaseous precursors can  also cause adverse human
health  effects.  Potential benefits  to visibility, hu-
man health, materials, and cultural resources from
controlling   acidic  deposition  are  discussed  in
Chapter 4.
The complex relationship between emissions and
deposition depends on a large number of physical,
chemical, and biological processes. To understand
the environmental impact of the CAAA and to de-
velop and analyze strategies to reduce the effects
of  acidic deposition,  the  relationship  between
emissions and deposition must be understood not
only for the present, but also for future years. To
predict deposition  a model  must be able to de-
scribe  the  transformation  of anthropogenic and
biogenic emissions  by atmospheric processes  to
wet and dry deposition.  The goal of deposition
modeling is to simulate the source-receptor  rela-
tionships that  translate emissions into deposition
values  in space and over time. An understanding
of these complex interactions is  necessary to de-
velop a comprehensive approach to achieving the
environmental goals discussed in Chapter 2.

This chapter addresses  the following  requirement
of Section  404 (Appendix  B) of  Title  IV of the
CAAA:

  * Description of the state of knowledge with
     respect  to source-receptor  relationships
     necessary to develop a control program on
     such standard or standards and  additional
     research  that  is  on-going  or  would be
     needed to make such a control program
     feasible

Section  3.2 describes state-of-the-art  atmospheric
modeling techniques, the uncertainties associated
with predictive modeling, precursor emissions and
emissions inventories, and deposition species. This
section describes RADM, the atmospheric model
used for this study, and  examines  its use in model-
ing of acidic  deposition and the results of  its
evaluation.

Section  3.3  presents RADM results that  explore
and define source-receptor relationships. Relation-
ships prior to implementation of the  1990 CAAA
are compared and contrasted to those  expected af-
ter  full implementation of the Act in 2010. The
model discussion is followed in Section  3.4 by a
description of the inventories used to evaluate al-
ternative emissions control scenarios.
                                               59

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
In Section 3.5, source-receptor relationships from
RADM are used  to assess the environmental im-
pact of the CAAA and to predict deposition values
for acidic sulfur and  nitrogen under various  alter-
native emissions scenarios for the year 2010.  Emis-
sions scenarios analyzed include full implementa-
tion of Title IV with trading, full implementation of
Title IV without trading, national reductions in util-
ity and industrial SO2 emissions beyond  Title IV,
and  national reductions in  utility and  industrial
NOX  emissions  beyond  Title IV. Regionally tar-
geted emissions reductions for SO2  are examined
in Section 3.6.

3.2 THE  REGIONAL ACID DEPOSITION MODEL
RADM has been developed over the past  10 years
under the auspices of the National Acid Precipita-
tion  Assessment Program (NAPAP) to address pol-
icy and  technical  issues  associated with  acidic
deposition. The model is  designed to  provide a
scientific basis for predicting changes in deposition
resulting  from changes in  precursor emissions, to
predict the influence of emissions sources in one
region on acidic deposition in other geographic
regions, and to predict the levels of acidic deposi-
tion  in certain sensitive receptor regions.

A key requirement for the  model selected for NA-
PAP was the  ability to assess changes in sulfate in
response to projected changes in SO2 emissions.
Based on knowledge  gained from models and ex-
perimental measurements, a  reduction  in  SO2
emissions is expected to lead to a less than equiva-
lent  reduction  in  sulfate  deposition. This  non-
equivalency  between emissions  reductions  and
decreases in  deposition is  due to the nonuniform
spatial distribution of  emissions reductions and the
complexities  of  atmospheric  chemistry.  While
simpler  models can  predict the  nonequivalency
due  to spatial nonuniformities, a  complex model
such  as  RADM is needed to calculate the  addi-
tional affects of atmospheric chemistry on deposi-
tion.

The  development, application, and evaluation of
RADM has been  documented extensively by NA-
     89,90,91 RADM  continues to undergo  periodic
peer   reviews,   evaluations,   and   improve-
ments.92-93'94 Understanding and modeling acidic
deposition  requires  consideration of  a complex
range of physical and chemical processes and their
interactions, including:

  * The emissions of  precursor chemicals that
     produce  and  regulate acidity  in  atmos-
     pheric deposition;

  » The meteorological processes that transport
     and mix emitted species in space and time;

  » The physical and  chemical transformations
     that alter the physical phases and chemical
     properties of emitted species;

  * The meteorological factors and properties
     of the  Earth's surface that  lead  to deposi-
     tion of  acidic substances.

RADM is an Eulerian model in  which concentra-
tions of gaseous and particulate  species are calcu-
lated  for specific  fixed  positions in  space  (grid
cells) as a function of time. The concentration of a
specific pollutant in  a grid cell at a specified time
is  determined  by: the  emissions input rate;  the
transport of  that species by wind  into and out of
the grid in three dimensions; movement by turbu-
lent motion  of the atmosphere; chemical reactions
that either produce  or deplete the chemical;  the
change in concentration due  to vertical transport
by  clouds;  aqueous  chemical transformation and
89 Renne, D., et al. December 1989. Models Planned
   for Use in the NAPAP Integrated Assessment. Sec-
   tion 4: Atmospheric Models. National Acid Precipi-
   tation Assessment Program.
90 Chang, J.S., P.B.  Middletori, W.R.  Stockwell, C.J.
   Walcek, J.E. Pleim, H.H. Lansford,  F.S. Binkowski,
                                      (continued)
   S. Madronich,  N.L. Seaman, and D.R. Stauffer. De-
   cember 1990.  The Regional Acid Deposition Model
   and Engineering Model. SOS/T Report 4. In: Acidic
   Deposition: State of Science  and Technology. Na-
   tional Acid Precipitation Assessment Program.
91  Dennis,  R.L,  W.R.  Barchet,  T.L. Clark,  and S.K.
   Seilkop.  September  1990.  Evaluation  of Regional
   Acid Deposition Models (Part  I). SOS/T Report 5. In:
   Acidic Deposition: State of Science and Technology.
   National Acid Precipitation  Assessment Program.
92  Dennis,  R.L.,  J.N. McHenry, W.R. Barchet, F.S.
   Binkowski,  and  D.W.  Byun.  1993.  Correcting
   RADM's sulfate underprediction: Discovery and cor-
   rection of model errors and testing the  corrections
   through comparisons against field data. Atmospheric
   Environment 27A(6):975-997.
93  McHenry, J.N., and R.L. Dennis. 1994. The relative
   importance of  oxidation pathways and clouds to at-
   mospheric ambient sulfate  production as  predicted
   by the Regional Acid Deposition Model. Journal of
   Applied Meteorology 33(7):890-905.
94  External Review Panel report  on RADM evaluation
   for the Eulerian Model  Evaluation Field Study Pro-
   gram.
                                                 60

-------
                                               CHAPTER 3: SOURCE-RECEPTOR RELATIONSHIPS AND DEPOSI-
                                                 TION REDUCTIONS UNDER VARIOUS EMISSIONS SCENARIOS
scavenging; and removal by dry deposition. These
physical and chemical process are depicted in Ex-
hibit 16.
sional  data  assimilation  (4DDA)  to  produce the
most accurate recreation of past weather.95 An ag-
gregation technique, described fully by Dennis et
             EXHIBIT 16. PHYSICAL AND CHEMICAL PROCESSES CONTRIBUTING TO ACIDIC DEPOSITION
                     SOURCES
s~
\ '
Gaseous
Pollutants


in
Atmosphere
|W- Dry D'-xJsition
^
Depo
't h
\

\/
Pollutants
Cloud Wat
and
Precipitatu
Wet
sition v
M 1 A .
\ '
Paniculate
Pollutants
in
Atmosphere
/ e
o
VI
o
in ex
er S
>~
3n £
i
. — • —
                                            Natural
                 RECEPTORS
              Anthropogenic
The version of RADM used for NAPAP and for the
analyses presented in this report covers an area of
2,800 by 3,040 km east of central Texas and south
of James Bay,  Canada to the southern tip of Flor-
ida. RADM divides this area into 80 by 80 km grid
cells.  For sulfur deposition modeling the  distance
from ground  level to 16  km in  altitude  was re-
solved into 6 vertical layers; for nitrogen deposi-
tion 15 layers were used. The RADM domain, pic-
tured in Exhibit 17, consists of 35 by 38 horizontal
grid cells. The  model is run with either 7,980 or
19,950 cells depending upon whether 6 or 15 ver-
tical layers are employed.  For each grid cell, pre-
dictions are generated at dynamically determined
time-steps of  seconds to minutes and are output
hourly by RADM with 41  chemical  species being
transported. Hourly wet  and dry deposition values
are also generated for each surface cell for 12 spe-
cies (6 wet and 6 dry).

The meteorological fields used to drive the RADM
are from  the  Pennsylvania  State University-Na-
tional Center for Atmospheric Research Mesoscale
Model (MM4).  The MM4 is run  using 4-dimen-
al., developed during NAPAP is used to develop
annual estimates of acidic deposition (see footnote
91). Meteorological cases with similar wind flow
patterns were grouped by applying cluster analysis
to classify the wind flow  patterns from 1982  to
1985, resulting in  19 sampling groups, or strata.
Meteorological cases were randomly selected from
each stratum; the number  selected was based on
the number of wind flow patterns in  that stratum
relative  to the number of patterns in each of the
other  strata, to  approximate  proportionate  sam-
pling. A total of 30 cases were used in the current
aggregation approach. Deposition results for these
cases were weighted according to the strata  sam-
pling frequencies to form annual averages.
95  Seaman,  N.L.,  and D.R. Stauffer. 1989. Develop-
   ment of Four-Dimensional Data Assimilation for
   Regional Dynamic Modeling Studies.  Final Report to
   the U.S. Environmental Protection Agency, Contract
   CR-814068-01, the Pennsylvania State  University,
   102 pp.
                                               61

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                              EXHIBIT 17. THE RADM MODELING DOMAIN
Development and  implementation of an  acidic
deposition control strategy requires an understand-
ing of the interaction among emissions of several
chemical species, the spatial and temporal patterns
of those emissions, and atmospheric transport of
those species to regions where deposition occurs.
It is important, therefore, to understand the uncer-
tainty and reliability of model predictions. The fol-
lowing  subsections briefly  describe the key input
and functional components upon which RADM
predictions are based. The discussion on emissions
and atmospheric  chemistry explains why certain
emissions are important, distinguishes anthropo-
genic  (controllable) from  natural  (background
sources), and describes why detailed spatially and
temporally  resolved  emissions   inventories  are
needed to accurately predict deposition. Subsec-
tions 3.2.2 and 3.2.3 describe the development of
source-receptor relationships and  how these rela-
tionships can be used to identify emissions sources
responsible for deposition, whether proximate or
hundreds of kilometers distant. The reliability and
                                                62

-------
                                               CHAPTER 3:  SOURCE-RECEPTOR RELATIONSHIPS AND DEPOSI-
                                                  TION REDUCTIONS UNDER VARIOUS EMISSIONS SCENARIOS
confidence of the scientific  community in  using
modeling results for decision-making is discussed
in subsection 3.2.4.

3.2.1  Emissions and Atmospheric Chemistry
The principal  acids in deposition  are  sulfuric
(H2SO4)  and  nitric (HNO3) acids. Thus, emissions
of compounds containing sulfur and nitrogen have
received primary  emphasis  in acidic deposition
control  strategies.  However, emissions of VOCs
and their oxidation products are extremely impor-
tant because they are involved in  reactions that
produce  the oxidizing species that lead to forma-
tion of sulfuric and nitric acids in the atmosphere.
Key environmentally important species predicted
by RADM are:

   * AMBIENT CONCENTRATIONS: SO2, NO,  NO2,
     HN03, 03, H202, NH3, PAN, HCHO, CO,
     aerosol SO42~

   * WET DEPOSITION: SO42-,  NO3-  as HNO3,
     NH3, H+

   * DRY DEPOSITION: SO2,  SO42-,  HNO3, O3,
     NO2

The RADM chemistry component consists of 140
reactions among 60 species, 40 of  which are or-
ganic compounds.  Chemical decomposition by so-
lar radiation (photolysis) is  included in the model
chemistry as  are aqueous-phase reactions which
occur in  clouds.  These latter reactions are particu-
larly  important  in  sulfuric  acid formation.  The
chemically   derived  nonequivalency  between
emissions reduction and deposition decreases is
due in part to local depletion of hydrogen perox-
ide, a compound produced  by atmospheric photo-
chemical (involving sunlight) reactions that is  im-
portant in the oxidation of  aqueous-phase SO2 to
sulfate.96 This nonlinear oxidant limitation affects
only  wet deposition of sulfate. Details  of  the
RADM chemistry are described in NAPAP State of
Science  and  Technology Report  4 (see footnote
90).

Inputs to RADM include hourly emissions  of SO2,
sulfate, nitric oxide (NO), nitrogen dioxide (NO2),

96 McHenry, J.N., and R.L. Dennis. 1994. Cloud  and
   chemistry pathway characterization of the nonlinear
   response of sulfur deposition and sulfate air concen-
   trations to changes in SO2  Emissions in the RADM.
   In Atmospheric Chemistry Extended Abstracts. AMS
   Conference  held  January  1994,  Nashville,   TN.
   pp. 203-208.
ammonia  (NH3), carbon  monoxide (CO), particu-
late matter,  and 15 classes of VOCs from natural
and anthropogenic sources.  Emissions  inventories
used for RADM were derived from the 1985 NA-
PAP Emissions Inventory, the most comprehensive,
highest quality air  emissions inventory  ever as-
sembled. (RADM inputs have since been updated
to include more recent sector-specific  inventories
such as the  National Allowance Data Base devel-
oped under the Acid Rain Program  for utility emis-
sions.)  The basis and key assumptions for the 1985
NAPAP Emissions Inventory are described below.
A detailed description of this  inventory and the
data processing for use in RADM are described in
NAPAP State  of Science  and Technology  Re-
port 1,97

Natural Emissions Sources
Natural emissions of acidic precursor species, or-
ganic matter, and alkaline materials (dust) are gen-
erated  by vegetative  matter, microbes, geothermal
activity, natural combustion (such  as forest  fires),
lightning, and salt-water organisms. These sources,
in contrast to anthropogenic sources, are widely
distributed, small, sporadic, and subject to large
seasonal  and  weather-related  variations.  Total
VOC  emissions from biogenic  sources are  esti-
mated  to  be of the  same order of magnitude  as
VOC   emissions from anthropogenic  sources.98
Principal  biogenic   sources   are  trees,  shrubs,
grasses, agricultural crops, decaying leaf litter, and
vegetation in fresh and salt water.  Some biogenic
compounds  are very reactive in the atmosphere;
others  are relatively inert. Emissions for individual
grid cells  and specific VOCs or VOC classes are
calculated   from   estimates   of  biomass   data
(vegetation type, species, land-use coverage, and
leaf area  index), adjusted for seasonal variation,
temperature, solar intensity, soil conditions includ-
ing moisture, and elevation.

Natural emissions of SO2,  sulfates, and  nitrogen
oxide  have been found to be less  important than
anthropogenic sources. Natural emissions of sulfur,
which  are not well understood, are  estimated to be
97 Placet, M., R.E.  Battye, and F.C.  Fehsenfeld. De-
   cember 1990.  Emissions Involved in Acidic Depo-
   sition Processes. SOS/T Report 1. In: Acidic Deposi-
   tion: State of Science  and Technology.  Volume I.
   National Acid Precipitation Assessment Program.
98 Novak,  J.H., and  T.E.  Pierce.   1993.  Natural
   Emissions of Oxidant Precursors. In Water, Air, and
   Soil Pollution. Vol. 67, pp. 57-77.
                                                63

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
about 6 percent of anthropogenic emissions and
are not included in the  inventory.  Natural  emis-
sions  of  nitrogen compounds, which are  also
poorly characterized,  are estimated to be 1  to 2
percent of total nitrogen emissions. The 1985 NA-
PAP inventory does  include  nitrogen  emissions
from soil because these emissions have been stud-
ied, although  large uncertainties in estimates still
exist.  Lightning-generated NOX may  be episodi-
cally important, but is not understood well enough
to be  included.  Alkaline dust emissions  (from
sources such as erosion  and unpaved roads) are
also not included  in the inventory. The principal
natural source of ammonia, an important nitrogen-
containing atmospheric reactant, is animal excre-
ment, from feedlots, farm animals, and wildlife;
however, the paucity of data on wildlife densities
does not allow this source  to  be  estimated  with
any degree of reliability. "Natural"  emissions  of
ammonia are included  in  the inventory, but the es-
timates  have a large  degree of uncertainty.  Am-
monia is considered in  RADM simulations because
it is an  important  contributor to nitrogen  deposi-
tion and affects rainwater pH and the production
of secondary sulfate and nitrate aerosols.

Anthropogenic Emissions Sources
Acidic deposition  precursor species and reactive
atmospheric chemicals are  generated by energy
production, industrial  processes,  mobile sources,
and waste disposal. Major sources  (such as power
plants) that emit large quantities of pollutants at
specific, well-defined  locations are called "point
sources." Small emissions sources such as residen-
tial boilers  and mobile  sources are  grouped to-
gether as "area sources." In general, emissions from
point sources are reported to EPA by states, while
EPA calculates emissions from  area  sources. The
RADM model includes a  species allocation  mod-
ule" which accounts for the small amount of pri-
mary  sulfate (about 1  percent  of  SO2 emissions)
that is emitted with SO2. This primary sulfate is
negligible in comparison to sulfate formed in the
atmosphere. Similarly, the allocation module splits
NOX emissions into a 5:95 ratio of NO2 to NO. As
is the case with  biogenic VOC emissions,  some
VOC species are very  reactive  in the  atmosphere,
others are not. The model calculates VOC species
or classes by grid cell  based on source category.
Major emissions source categories in the United
States and their contribution to total emissions of
each pollutant in the year 1992 are summarized
below for SO2, NOX, and VOC.100

   * S02
     » Electric Utility Fuel Combustion (69.7%)
     # Industrial Fuel Combustion (13.6%)
     # Metals Processing (3.8%)
     • Highway Vehicles  (3.5%)
     » Other Fuel Combustion (2.6%)
     * All Other Sources (6.8%)

   * NOX
     * Electric Utility Fuel Combustion (32.3%)
     * Highway Vehicles  (32.3%)
     » Industrial Fuel Combustion (15.2%)
     * Off-Highway Vehicles (12.3%)
     » Other Fuel Combustion (3.2%)
     » All Other Sources (4.7%)

   * VOC
     » Highway Vehicles  (26.8%)
     » Solvent Utilization (26.7%)
     » Waste Disposal and Recycling (10.2%)
     » Off-Highway Vehicles (9.4%)
     » Storage and  Transport (8%)
     * All Other Sources (18.9%)

Temporal and Spatial Allocation of Emissions
Data
To create modeling inventories the annual  inven-
tories must be resolved spatially, temporally, and
by chemical  species. Allocation of  emissions to
grid cells for point sources is relatively straightfor-
ward. The geographic (latitude-longitude) location
of  each  point source  determines   its  grid  cell
placement. Since area sources are too small to be
included  individually  in  the  annual  inventory,
emissions from these  sources  are calculated  by
multiplying an emissions factor by an activity  pa-
rameter which reflects the operating rate of each
source. The activity parameter is determined from
surrogates such as population (number of gasoline
service stations), housing (residential  fuel combus-
tion), and agricultural  land  area (ammonia fertil-
izer application). A process was developed  by
99 Walters, R., and M. Saeger. 1990. The NAPAP Emis-
   sions Inventory: Development of Species Allocation
   Factors.  EPA-600/7-89-010f.  U.S.  Environmental
   Protection Agency, Research Triangle Park, NC.
100Office of Air Quality Planning and Standards. Octo-
   ber 1993. National Air Pollutant Emissions Trends,
   1900-1992.  EPA  Report No. 454/R-93-032.  U.S.
   Environmental Protection Agency. Research Triangle
   Park, NC.
                                                64

-------
                                               CHAPTER 3: SOURCE-RECEPTOR RELATIONSHIPS AND DEPOSI-
                                                  TION REDUCTIONS UNDER VARIOUS EMISSIONS SCENARIOS
EPA101 to apportion area source emissions to indi-
vidual grid cells from county-level data using these
surrogate parameters. Exhibit 18 is a density map
of SO2 emissions for the RADM region.


      EXHIBIT 18. MAP OF ANNUAL SULFUR EMIS-
        SIONS DENSITY IN 1985 (TONS/YEAR)
                            45000 - 160000 |
Day- and hour-specific gridded emissions are cre-
ated for each of the 30 RADM aggregation cases.
Temporal allocation factors  were developed for
NAPAP102  that  provide  day-specific   estimates
based on tabulation of representative relative diur-
nal emissions patterns by day of the  week and by
season for each source. Where emissions strongly
respond to temperature and other meteorological
conditions, the  day-specific meteorology is used in
the emissions estimation procedure.  Examples in-
clude  volatile   organics  that  evaporate  easily,
sources for which ambient conditions affect per-
formance and  hence emissions,  such as mobile
sources, and sources for which temperature affects
biological processes, such as biogenic emissions,
Plume rise from major industrial and utility sources
101Modica, L, and D.R. Dulleba. April 1990. The 1985
   NAPAP Emissions Inventory: Development of Spatial
   Allocation Factors. EPA Report No. 600/7-89-01 Ob.
   Air and Energy Research Laboratory, U.S.  Environ-
   mental Protection Agency, Research  Triangle Park,
   NC.                                <
102  Fratt,  D., D.F. Mudgett, and  R.A. Walters. 1990.
   The   1985   NAPAP   Emissions   Inventory:
   Development of Temporal Allocation Factors. EPA
   Report  No.  600/7-89-01 Od.   U.S.  Environmental
   Protection Agency, Research Triangle  Park, NC.
is also computed hourly, based on the hourly me-
teorology of each day.

3.2.2  Modeling Source-Receptor Relation-
       ships and Source Attribution
Eulerian, or fixed-grid models, are very suitable for
representing the full, complex nonlinearity of the
photochemistry  involved in the oxidation  of pri-
mary emitted species  to acidic  substances. The
gas- and aqueous-phase oxidation of sulfur  is non-
linear; the nonlinearity comes from  competition
for  scarce oxidizers, such  as hydrogen peroxide.
The most accurate modeling of source-receptor re-
lationships must maintain the overall  competition
for  oxidants represented by the concentrations
produced by all  the SO2 emissions, while tracking
the particular emissions from the  SO2 sources of
interest.  In  other words,  the  influence  of SO2
sources cannot be studied individually, but must
be  considered altogether. This is  because, for  a
nonlinear system,  the sum  of the deposition from
individual point  source emissions  is not expected
to be the same as the total deposition  from all
point emissions computed simultaneously.

Eulerian  models  have not historically been used to
study source-receptor  relationships.  The Tagged
Species Engineering Model103 was developed  un-
der  NAPAP to  study  such relationships.  The
Tagged  Species  Model  gives the Eulerian  RADM
modeling system the capability to  identify,  for as-
sessment purposes, the  concentration  and deposi-
tion fields attributable  to specified SO2 emissions
source regions in the presence of the full concen-
tration fields.  The Tagged  Model preserves  the
oxidant competition across space and  time. A tag-
ging concept is applied in which additional, iden-
tical mass conservation equations are  solved for a
portion of the sulfur concentration field that origi-
nates from specific geographical locations  within
the full modeling domain. This allows tagged con-
centration fields and tagged  wet and dry deposi-
tion to  be  identified and  tracked in  the  model
separate  from, yet as portions of,  the  total sulfur
chemical environment that  is nonlinear  and that
produces the complete  concentration  and deposi-
tion fields.  Exhibit 19  shows the  tagged  RADM
regions created for the  Engineering Model  and
103McHenry, J.N.,  F.S.  Binkowski,  R.L. Dennis,  J.S.
   Chang, and  D. Hopkins.  1992. The tagged species
   engineering  model   (TSEM).  Atmospheric Envi-
   ronment 26A(8):1427-1443.
                                                65

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                                  EXHIBIT 19. TAGGED RADM SUBREGIONS*
                  a Geographical description of numbered subregions:
                     1    Montreal Area                  28
                     2    NH/MA Border                  29
                     3    NYC/CT                        30
                     4    Southeast NY                   31
                     5    Southern Tier NY                32
                     6    Niagara Area                   33
                     7    Detroit Area                    34
                     8    Central/Southern NJ              35
                     9    Eastern PA                      36
                    10    Southern PA/MD                37
                    11    Central PA                      38
                    12    Northwest PA                   39
                    13    Southwest PA/Northern WV       40
                    14    Cleveland Area                  41
                    15    OH/WV/PA Border              42
                    16    Northwest OH/Eastern IN         43
                    17    Chicago Area                   44
                    18    Eastern VA                     45
                    19    Western VA/Eastern WV          46
                    20    OH/WV/KY Border              47
                    21    KY/WV/VA Border               48
                    22    Cincinnati Area                 49
                    23    Central KY                     50
                    24    Central IL/IN Border              51
                    25    Southwest IN/KY Border          52
                    26    St. Louis Area                   53
                    27    Northeast NC
Northwest NC
Blue Ridge Area
NC/TN/CA Border
Central TN
Central TN/KY Border
Central TN/AL Border
IL/MO/TN Border
Memphis Area
Southeast SC
Central SC/NC Border
Northeast GA
Northeast AL/Northwest CA Border
Western  AL
Mobile Area
Baton  Rouge Area
Northeast TX
Lake Ontario/NY Shore
Adirondacks
VT/NH
Southeast NC
Southeast GA
Southern AL/GA Border
Northern MS
Northern FL Peninsula
FL Panhandle
Southern MS
                                                     66

-------
                                              CHAPTER 3: SOURCE-RECEPTOR RELATIONSHIPS AND DEPOSI-
                                                 TION REDUCTIONS UNDER VARIOUS EMISSIONS SCENARIOS
their  geographical  description.  The use  of  the
Tagged Model in this study represents the first ex-
tensive use of a Eulerian model to study source-re-
ceptor relationships.

3.2.3  Transport, Chemistry, and Source-Re-
       ceptor Relationships
The lifetime in the atmosphere of the sulfur diox-
ide from  an emissions source is several days. The
main  loss  mechanisms  are the  incorporation  of
SO2 and  sulfate in clouds and the subsequent wet
deposition  and dry deposition  of these  species.
Thus,  SO2  is transported over many hundreds of
kilometers before concentrations from a source are
substantially reduced by wet and dry deposition.
As a result, the deposition at any one receptor area
is  coming from a very  large number of  sources,
spread over a large geographic  area. Source re-
sponsibility cannot be determined, therefore, from
monitoring data. Too much is mixed together. Al-
though rough estimates of source-type responsibil-
ity can be  developed with source apportionment
techniques  using  unique chemical  "fingerprints,"
all assessments of source responsibility or source
attribution require the use of an air quality model.

Calculations from the Tagged Species Model illus-
trate  the  distances  over  which  SO2  emissions
sources can have an influence. A map of the com-
puted proportion of total annual  sulfur  deposition
contributed by the Ohio/West Virginia/Pennsylva-
nia source  subregion (RADM Subregion 15) along
the upper  Ohio  River  Valley  is  shown  in  Ex-
hibit 20.  Exhibit 21 shows the distance covered to
deposit increasing fractions of the total  deposition
in  the eastern United States. The subregion's range
of  influence is more than 1,000 km. Typically, the
range of  influence of a  sub-region extends out to
between  500 and 1,200km. The Tagged Species
Model analyses indicate that about two-thirds of
the total sulfur  deposition  from major  sources
along the Ohio River Valley occurs within 500 to
700 km.  For the southern source regions, the dis-
tance  to deposit about two-thirds of a  source's spa-
tially integrated deposition  is somewhat  less, about
300 to 500km. The difference  in scale of influ-
ence is primarily due to meteorology.

A  number of meteorological factors influence the
existence of dominant transport directions and de-
termine how a group of  SO2 emissions sources in-
fluence nearby regions.  Key factors are the posi-
tion of the  jet stream, which moves storms across
the upper Mid-West; the influence of the Appala-
     EXHIBIT 20. PROPORTION OF ANNUAL SUL-
   FUR DEPOSITION CONTRIBUTED BY RADM SUBRE-
      GION 15 (OH/WV/PA BORDER REGION)
    EXHIBIT 21. PERCENTAGE CUMULATIVE RANGE
      OF INFLUENCE OF RADM SUBREGION 15
          (OH/WV/PA BORDER REGION)
chian Mountains  on winds and  rainfall patterns;
the Bermuda  highs  (stagnation) that move Ohio
River Valley  emissions in a clockwise direction;
and the ocean  and Gulf Coast weather that pro-
duces lighter winds and  more  convective  con-
ditions,  including a typically large proportion of
                                               67

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
corrective clouds across the southern states. As a
result, source-receptor relationships are skewed to
the northeast  (Exhibit 22a)  starting  in  northern
West Virginia and show a predominantly northerly
to easterly  quadrant  of  flow  along  the  lower
Ohio River  Valley (Exhibit 22b),  yet  are nearly
symmetrical  in the Southeast (Exhibit 22c). Thus,
the patterns  and ranges of source influence can
vary. Models, such as those in the RADM system,
help to  interpret and  explain the deposition at re-
ceptors of interest.

An alternative  approach  to  determining  efficient
and cost-effective strategies to achieve deposition
targets relies on the use of optimization models.104
An optimization model for acidic deposition could
simultaneously minimize SO2 removal  cost and
average exceedance of target deposition rates over
the receptor model  domain.105  Such  a  model
could calculate  costs  and  emissions reductions
necessary to achieve  a regionally averaged target
load comparable to the average  annual deposition
level calculated by RADM. This model was inves-
tigated for this report but not used because of the


     EXHIBIT 22A. SOURCE-RECEPTOR RELATION-
        SHIPS IN THE NORTHEAST: CUMULA-
         TIVE PERCENT SULFUR DEPOSITION
104Streets, D.C., D.A. Hanson, and L.D. Carter. 1984.
   Targeted strategies for control of acidic deposition.
   Journal of the Air  Pollution  Control  Association
   34(12):1187-1197.
105 Ellis, J.H. 1988. Multiobjective mathematical  pro-
   gramming models for acid  rain control. European
   Journal of Operational Research 35(3):365-377.
     EXHIBIT 22e. SOURCE-RECEPTOR RELATION-
    SHIPS IN THE LOWER OHIO VALLEY: CUMULA-
         TIVE PERCENT SULFUR DEPOSITION
EXHIBIT 22c. SOURCE-RECEPTOR RELATIONSHIPS IN THE
SOUTHEAST: CUMULATIVE PERCENT SULFUR DEPOSITION
computational difficulty in combining the  non-
linear RADM transfer coefficients into a linear pro-
gramming optimization model. Although an  opti-
mization model  could  have been employed using
linear transfer coefficients, at the time this report
was being developed,  no  linear  transfer  coeffi-
cients that approximated the RADM transfer  coef-
ficients  were available. Optimization models are,
however, used extensively when important critical
                                                68

-------
                                               CHAPTER 3:  SOURCE-RECEPTOR RELATIONSHIPS AND DEPOSI-
                                                  TION REDUCTIONS UNDER VARIOUS EMISSIONS SCENARIOS
or target load decisions have already been made
(e.g.,  in European countries). Development of lin-
ear transfer coefficients consistent with RADM are
beyond the scope of this study.

3.2.4  Confidence in Results
The United States and Canada have been  cooper-
ating  in a  bi-national Eulerian Model Evaluation
Field  Study (EMEFS) and  model evaluation  since
1985.53 U.S.  data collection for EMEFS was con-
ducted under the auspices of NAPAP.  Phase  1  of
the evaluation of the two advanced acidic deposi-
tion models  ADOM  (Acid  Deposition and Oxi-
dants Model) and RADM  (Regional Acid  Deposi-
tion Model) has been  completed and described.106
Phase 1 was primarily based on 1988 EMEFS data.
Improvements to RADM resulted and  uncertainty
analyses regarding the scientific assumptions in the
model were prepared  that went beyond the earlier
NAPAP results (outlined in reference 3). Phase  2,
the final phase of the evaluation,  has  been  com-
pleted and is based on  the 1990  EMEFS  data  as
well as data of high diagnostic value from the
1988 period. A cooperative effort in the evaluation
of the advanced models will continue under the
U.S.-Canada Air Quality Agreement.

At the end of Phase 1  and Phase 2, the evaluation
results were subjected to  extensive external peer
review. The  review  focused  on  how well the
evaluation  had followed the preset protocol, the
resulting credibility of the models, and the appro-
priateness of model applications. The review panel
members determined they would have  confidence
in the ability of the models to represent (1) total
sulfur loading of the  atmosphere,  suggesting that
the emissions inventory and average lifetimes  of
sulfur species are roughly correct; (2) annual sulfur
deposition, although there is some seasonal bias,
with deposition  being underestimated  in summer
and overestimated in winter; and (3) total nitrogen
loading of the atmosphere, suggesting that the ni-
trogen budget and  average  lifetime  of  nitrogen
species is roughly correct.

From  the  Phase  1  review, the external  review
panel felt that the models could be used for esti-
mating annual deposition of sulfur and nitrogen.
106 pacjfjc Northwest Laboratories. 1991  The Eulerian
   Model Evaluation Field Study.  Interim report PNL-
   7914.  Prepared for the  U.S. Environmental Protec-
   tion Agency,  Research Triangle  Park,  NC.  IAC
   DW89933040-01, 81 pp.
The reviewers concluded that, while both models
were  adequate for the study of large-scale and
long-distance  source-receptor  relationships, the
models are so complex that their application to the
problem of source-receptor relationships might be
more  limited. A valuable application would be for
the complex models to serve as a check on the ac-
curacy of the simpler, but much faster, Lagrangian
methods or on the source-receptor matrices pro-
duced by the Lagrangian methods. Regarding non-
linearities  in  the  sulfur deposition,  the  reviewers
agreed that they may be of the order of 10 percent
to 15  percent.  However, they noted there is no ob-
servational means of fully testing the validity of the
model estimates of the relatively modest nonlinear
effects. The panel concluded that a great deal had
been  accomplished and significant  improvements
had been  made  in the models during the evalu-
ation  process. The protocol and evaluation have
been essential in winning scientific support for the
models, and the models have shown evidence of
converging towards operational use. This process,
however, is more advanced for sulfur than for ap-
plications involving nitrogen.107

As part of  the model  evaluation process,  bounding
studies were performed to assess the risk that the
predicted changes in air concentrations and depo-
sition  would  be sensitive to uncertainties  in the
scientific descriptions in RADM (see footnote 91).
The bounding analysis showed that errors that
could contribute  to  scatter  in  comparisons with
measurements do not result in  the  same level of
uncertainty in the relative change predictions. The
uncertainty  regarding  the  predicted change  is
much less.  The effect of known model errors that
could contribute most  to biasing or changing the
model's sensitivity to  emissions change was ex-
plored.  The results were that the bounded  range
of RADM predictions of relative change is  roughly
10 percent around the best estimate of deposition
change.  That is, a model  estimate of change in
sulfur deposition of 40 percent  has an uncertainty
the order  of ±4  percent.  There is  greater confi-
dence in the upper bound, but less in the  lower
bound because it is affected by our lack of com-
plete  understanding  of the  nonlinear processing
affecting   sulfur  deposition.  The  narrow  range
would suggest that there is little  risk that the model
will misguide users regarding the predicted change
107 U.S. Environmental Protection Agency. 1994. Pro-
   gress Report  for  the  U.S.-Canada  Air  Quality
   Agreement.
                                                69

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
in sulfur deposition, despite shortcomings uncov-
ered in the model evaluation. This appraisal of the
bounding results  may  change if  significant  new
insight or knowledge develops in the future.

3.3  SOURCE ATTRIBUTION
3.3.1  Changes from 1985 to 2010
Fifty-three tagged regions were identified  in Ex-
hibit 19  for  which  the major point source  SO2
emissions (utility  and major industry) have been
explicitly tracked, using the RADM Tagged  Spe-
cies Engineering Model. These regions account for
84 percent of the  major point source  emissions in
the United States during  1985 and 76 percent in
2010. Exhibit 23  shows the percentage contribu-
tions of the 53 tagged regions to U.S. major point
SO2 emissions in  the RADM domain, to total  U.S.
SO2  emissions in  the RADM domain, and to all
North American SO2 emissions in the Northeast,
respectively. Thus, although the 53 tagged regions
accounted for three-quarters of the 1985 total SO2
emissions, by  2010,  after implementation  of the
1990   CAAA,  they  will   account  for  about
60 percent of the U.S.  emissions  of SO2 from all
sources.
   EXHIBIT 24. PERCENT REDUCTION IN TAGGED RE-
     GIONS FROM 1985 TO 2010 AS A FUNCTION
      OF RELATIVE CONTRIBUTION OF EACH RE-
         GION TO ALL TAGGED EMISSIONS

V)
•5
1
00
«-


"^
•£
o

1








1
1 •
1
1
1 •
1
1 ,
1 %
1
1
L • ^ » -
•\ * •
, 	 	 ) 	 .%_-_ — - — • 	

         -100       -SO       0        50       100

            Percent Reduction from 1985 to 2010
                  • 53 Emissions Regions
tween  1985 and 2010, thereby increasing in  im-
portance in analyses  regarding additional  emis-
sions reductions beyond the CAAA. The result  is a
modest leveling out of responsibility for the emis-
sions, as shown in Exhibit 25.
         EXHIBIT 23. PERCENT CONTRIBUTION TO SULFUR EMISSIONS OF 53 TAGGED RADM REGIONS
Year
1985
2010
Year
1985
2010
RADM SO2 Emissions Data
53 Regions: Percent of
U.S. Major Point Sources
83.6%
75.7%
53-Region Total
Tagged Emissions
1 5,420,000 tons
9,265,000 tons
53 Regions: Percent
of Total U.S. Sources
74.8%
63.6%
Total Major U.S.
Point Emissions
1 8,452,000 tons
1 2,245,000 tons
53 Regions: Percent of Total
North American Sources
66.6%
54.3%
Total Emissions from
All U.S. Sources
20,323,000 tons
14,557,000 tons
The  listing of emissions by tagged-source region
shows that emissions per unit area are being lev-
eled out by Title IV. This is shown in Exhibits 24
and 25.  Exhibit 24 quantifies this by comparing the
percent  reduction between 1985  and 2010  as  a
function of the 1985 SO2 emissions contributed by
each of  the 53 regions. The top 8 emitting regions
in 1985  will have the largest percent reductions by
2010, around 50 to 60 percent. A major fraction of
the regions, those with emissions between  100,000
and  450,000  tons/year, have their SO2 emissions
reduced between 10 and  50 percent, forming  a
second  tier of reductions. Emissions from most of
the smallest emitting regions (<100,000 tons/year)
actually increase (negative percent change) be-
Exhibit 26 shows the percentage contributions of
the top 10 emitting regions of 1985 and 2010 to
deposition in the three sensitive  regions. Eight of
the top 10 in 1985 are still in the top 10 in 2010.
But, the top 10  emitters are responsible  for  a
smaller fraction of the deposition in the sensitive
regions in 2010 than they were in 1985. The frac-
tion of the total deposition attributable  to the top
10  emitters goes down by 32 percent, 24 percent,
and 52 percent for the Adirondacks, the mid-Appa-
lachians, and the  Southern  Blue  Ridge,  respec-
tively.  The change is largest for the Southern Blue
Ridge,  resulting in the top 10 emitting regions be-
ing responsible for  only  16  percent of  the  sulfur
deposition in 2010.
                                               70

-------
                                               CHAPTER 3: SOURCE-RECEPTOR RELATIONSHIPS AND DEPOSI-
                                                  TION REDUCTIONS UNDER VARIOUS EMISSIONS SCENARIOS
           EXHIBIT 25. PERCENTAGE OF TAGGED EMISSIONS BY TAGGED REGIONS" FOR 1985 AND 2010
    10.0
  g
 '1
  E
 "D
  0)
  0)
  BJO
  (0
  4_<
  0)
  u
     8.0
     0.0
             3   5  7   9  11  13  15  17  19  21  23  25  27 29  31  33  35  37 39  41  43  45  47  49  51  53
                                          Region in Rank Order
                                               2010
     1985
                         EXHIBIT 26. CONTRIBUTION OFTOP-IO SO2 EMITTING RE-
                           GIONS TO SULFUR DEPOSITION IN SENSITIVE REGIONS
Year
1985
2010
Top-10 SO2 Emitting Regions3
1 5,26,39,1 3,22,20,25,32,7,24
1 5,1 3,22,20,51 ,26,25,39,7,1 7
Adirondacks
30.2%
20.6%
Mid- Appalachi-
ans
49.4%
37.3%
Southern
Blue Ridge
30.8%
16.1%
         a See Exhibit 19 for geographical descriptions of RADM subregions.
3.3.2  Regional Emissions Distribution in
       2010
With greater emissions reductions coming from the
heavier-polluting  regions, the relative importance of
long-range transport is  expected  to decrease in
2010  compared to 1985.  The character of the
source contributions in  2010 is important to any
analyses  of further emissions control  to  reduce
deposition. Two aspects stand out: first, in the mid-
Appalachians and Southern Blue Ridge significant
contributions  to  sulfur  deposition  come  from
sources near the sensitive aquatic regions; second,
the local  versus  long-range character of  the
sources of deposition changes when moving south
from the Adirondacks to the mid-Appalachians and
the Southern Blue Ridge.  The importance of the
top emitting regions decreases as one moves north
or south away from the mid-Appalachians. This re-
sults from a combination of meteorology'(transport
directions), proximity to large emissions  sources,
and the pattern of emissions in 2010.

Exhibit 27 shows that, as one moves from the Adi-
rondacks to the mid-Appalachians  and Southern
Blue Ridge, emissions sources near sensitive areas
are responsible for a greater percentage of deposi-
tion relative to the contributions from the top 10
                                               71

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                     EXHIBIT 27. COMPARISON OF PROXIMATE AND MAJOR EMITTING RE-
                        GIONS TO SULFUR DEPOSITION IN SENSITIVE AREAS IN 2010
Geographic Source of Deposition
A. Sources local to sensitive region
B. Sources local and contiguous to sensitive re-
gions
C. Sources from top-10 deposition contributing
regions (utilities only)
D. Sources from top-1 0 deposition contributing
regions (utilities plus industry)
Adirondacks
3.7%
7.8%
20.0%
28.0%
Mid- Appalachi-
ans
24.0%
28.0%
37.0%
46.0%
Southern
Blue Ridge
12.0%
26.0%
34.0%
41.0%
emissions regions. This can be seen by noting the
contribution to deposition from source regions as
one moves from row A to row D. For the; Adirond-
acks, the 10 regions contributing the most to depo-
sition are responsible  for about 3.5  times  more
deposition than do nearby sources (row D versus
row B).  In the mid-Appalachians and th€' Southern
Blue  Ridge the top emitting  regions contribute
only 60 percent more  deposition than (Jo nearby
regions.  In these two receptor  regions several of
the regions contributing large amounts of deposi-
tion are either local or contiguous to the receptor
regions. The  importance of the contribution from
the top  emitting regions (primarily  located along
the Ohio River Valley) to the  deposition  in the
sensitive regions,  and  hence,  the  importance of
long-range transport, decreases as one moves from
north  to south.  This source attribution  insight  is
used  in Section 3.6 to analyze a  regionally tar-
geted emissions reduction approach to achieving
deposition reductions of sulfur.

3.4  EMISSIONS REDUCTIONS SCENARIOS
This section describes emissions  scenarios created
to evaluate the impact of Title IV on sensitive re-
gions and the environmental impact of additional
emissions reductions beyond those mandated  by
Title IV. This analysis concentrates on emissions of
SO2 and deposition of sulfur because Title IV fo-
cuses on SO2 emissions and because emissions in-
ventories  and  source-receptor  relationships are
better characterized for sulfur than for nitrogen,
the other  key  pollutant  contributing to  acidic
deposition. A scoping analysis of nitrogen  deposi-
tion  is  included. The  scenarios are used in this
chapter to compare deposition  levels in sensitive
regions. Cost and economic impacts of sulfur re-
ductions scenarios are presented in Chapter 5, Im-
plementation.
Two sets of scenarios for SO2 emissions in 2010
(the year Title IV will  be fully implemented) were
developed. The first was created to  evaluate the
environmental impacts (i.e., changes in deposition)
resulting  from the  trading  of  SO2  emissions
allowances. The second represents additional SO2
emissions reductions beyond  those mandated  by
Title  IV.  The  environmental  impacts  of these
scenarios  are compared to the  pre-CAAA  case
(1980).  One scenario  was developed to compare
NOX  reductions with 1990  baseline  emissions
levels. Base years for SO2 and NOX emissions were
selected based on  the availability of data at the
time of this analysis.

Emissions for each scenario were projected from
existing EPA emissions inventories. The National
Allowance Data Base (NADB) was used as the ba-
sis for electric utility SO2 emissions estimates and
projections. EPA developed the NADB to allocate
and track SO2  allowances  issued under Title  IV.
The  NADB was prepared  by  updating the utility
emissions data base (the 1985  National  Unit Ref-
erence File or NURF) included in  the 1985 NAPAP
Emissions Inventory. The basis for non-utility SO2
emissions estimates is the 1985 NAPAP Emissions
Inventory  for Canadian  and U.S. emissions. The
basis for NOX emissions estimates for utilities and
industrial sources is EPA's  1990 Interim Emissions
Inventory.108 As with the NADB,  the  1990 Interim
Inventory was  developed  by updating the 1985
NAPAP Emissions  Inventory. The 1985  inventory
108U.S. Environmental Protection Agency. June 1992.
   Regional Oxidant Modeling—Emissions Inventory
   Development and [mission Control Scenarios.
   U.S. Environmental Protection Agency, May 1989.
   Regional Ozone  Modeling  for  Northeast Trans-
   port—Development of  Base  Year Anthropogenic
   Emissions Inventory.
                                                72

-------
                                              CHAPTER 3: SOURCE-RECEPTOR RELATIONSHIPS AND DEPOSI-
                                                 TION REDUCTIONS UNDER VARIOUS EMISSIONS SCENARIOS
was updated for 1990 using industry growth rates,
EPA's Mobile 4.1 model for mobile sources, and
by adding electric utility units that became opera-
tional between  1985 and 1990. The  1980 SO2
emissions inventory used to calculate 1980 sulfur
deposition was developed for EPA's retrospective
cost/benefit analysis of the CAA conducted pursu-
ant to section 812 of the 1990 Amendments. This
inventory was  constructed  by backcasting emis-
sions from the 1985  NAPAP emissions inventory.

The base year  for  evaluating  the environmental
impacts of the CAAA and the benefits of additional
SO2  emissions  reductions beyond the  CAAA is
2010. Utility emissions for 2010  were  forecasted
in EPA's 1993  Base Case Analysis,109 which was
developed to support rulemaking under Title IV.
Existing  and  planned  electric
utility boilers identified in NADB
Version 3.11 plus generic plants
required  to  meet   growth   in
electricity demand were used as
a basis for the SO2 forecasts.  For
EPA's 1993 Base Case Analysis,
electric  utility  SO2  emissions
were projected from the NADB
inventory using ICF's Coal and
Electric Utilities Model (CEUM).
EPA's   analysis  assumed   full
implementation  of   allowance
trading   (i.e.,   electric  utilities
would engage in allowance trad-
ing in  order  to minimize  the
overall  cost  of  reducing  SO2
emissions by   10  million  tons
below 1980 levels).
           (i.e., by 2-digit SIC  code and state).  Next,  the
           grown emissions were adjusted to reflect the re-
           tirement  of  existing  sources,  new  emissions
           (assumed   to   be   subject   to   New  Source
           Performance Standards [NSPS]) to replace  those
           lost due  to retirement, and  the application  of
           additional  controls required by the CAAA. In total
           non-utility  SO2   emissions   did   not  change
           significantly between  1990 and 2010.

           Projected  nationwide total annual  emissions  of
           SO2 with and without implementation of Title IV
           are shown in Exhibit 28. The projections are  based
           on CEUM predictions of utility emissions with and
           without  Title IV  and  predictions of non-utility
           emissions calculated  as described in the previous
           paragraph.  Annual SO2 emissions decreased by
       EXHIBIT 28. ESTIMATED U.S. SO2 EMISSIONS
    WITH AND WITHOUT TITLE IV FROM 1980 TO 2015
Projections of  2010  non-utility
SO2 emissions from the 1990 In-
terim  Inventory were based on  a
straightforward  approach devel-
oped  by  EPA.110 First, emissions
from the 1990 Interim Inventory
were grown according to the Bu-
reau of Economic Affairs  (BEA)
industrial earnings growth  factor
20
18 •
16
14
  1980   1985    1990
109  ICF  Resources,  Inc.  February  1994.  Economic
   Analysis of the Title IV Requirements of the 1990
   Clean   Air   Act  Amendments.   Prepared  for
   U.S. Environmental Protection Agency, Office of Air
   and Radiation, Acid Rain Division.
110U.S. Environmental Protection Agency, May 1993.
   Regional  Interim  Emissions  Inventories  (1987-
   1991). Volume I:  Development of Methodologies.
1995   2000

    Year
2005   2010   2015
           about 14 percent between 1980 and  1990. With-
           out Title IV annual emissions would slowly begin
           to  increase after 1990 and almost reach 1980 lev-
           els by 2010.  Under Title IV SO2 emissions will de-
           crease dramatically after 1990, achieving a 10 mil-
           lion ton reduction from 1980 levels by 2010.
                                               73

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
Under Title IV utilities are allowed to defer the use
of allowances to future years,  i.e. bank allow-
ances. The Nitrogen Bounding Study, discussed in
Chapter 2, calculates the aquatic  impact of deposi-
tion in the year 2040 based on deposition values
between 2010 and 2040.  EPA has projected SO2
emissions only through 2010 because the assump-
tions  underlying the  CEUM  are assumed  to  be
valid only through that year. Exhibit 29 shows na-
tional utility SO2 emissions predicted by CEUM for
the years 1990-2010. The model  predicts that sev-
eral million allowances will be banked in the early
years  of the allowance trading program (resulting
in early emissions reductions) for use between the
years  2000 and  2010,  with almost  1 million
banked allowances used in 2010. Thus, SO2 emis-
sions  in the year 2010 will be  almost  I million
tons higher than in subsequent years.  A no-bank-
ing (post 2010 full implementation) scenario was
created  for purposes of this analysis to represent
typical emissions for the years 2010 through 2040,
and thus to better reflect  deposition modeled in
the Nitrogen Bounding Study.

The emissions inventories  for the alternative sce-
narios for SO2 in  2010 were developed for the
RADM  domain  (eastern United States, see  Ex-
hibit 17)  as follows and are summarized in  Ex-
hibit 30:

   * 2010  SCENARIO:  This  scenario represents
    utility SO2 emissions in 2010  forecast by
    the CEUM in the 1993 Base Case Analysis
    described above.  Utility emissions in  this
    scenario total 9.5 million tons in 2010. The
    following two scenarios  were calculated
    from this inventory.

   » PosT-2010 FULL IMPLEMENTATION SCENARIO:
    Utility SO2 emissions from  the 2010 sce-
    nario were reduced to allowance levels in
    2010 (the 8.95 million ton  cap mandated
    by Title IV) by removing emissions  banked
    from previous years (see Exhibit 29). This
    scenario was  developed  to represent  full
    implementation of Title IV after 2010. Util-
    ity emissions in years beyond 2010 should
    remain near this level.  Non-utility emis-
    sions were unchanged from the 2010 sce-
    nario.

   » 2010 NO-TRADING SCENARIO: Existing units
    operating in 2010 were  forecast to emit
    SO2 at their allowance levels.  Allowances
    issued to units retiring before  2010 (0.14
    million) were assigned to new units. This
    scenario also  assumes (based on the EPA
    2010 Base Case Forecast Without Title IV)
  that oil and gas-fired units, which were al-
  located allowances, would have emissions
  that are 0.3 million tons less than their al-
  lowance  allocations. Of the total of 0.44
  million tons  of  allowances, 0.19  million
  tons were assumed to be allocated to new
  units, leaving 0.25  million  tons unused.
  Thus total nationwide utility SO2 emissions
  in this scenario  are 8.95 million tons less
  the 0.25  unused allowances or 8.7 million
  tons.   Non-utility   emissions   were  un-
  changed  from the 2010 scenario.

* ADDITIONAL   UTILITY   SO2    REDUCTION
  SCENARIO (UTILITY SO2 EMISSIONS REDUCED BY
  50 PERCENT FROM  THE  POST-2010 FULL  IM-
  PLEMENTATION SCENARIO): Total electric util-
  ity SO2   emissions  were  reduced  by
  50 percent from  the Post 2010 Full  Imple-
  mentation Scenario. To allocate this reduc-
  tion units included in the baseline scenario
  able to  achieve significant  reductions  in
  SO2 emissions (i.e.,  those with  emissions
  rates forecast to  be greater than 0.8 pounds
  of  SO2  per  million  British thermal  unit
  [Ibs/MMBtu])  were  identified.  From this
  group, the  set of boilers able to most cost-
  effectively reduce SO2 were selected based
  on Cadmus'  Generic Retrofit Scrubbing
  Cost Model.  The 50 percent utility  reduc-
  tion was then pro-rated by state. Baseline
  emissions for the set of boilers with SO2
  emissions rates greater than 0.8 Ibs/MMBtu
  were reduced in  proportion to the pro-rated
  state reduction.

» ADDITIONAL  INDUSTRIAL  SO2   REDUCTION
  SCENARIO  (INDUSTRIAL   SO2    EMISSIONS
  REDUCED  BY 50 PERCENT FROM THE POST-2010
  FULL IMPLEMENTATION SCENARIO): SO2  emis-
  sions  in  the  RADM  domain for the 2010
  projection described  above were estimated
  to total about 4 million tons. A 2 million
  ton reduction was achieved from industrial
  boilers   and   major   process   industries
  (categories projected to  emit  more than
  10,000 tons in 2010). As described in more
  detail  in  Chapter  5,  cost-effectiveness
  measures (annual cost per ton of SO2  re-
  moved) were calculated for each sector  to
  allocate the 2 million ton reduction. Based
  on this  analysis,  industrial   boilers ac-
  counted  for  63 percent  and  industrial
  sources  37 percent of the  emissions  re-
  moved.

» ADDITIONAL UTILITY  AND  INDUSTRIAL  SO2
  REDUCTION  SCENARIO  (UTILITY  AND  IN-
  DUSTRIAL    SO2  EMISSIONS  REDUCED   BY
  50 PERCENT FROM THE POST-201 0 FULL IMPLE-
                                               74

-------
                                         CHAPTER 3:  SOURCE-RECEPTOR RELATIONSHIPS AND DEPOSI-

                                            TION REDUCTIONS UNDER VARIOUS EMISSIONS SCENARIOS
o
o

e
s
l/>


O
u
5

I
r
x
                                                                                     O
                                                                                     CM
                                                                                     in
                                                                                     o
                                                                                     o
                                                                                     CM
                                                                                     o
                                                                                     CM
                                                                                     IO
                                                                                     O)
                                                                                     O)
                                                                                     o
                                                                                     O)
                                                                                     0)
       CM
CM
CM
O    00
CM    -r-

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
               EXHIBIT 30. SO2 EMISSIONS IN THE U.S.
             RADM DOMAIN (EASTERN UNITED STATES)
Scenario
1980
1985 NAPAP
2010
Post-2010 full implementation scenario
No Trading
Additional utility SO2 reduction
Additional utility and industrial SO2 reduction
SO2 Emissions
(million tons)
24.8
20.3
14.6
14.0
13.7
9.7
7.8
     MENTATION SCENARIO): Combination  of  the
     two previous scenarios. As shown in Sec-
     tion 3.2.1, utility and industrial sources ac-
     count for  87 percent of total nationwide
     SO2 emissions. Thus, a 50 percent  reduc-
     tion in utility and industrial SO2 emissions
     represents a 43.5 percent reduction in total
     SO2 emissions.

  *  ADDITIONAL UTILITY  AND  INDUSTRIAL NOX
     REDUCTION  SCENARIO  (UTILITY  AND   IN-
     DUSTRIAL  NOX  EMISSIONS  REDUCED  BY  50
     PERCENT FROM THE 1990  INTERIM EMISSIONS
     INVENTORY): 1990 emissions from each unit
     were reduced by half. As shown in Section
     3.2.1, utility and  industrial sources account
     for 47.5 percent  of total nationwide NOX
     emissions.  Thus,  a 50 percent reduction in
     utility and industrial NOX emissions repre-
     sents only a  24 percent reduction  in total
     NOX emissions.

3.5  DEPOSITION REDUCTIONS UNDER: VARIOUS
     NATIONAL EMISSIONS REDUCTIONS
     SCENARIOS
Total sulfur or nitrogen deposition to each 80 by
80 km grid cell in the RADM region was calcu-
lated for each scenario in Exhibit 30. Sulfur  depo-
sition in the units of kilograms of sulfur per hectare
per year (kg-S/ha/yr) is the sum of wet and dry sul-
fate;  nitrogen deposition (in units of kg-N/ha/yr)
consists of wet and dry nitrate and wet and dry
ammonia. This section is divided into three parts:
sulfur deposition reductions under Title IV and an
analysis  of trading;  additional  sulfur deposition
emissions beyond Title IV; and nitrogen deposition
reductions.
               3.5.1   Impact of SO2 Allow-
                      ance Trading on Sulfur
                      Deposition
               Exhibit 31  is a map of the spatial
               distribution of total sulfur deposi-
               tion in the RADM region for the
               pre-CAAA  year  of  1980.  The
               highest  sulfur deposition  levels,
               more than 20 kg-S/ha/yr, were in
               the  industrial area encompassing
               parts of West Virginia,  Pennsyl-
               vania, Ohio, Kentucky, and Indi-
               ana and in the mid-Appalachians.
               Deposition  levels  were  about
20 percent lower in  the Southern Blue Ridge and
Pocono Mountains and about 40 percent lower in
the Adirondacks. Thus, the highest deposition lev-
els are in or just downwind of the highest emitting
areas.

     EXHIBIT 31. ANNUAL AVERAGE RADM TO-
     TAL SULFUR DEPOSITION (kG-S/HA): 1980
Exhibit 32 is the same map projected for the post
2010 full implementation scenario. The spatial dis-
tribution of emissions in 2010 is similar to that of
1980. In 2010  deposition  levels in  the Southern
Blue Ridge are  projected to be about 15 percent
lower  and  those  in  the  Adirondacks  about
40 percent lower than  in  the industrial mid-West
and mid-Appalachians.  Similar spatial distributions
are also found for the 2010 and 2010 no-trading
scenarios. Exhibit 33 is a map of percentage reduc-
tions in sulfur deposition from  1980 to the post
2010 full implementation scenario. This map again
demonstrates that the largest deposition reduc-
                                               76

-------
                                            CHAPTER 3: SOURCE-RECEPTOR RELATIONSHIPS AND DEPOSI-
                                               TION REDUCTIONS UNDER VARIOUS EMISSIONS SCENARIOS
EXHIBIT 32. ANNUAL AVERAGE RADM-PREDICTED
  TOTAL SULFUR DEPOSITION (KG-S/HA): POST-
      2010 FULL CAAA IMPLEMENTATION
EXHIBIT 33. PERCENTAGE REDUCTIONS IN SULFUR
   DEPOSITION FROM CAAA IMPLEMENTATION
  EXHIBIT 34. IMPACT OF TRADING ON SULFUR DEPOSITION IN SENSITIVE REGIONS
Emissions Scenario
1980
1985 NAPAP
2010 (with trading)
Post-2010 full implementation
2010 (without trading)
Annual Average Deposition Level (kg-S/ha)
Adirondacks
11.0
9.8
7.1
6.9
6.8
Mid- Appalachi-
ans
19.0
17.0
12.0
11.0
11.0
Southern
Blue Ridge
14.0
13.0
9.9
9.7
9.2
tions, over 40 percent, are in the highest emitting
regions.

Exhibit 34 compares annual average sulfur deposi-
tion for the 2010  (with trading), no-trading,  and
post 2010 full  implementation scenarios for the
three sensitive regions analyzed for this report. The
1980 and 1985 deposition values are included for
reference. The most consistent comparison  to as-
sess the  impact of allowance  trading is between
the post 2010 full  implementation and no-trading
scenarios. For the  post 2010 full implementation
scenario  sulfur deposition in the Adirondacks and
mid-Appalachians  is  reduced about 40 percent
from 1980 and reduced about 30 percent in the
Southern Blue Ridge.  The deposition values  and
percent reductions  are essentially the same for the
no-trading case in  the Adirondacks  and the mid-
Appalachians. The lower total RADM-wide SO2
emissions for the  no-trading case  shown in Ex-
hibit 30 most likely accounts for the 0.1 kg-S/ha
difference in deposition in the Adirondacks.  In the
Southern Blue Ridge the deposition (9.2 kg-S/ha) is
lower and the  percent reduction (34 percent) is
higher for the no-trading case.

A more detailed spatial view of the  differences in
sulfur deposition between  the  post 2010 full  im-
plementation and no-trading scenarios for 2010 is
given in the map in Exhibit 35. The areas shaded
diagonally and in solid black depict areas where
projected  regional  deposition would be higher
with trading. The areas shaded with a plaid pattern
are those where projected deposition  would be
lower with trading. The largest increases associ-
ated with trading,  somewhat more  than 1.2  kg-
S/ha or about 10 percent of total  deposition, are
near the Southern Blue Ridge and in  northern Ala-
bama. Most of the  increases are 0.2-0.8  kg-S/ha,
less than  10 percent of the total sulfur deposition.
Note that the allowance trading program cannot
result  in  exceedance of  a  National  Ambient
                        Air Quality    Standard
                        (NAAQS), and thereby
                        will not create or  inten-
                        sify local  air quality or
                        public health problems,
                        because  sources  must
                        always comply with the
                        requirements   of   the
                        NAAQS as well as the
                                            77

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
    EXHIBIT 35. ANNUAL AVERAGE RADM TOTAL SULFUR DEPOSITION (KG-S/HA) BY GEOGRAPHIC AREA IN 2010:
 RANGES OF DIFFERENCES IN DEPOSITION BETWEEN POST 2010 FULL IMPLEMENTATION AND NO TRADING SCENARIOS
                                                                   -0.80 TO  -1.16

                                                                   -0.50 TO  -0.80
                                                                EH -0.20 TO  -0.50

                                                                D -0.20 TO  0.20

                                                                0 0.20 TO 0.50

                                                                il 0.50 TO 0.90
                                                                • 0.90 TO 1.53
Prevention of Significant Deterioration Program
(e.g.,  in  National  Parks and  Wilderness  Areas).
There are a few small regions where deposition is
projected to decrease due to trading. These are in
southern  New England,  southern  Florida,  and
northwestern Tennessee. Exhibit 35 shows that the
lack of a difference in deposition in the mid-Appa-
lachians  between  post 2010 full implementation
and no-trading scenarios listed in Exhibit 34 is due
to increases in one area being offset by decreases
in another. Nonetheless, the modeling  estimates of
the differences in  sulfur deposition between the
post 2010 full implementation and no-trading sce-
narios indicate that they  are expected to be less
than 10 percent,  and in many cases much less in
the sensitive aquatic regions. It  is also  apparent
from Exhibit 35 that trading is projected to have
virtually no impact on sulfur deposition in Canada.

3.5.2  Effect of Additional SO2 Emissions Re-
       ductions on Sulfur Deposition
This section  presents deposition values for  the
additional  SO2   reduction  scenarios  described
above and compares each to the post 2010  full
implementation  scenario.  The additional reduc-
tions scenarios were chosen to  demonstrate  the
                                               78

-------
                                               CHAPTER 3: SOURCE-RECEPTOR RELATIONSHIPS AND DEPOSI-
                                                  TION REDUCTIONS UNDER VARIOUS EMISSIONS SCENARIOS
magnitude of further emissions reductions and the
spatial distributions associated with each. Exhib-
its 36  and  37  are  sulfur  deposition  maps for
50 percent additional utility SO2 reductions, and
50 percent utility  plus  50 percent  industrial SO2
reduction scenarios.  Exhibits 38 and 39 are maps
of percentage reductions in sulfur emissions rela-
tive to the post 2010 full implementation  case for
each  scenario. Exhibit 40 summarizes the annual
average deposition to  the three sensitive regions
for each scenario,  and  Exhibit 41  lists the percent-
age decrease  in sulfur deposition to each region
relative to 1980.

By comparing the sulfur deposition maps for the
additional  reductions  scenarios (Exhibits 36 and
37) to the deposition map for the  post 2010 full
implementation  case (Exhibit 32),  it is  apparent
that the general spatial distribution of sulfur depo-
sition remains similar with the highest deposition
values falling in the industrial areas of the  mid-
West, the regions  with the highest  emissions. Ex-
hibits 38 and 39 show that the percentage reduc-
tions increase as one moves  south, with the largest
percentage reductions  occurring from the  lower
Ohio Valley  through  Florida, with  significant re-
ductions  in central New York State. The reduction
in industrial emissions has the greatest impact in
the Southern  Blue Ridge and  South Atlantic re-
gions.

3.5.3  Decrease  in Total Nitrogen Deposi-
       tion from  Decreases in NOX Emissions
Nitrogen deposition  is primarily the sum of wet
and dry deposition of nitrate and ammonia. (There
is also a  small  contribution from  dry  NO2.) As
noted  in  Section  3.2.1, emissions  estimates for
ammonia  are highly uncertain  both in terms  of
magnitude and source. Therefore, RADM replaces
predicted ammonia with observed ammonia depo-
sition.  In deposition  modeling the contribution of
ammonia to nitrogen deposition is treated as back-
ground along with nitrogen  deposition  from oxi-
dized species from natural sources and agricultural
emissions.  Background from  these sources is esti-
mated to be 4, 4.3, and 2.8 kg-N/ha in the Adiron-
dacks,  mid-Appalachians,  and  Southern   Blue
Ridge, respectively.111 Therefore, the scenario ana-
  1 Van Sickle, ]., M.R. Church. 1995. Methods for Es-
   timating the Relative Effects of Sulfur and Nitrogen
   Deposition on Surface  Water Chemistry. Environ-
   mental Research Laboratory, Corvallis, OR.
lyzed in this section  focuses on  anthropogenic
NOX emissions reductions.

Exhibit 42 shows historical and projected nation-
wide  total NOX  emissions for the period 1980 to
2010  with and without implementation of Title IV.
The emissions projections include Title I (i.e.,  rea-
sonably achievable control technologies) and  mo-
tor vehicle NOX reductions mandated by Title II.
The difference in the two projections is the 2  mil-
lion ton  per  year  reduction  in utility NOX emis-
sions  estimated to come from Title IV. Unlike  util-
ity SO2  emissions, NOX emissions are not capped
by the CAAA.

As shown in Section  3.2.1, utilities and highway
vehicles  are  each  responsible for about one-third
of nationwide emissions of  NOX  and industrial
sources  for  about one-sixth  of nationwide  NOX
emissions. Exhibits 43-^5 show  nitrogen  deposi-
tion maps of the RADM region for each of these
source categories  in  1990.  Deposition  resulting
from  utility emissions is strongly concentrated in
the Ohio Valley and falls away almost uniformly
with distance. Industrial  deposition is concentrated
on the Gulf Coast, reflecting the  concentration of
high NOX emitting industry in that area. Deposi-
tion from mobile sources is concentrated along the
East Coast as a result of automobile emissions in
urban areas.  The principal  contributors to deposi-
tion in the Southern Blue Ridge and mid-Appala-
chians  are   utility   and  industrial  emissions.
The Adirondacks  are  affected  about equally by
utility and mobile source emissions.

It is estimated that  under Title II of the Act, mobile
source emissions in 2010 in the eastern half of the
United States will be reduced by about 15 percent
from  1990 levels. Since mobile sources represent
about one-third of total NOX emissions, this reduc-
tion corresponds to a reduction  in total NOX of
about 5 percent. Under  Title IV,  NOX emissions
rates from utility boilers will be reduced based on
the degree of reduction  available through the ap-
plication of control technology. EPA estimates  that
implementation of these regulations will result in
utility NOX emissions in  2010 which  are approxi-
mately 20 percent less  than  would have existed
without  Title IV.  Implementation  of  Title I  to
achieve the ozone standard in the  Northeast  and
other  nonattainment areas  is expected to involve
significant reductions in  NOX emissions from  util-
ity, industrial, and  mobile sources. These  reduc-
tions are likely to be region specific.
                                                79

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
  EXHIBIT 36. RADM-PREDICTED ANNUAL AVERAGE
      TOTAL SULFUR DEPOSITION (KG-S/HA) IN
    2010 UNDER ADDITIONAL UTILITY SO2 EMIS-
           SIONS REDUCTION SCENARIO
EXHIBIT 37. RADM-PREDICTED ANNUAL AVERAGE
   TOTAL SULFUR DEPOSITION (KG-S/HA) IN
 2010 UNDER ADDITIONAL UTILITY AND INDUS-
  TRIAL SO2 EMISSIONS REDUCTION SCENARIO
   EXHIBIT 38. PERCENTAGE REDUCTIONS IN SULFUR
DEPOSITION FROM POST-2010 FULL IMPLEMENTATION -
UNDER ADDITIONAL UTILITY SO2 REDUCTION SCENARIO
 EXHIBIT 39. PERCENTAGE REDUCTIONS IN SUL-
 FUR DEPOSITION FROM POST-201 0 FULL IMPLE-
   MENTATION UNDER ADDITIONAL UTILITY
 AND INDUSTRIAL SO2 REDUCTION SCENARIO
                                             80

-------
                                   CHAPTER 3:  SOURCE-RECEPTOR RELATIONSHIPS AND DEPOSI-
                                      TION REDUCTIONS UNDER VARIOUS EMISSIONS SCENARIOS
EXHIBIT 40. SULFUR DEPOSITION TO SENSITIVE REGIONS UNDER VARIOUS EMISSIONS SCENARIOS
Emissions Scenario
1980
1985 NAP AP
Post-2010 full implementation
CAAA implementation plus additional
utility SC>2 reduction
CAAA implementation plus additional
utility and industrial SO2 reduction
Annual Average Deposition Level (kg-S/ha)
Adirondacks
11.0
9.8
6.9
5.5
4.7
Mid-Appalachi-
ans
19.0
17.0
11.0
8.1
6.9
Southern
Blue Ridge
14.0
13.0
9.7
6.8
5.5
<
        EXHIBIT 41. PERCENT REDUCTIONS IN SULFUR DEPOSITION TO SENSITIVE RE-
            GIONS FROM 1980 LEVELS UNDER VARIOUS EMISSIONS SCENARIOS
Emissions Scenario
Post-2010 full implementation
CAAA implementation plus additional
utility SC>2 reduction
CAAA implementation plus additional
utility and industrial SO2 reduction
Percent Reduction
Adirondacks
39
51
58
Mid- Appalachi-
ans
41
56
63
Southern
Blue Ridge
31
52
60
                  EXHIBIT 42. ESTIMATED U.S. NOX EMISSIONS
               WITH AND WITHOUT TITLE IV FROM 1980 TO 201 0
              26 T
              24 -•
          a   22 -.
              20 ••
          o
              18 •
              16 ••
              14
                 1980
2005   2010
                                    81

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
   EXHIBIT 43. PERCENT CONTRIBUTION OF UTILITY
     SOURCES TO NITROGEN DEPOSITION IN 1990
   EXHIBIT 45. PERCENT CONTRIBUTION OF MOBILE
     SOURCES TO NITROGEN DEPOSITION IN 1990
 EXHIBIT 44. PERCENT CONTRIBUTION OF INDUSTRIAL
     SOURCES TO NITROGEN DEPOSITION IN 1990
A detailed inventory of NOX emissions in the year
2010 suitable for RADM modeling which consid-
ered full implementation (all Titles) CAAA was not
available for this report. As a scoping exercise for
this report two emissions scenarios were used: the
baseline 1990 Interim Inventory and an  inventory
in which NOX emissions from industrial and utility
sources is reduced by 50 percent from 1990 levels.
This reduction inventory, as described  in Section
3.4, corresponds to a 24 percent decrease in total
NOX emissions. Nitrogen deposition from ammo-
nia was assumed to  remain constant. Exhibits 46
and 47 are maps of nitrogen deposition from the
1990 Interim Inventory and from the utility and in-
dustrial emissions reduction scenario.  Exhibit 48
is  a map of percentage reductions  in nitrogen
deposition due to the control scenario.  Exhibit 49
summarizes  nitrogen deposition and percentage
reductions for the baseline  and  control scenarios.
Percentage reductions are less than for the  similar
SO2 scenario for two reasons: ammonia  emissions,
which  account for between  one-fifth  and  one-
quarter  of nitrogen  deposition  in  the eastern
United States, are assumed to be constant and util-
ity and industrial  sources account for only about
50 percent of total NOX emissions, whereas these
categories account for 87 percent of SO2  emis-
sions. The greatest percentage reduction is  found
in the mid-Appalachians, followed by  the Southern
Blue Ridge, and then the Adirondacks,  reflecting
the relative  importance of utility  and  industrial
NOX emissions to deposition in these regions.

3.6  EMISSIONS REDUCTIONS STRATEGIES TO
     ACHIEVE GEOGRAPHICALLY TARGETED
     SULFUR DEPOSITION LOADS
In the previous section deposition values were pro-
jected for nationwide emissions  control scenarios.
That is, similar to Title  IV, the reductions resulted
from  control  approaches which  specified  emis-
sions  reductions by source  category  not by geo-
graphic location. This section describes and ana-
lyzes geographically targeted emissions reductions
to  achieving  deposition reductions.  Analysis of
                                               82

-------
                                          CHAPTER 3:  SOURCE-RECEPTOR RELATIONSHIPS AND DEPOSI-
                                             TION REDUCTIONS UNDER VARIOUS EMISSIONS SCENARIOS
EXHIBIT 46. ANNUAL AVERAGE RADM TOTAL NI-
   TROGEN DEPOSITION (KG-N/HA) IN 1990
EXHIBIT 47. RADM-PREDICTED ANNUAL AVER-
     AGE TOTAL NITROGEN DEPOSITION
 (KG-N/HA) UNDER UTILITY AND INDUSTRIAL
   NOy EMISSIONS REDUCTIONS SCENARIO
                       EXHIBIT 48. PERCENTAGE REDUCTIONS IN NITROGEN
                          DEPOSITION UNDER UTILITY AND INDUSTRIAL
                            NOy EMISSIONS REDUCTIONS SCENARIO
          EXHIBIT 49. NITROGEN DEPOSITION TO SENSITIVE REGIONS UNDER BASE CASE AND ADDI-
                TIONAL UTILITY AND INDUSTRIAL NOX EMISSIONS REDUCTION SCENARIO
Emissions Scenario
1990 Base Case
Additional utility and industrial NOX reduc-
tions from 1 990 base case
(% reduction from base case in parenthesis)
Annual Average Deposition Level (kg-N/ha)
Adirondacks
9.5
8.1
(14%)
Mid- Appalachi-
ans
14.3
11.3
(21%)
Southern
Blue Ridge
11.9
9.9
(16%)
                                           83

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
these reductions relies on RADM source-receptor
relationships to target source regions for emissions
reductions  needed  to achieve a  specified  target
load.

The  assessment of geographically targeted reduc-
tions strategies relies  on the ability of ihe RADM
Tagged Engineering Model to identify the contribu-
tion  of emissions from tagged source regions to
deposition  in selected target regions. This assess-
ment is based on results for the 53 tagged regions
shown in Exhibit 19. To illustrate the potential ap-
plicability of this geographic targeting, deposition
goals for the three  sensitive aquatic regions were
selected  to be  equal  to  the  deposition  levels
achieved by the source-category specific  SO2 re-
duction  scenarios analyzed in  Section 3.5. The
maintenance load  scenario  is  described in  Ex-
hibit 50.

As noted in Section  3.3, implementing Title IV will
result in a higher proportion of sulfur deposition in
sensitive aquatic  regions  coming  from  nearby
sources in 2010 than was the case in 1985. At the
same time, the top  10 emitting  regions' contribu-
tions to  deposition in those  areas will decrease in
importance from 1985 to 2010.  To assess  the fea-
sibility of defining  geographically targeted  emis-
sions reductions for given deposition targets, two
potential scenarios  were selected. The first  limits
the selection to obtaining the required emissions
reductions  from sources near a  sensitive region.
The other focuses on securing emissions reductions
from those RADM subregions most responsible for
deposition in a sensitive region.

Exhibit 40 lists the deposition level for each  sensi-
tive   aquatic   region   achieved  by  a  further
50 percent reduction  in utility SO2 emissions be-
yond the CAAA from utility sources in the United
States and the deposition level achieved by  a fur-
ther  50 percent reduction in utility and industrial
sources. These deposition levels were selected to
illustrate target loads  in order to compare nation-
wide and target emissions  reductions.  Exhibit 27
shows that  in 2010  after  implementation of the
CAAA, local and contiguous sources will  contrib-
ute 7.8 percent of deposition in the Adirondacks,
28 percent   in   the  mid-Appalachians,   and
26 percent in the Southern Blue Ridge. Thus, up to
a point it should be feasible to develop geographi-
cally restricted targeted areas for emissions reduc-
tions in the mid-Appalachians and the Southern
Blue Ridge as long as the reductions demanded
are significant, such as 95 percent or greater.  Be-
cause of the relatively smaller contribution of local
sources to deposition in  the  Adirondacks, reduc-
tions in emissions from  a  large geographic  area
would be required to achieve targeted levels of
deposition.

To achieve the same deposition  levels as  the sce-
nario defined by a 50 percent reduction  in utility
SO2 emissions beyond that achieved by the CAAA,
targeted source regions were identified by sequen-
tially removing 95 percent of the SO2 utility emis-
sions (remaining after implementation of Title IV)
from each  subregion until the  deposition levels
listed  in  Exhibit 40 were achieved.  For the  con-
tiguous scenario, emissions were removed starting
at the center of the sensitive region and continued
outward until the deposition goal was reached. For
the noncontiguous scenario, emissions were re-
moved by subregion  in  order of contribution to
deposition to the sensitive region.

Exhibits 51 a and 51 b  show the results of  targeted
emissions reductions to achieve  the same deposi-
tion as the 50 percent additional SO2 utility case.
The exhibits show that geographically targeted re-
ductions  can  be  achieved  for all of the  sensitive
aquatic areas. As anticipated,  it  takes  a larger
number of tagged-regions  and greater emissions
reductions to achieve the target for the Adirond-
acks than for the other two regions.  By comparing
the emissions  reductions required in the contigu-
ous vs. non-contiguous  scenarios,  it is  apparent
that only in the  Southern Blue Ridge are contigu-
ous reductions more efficient, requiring 25 percent
fewer emissions  reductions. This difference for the
Southern Blue Ridge can  be attributed to the inclu-
sion of RADM Subregion 51  (Northern Florida), a
major emitting region, in the non-contiguous re-
ductions. For the other two sensitive regions both
contiguous and non contiguous reductions rely es-
sentially  on the same geographic  areas of  the
country.  It is also interesting to note that the total
emissions reductions  to  achieve all  three target
loads  simultaneously are essentially equal  and are
less (i.e., amount of total emissions  reductions)
than 10 percent  more efficient than a  nationwide
50 percent reduction in utility SO2 emissions.
                                                84

-------
                                             CHAPTER 3:  SOURCE-RECEPTOR RELATIONSHIPS AND DEPOSI-
                                                TION REDUCTIONS UNDER VARIOUS EMISSIONS SCENARIOS
                       EXHIBIT 50. EXAMPLE: SELECTION OF MAINTENANCE LOADS
Critical load definitions are founded on scientific determinations of quantitative pollutant loadings be-
low which no significant harmful effects occur to critical ecological processes. Critical load definitions
depend solely on inherent ecological properties.  Target load definitions differ in that they incorporate
social, policy, economic, and related considerations along with  the scientific observations.  An exam-
ple of a possible target load would be a  decision to regulate  acidic deposition at levels adequate to
maintain proportions of ANC<50 ueq/l  waters,  roughly equivalent to pH>6.5, at or below the propor-
tions  found during the 1984-85 National  Surface Water Surveys (NSWS) for each of three  regions in
the eastern United States discussed here.  Other possible goals  for defining target loads could include,
for example, restricting deposition to produce a reduction in the  1984-85 proportions of ANC<50
ueq/l  waters by a specific percentage or to some fixed proportion endpoint. Also,  the definition could
be expanded to include a larger geographical region.

For purposes of illustration, assume that social,  policy, economic, scientific considerations led to a se-
lection of the first of the above options as a generally acceptable goal for a target load: to limit sulfur
deposition rates  sufficiently to maintain proportions of target population surface waters having ANC<
50 ueq/l in these regions through the year 2040 at proportions no  greater  than  those found  for the
same  subpopulation  by the 1984-85 NSWS. This maintenance  load also would allow for a possible
restoration of prior ANC levels  and reduction  in acidic conditions in some  of  these waters. The  nu-
meric proportions of target population waters  with ANC<50  ueq/l  found in 1984-85 that  would be
used  to evaluate attainment of this goal in each of the three regions are <55 percent in the Adirond-
acks,  <27 percent in the Mid Appalachians, and <6 percent in  the Southern Blue Ridge. It is important
to note, however, that the percentages reported here represent sensitive subpopulations of the  overall
number of surface waters in those regions.

The Nitrogen Bounding Study (NBS) projections, described in  detail in Chapter 2, can be used to de-
termine sulfur deposition values necessary to attain this goal by locating these percentages on the NBS
plots  shown in Appendix B. For each  region,  projections of  sulfur deposition  loadings necessary to
maintain the 1984-85 NSWS proportions of ANC<50 ueq/l waters depends on the assumed time to ni-
trogen saturation for  that region. For the  purposes of this  analysis, it has been assumed that nitrogen
deposition rates will  remain unchanged for the 1984-2040 period. In considering this approach, it is
critical to recognize that, while the NBS modeling results are the best available information, there are
significant uncertainties in the projected  relationships (see Section 2.5.3). Consequently, the  magni-
tude of the uncertainties in sulfur deposition maintenance loads used in this analysis cannot be quanti-
fied at this time.

Current scientific uncertainty does not allow quantifying the time to nitrogen saturation for any of the
three  sensitive regions  nor for any other  regions. However, it is reasonable to suggest that times to
saturation do vary among regions due to differences in temperature, moisture, soil fertility, forest age,
history of nitrogen deposition, and other  variables. Watersheds  in the Northeast  have  cooler  annual
temperatures,  snorter growing seasons, and long histories of elevated  nitrogen and sulfur deposition
levels. Consequently, watersheds in the Adirondacks may  include those having the shortest remaining
times  to nitrogen saturation. Watersheds in  the Mid-Appalachians and Southern Blue Ridge Province
may have longer remaining times to nitrogen saturation.

For illustrative purposes in this analysis, if the time to nitrogen  saturation is between 75  and  150 years
in the Mid-Appalachians, and between 200 and 300 years in the SBRP, then a sulfur deposition  load of
about 5 kg/ha/yr is projected by the NBS  results as potentially  maintaining surface waters with ANC<
50 ueq/l at 1984-85 proportions. NBS modeling suggests that for a time to nitrogen saturation  for the
Adirondacks of between 25 and 75 years, a  greater than 50 percent reduction in both sulfur and nitro-
gen may be necessary to maintain the 1984 proportion of  ANC<50 ueq/l lakes;  however, a 5 kg/ha/yr
sulfur load would likely provide some benefits.

This example of  a loadings  approach (i.e., maintenance load)  concentrates only on sulfur deposition,
while nitrogen deposition is held constant. However, NBS results for all three regions indicate that re-
ducing nitrogen deposition rates is projected to  provide likely benefits in reducing proportions of ANC
<0 ueq/l and ANC<50 ueq/l surface waters that may equal or exceed the potential benefits obtainable
from reducing sulfur deposition  alone. The amount of benefit would  depend on the actual amount of
reduction  in sulfur and nitrogen deposition obtained, and on  the actual time to  watershed nitrogen
saturation within each region.  Any efforts to develop acid deposition standards would likely include
both sulfur and nitrogen.
                                              85

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                        EXHIBIT 51 A. GEOGRAPHICALLY TARGETED ADDITIONAL UTIL-
                         ITY SO2 REDUCTION IN CONTIGUOUS RADM SUBREGIONS


Deposition
Subregions

SO2 Emissions
Reduction (tons)
c
Adirondacks
5.5 kg-S/ha
45,44,5,4,2,3,9,
11,12,14,7,8,10,
13,15,18,19,20,22
3,018,000
ensitive Region
Mid-Appalachi-
ans
8.1 kg-S/ha
13,19,28,20,
21,27,15,22

1,952,000

Southern
Blue Ridge
6.8 kg-S/ha
29,30,31,39,38,
37,28,21,23,27,
36,32,33,40
1,508,000

All Three Re-
ceptor Re-
gions
—
—

4,526,000a

Nationwide
Utility
—
—

5,047,000
   a This emissions total was derived without double counting those subregions contributing to deposition in
     more than one receptor region.
               EXHIBIT 51 B. GEOGRAPHICALLY TARGETED ADDITIONAL UTILITY SO2 REDUCTION
              IN MAJOR RADM SUBREGIONS CONTRIBUTING TO DEPOSITION (NOT CONTIGUOUS)


Deposition
Subregions

SO2 Emissions
Reduction (tons)

Adirondacks
5.5 kg-S/ha
15,13,22,20,
14,7,10,12,25,
3,9,5,2,17,44
3,160,000
ensitive Region
Mid-Appalachi-
ans
8.1 kg-S/ha
13,20,15,
22,28,27,10
2,080,000

Southern
Blue Ridge
6.8 kg-S/ha
31,29,39,38,28,
37,40,51,20,22
2,081,000

All Three Re-
ceptor Re-
gions
—
_

4,658,000a

Nationwide
Utility
—


5,047,000
   a This emissions total was derived without double counting those subregions contributing to deposition in
     more than one receptor region.
Exhibit 52 is a map of the geographically targeted
reductions  for the  contiguous reductions for all
three receptor regions. RADM subregions required
for the Adirondacks (indicated by short dashes) ex-
tend from the Canadian border south  through Vir-
ginia and Kentucky and from the Eastern seaboard
west to the middle of Indiana. Source regions for
the mid-Appalachians (solid line)  are fairly sym-
metric around the region and extent from the mid-
dle of  Ohio south through central  North Carolina
and  west from the North Carolina  Coast to  the
middle of  Indiana. For the Southern Blue  Ridge
source regions (longer dashed lines)  include most
of North and South Carolina,  Kentucky,  Tennes-
see,  and Alabama and the northern half of  Geor-
gia.  The map shows the striking overlap between
tagged regions required for achievement of the tar-
get load in the mid-Appalachians and the other
two  source  regions  and  the lack  of overlap
beween the Adirondack and Southern Blue Ridge
source regions. Achieving the target loads in  the
Adirondacks and Southern Blue Ridge would also
achieve the target load in the mid-Appalachians.

A similar analysis was conducted to determine if
the maintenance loads (see Exhibit 50) developed
for illustrative purposes and loads  defined by  the
deposition  levels  achieved  by the  nationwide
50 percent utility plus industrial SO2 emissions re-
duction could be achieved by targeting only utility
emissions  in the 53 tagged subregions. It was only
possible to achieve the deposition targets in  the
mid-Appalachians. It was possible to  achieve  the
maintenance load in the Adirondacks, but not in
the mid-Appalachians or the Southern Blue Ridge,
when  only  considering 95 percent reductions of
                                                86

-------
                                               CHAPTER 3:  SOURCE-RECEPTOR RELATIONSHIPS AND DEPOSI-
                                                  TION REDUCTIONS UNDER VARIOUS EMISSIONS SCENARIOS
     EXHIBIT 52. MAP OF EXTENT OF CONTIGUOUS
   GEOGRAPHIC REGIONS FOR ACHIEVING TARGETED
   DEPOSITION LOADS EQUIVALENT TO ADDITIONAL
       NATIONWIDE UTILITY SO2 REDUCTIONS
utility  emissions  in  the targeted  regions. Thus,
targeting scenarios that included industrial emis-
sions reductions were then analyzed.

For the 50 percent utility and industry SO2 reduc-
tions beyond the post 2010 full implementation
scenario, targeted source subregions were identi-
fied by sequentially removing  95 percent of the
SO2  major  source  (utility  plus  large  industrial
sources)  emissions from  each subregion.  Results
are shown in Exhibits 53a and 53b. It appears fea-
sible to  have geographically targeted  controls if
reductions  are from  both  utility  and  industrial
sources. For these targets, there is little geographic
"efficiency"   between  constraining  the  tagged
subregions to be contiguous  or focusing on the re-
gions  most responsible for deposition.  Thirty-one
emissions subregions are included in the former
case for all three sensitive receptor regions and 29
in the latter. As shown in Exhibit 54 the tagged
subregions cover the  same geographical expanse
as those involved in the utility  reduction targeted
scenario (compare Exhibit 52). The principal differ-
ence is that the overlap among regions is more ex-
tensive with reductions in some source regions re-
quired to achieve target loads in all three sensitive
regions. Again, emissions reductions from sources
in the western part  of the RADM domain are not
required.

For the  maintenance  load  analysis, a 95 percent
reduction  in major source emissions was used. The
results for the case  focusing on emissions  reduc-
tions from regions most responsible for deposition
are shown in Exhibit 55. Compared to the number
of tagged  emissions regions shown in Exhibit 53b,
the number  is somewhat less for the Adirondacks,
somewhat more for the Southern Blue Ridge, and
almost 3  times  larger  for the mid-Appalachians.
Nevertheless, the total number of tagged regions is
comparable,  31  for  the 50 percent utility plus in-
dustrial  source scenario and  33 for the mainte-
nance load scenario). Two-thirds of the subregions
contributing to the Adirondacks and two-thirds of
the subregions contributing to the Southern  Blue
Ridge also are major sources  of deposition  in the
mid-Appalachians. (A similar analysis using  only
contiguous  subregions  shows  a  greater   than
90 percent overlap  between  the Southern  Blue
Ridge and mid-Appalachians). There is no overlap
of the two sets of subregions identified for the Adi-
rondacks and the Southern Blue Ridge. Thus, these
regions could be individually  targeted;  however,
attempting to achieve  the  maintenance  load for
the mid-Appalachians  may be best done as part of
a strategy to achieve  it for all  three sensitive re-
gions at the  same time. Using a maintenance load
chosen for illustrative  purposes in this  report,  it
appears not to be very advantageous to geographi-
cally target regions  individually to achieve  a par-
ticular load.
                                                87

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                    EXHIBIT S3A. GEOGRAPHICALLY TARGETED ADDITIONAL UTILITY AND
                    INDUSTRIAL SO2 REDUCTION IN CONTIGUOUS RADM SUBREGIONS


Deposition
Subregions



Adirondacks
4.7 kg-S/ha
45,44,5,4,2,3,9,
11,12,14,8,10,13,
15,7,18,20,22
Sensitive Region
Mid- Appalachians
6.9 kg-S/ha
13,19,20,10,
15,28,27,18,22


Southern
Blue Ridge
5.5 kg-S/ha
29,30,31,39,38,
37,28,21,19,36,27,
15,20,22,23,40
                 EXHIBIT 53s. GEOGRAPHICALLY TARGETED ADDITIONAL UTILITY AND INDUS-
                     TRIAL SO2 REDUCTION IN MAJOR RADM SUBREGIONS CONTRIBUT-
                               ING TO DEPOSITION (NOT CONTIGUOUS)

Deposition
Subregions
Sensitive Region
Adirondacks
4.7 kg-S/ha
15,13,5,14,7,20,
22,44,10,12,45,
17,9,25,26,3
Mid- Appalachians
6.9 kg-S/ha
13,20,15,22,19,
28,27,18,10
Southern
Blue Ridge
5.5 kg-S/ha
29,31,39,38,37,
28,40,20,51,27,
21,30,22,15,25
                           EXHIBIT 54. EXTENT OF CONTIGUOUS GEOGRAPHIC
                            REGIONS FOR ACHIEVING TARGETED DEPOSITION
                           LOADS EQUIVALENT TO ADDITIONAL NATIONWIDE
                              UTILITY AND INDUSTRIAL SO2 REDUCTIONS
                   EXHIBIT 55. GEOGRAPHICALLY TARGETED REDUCTIONS WITH A MAINTE-
                   NANCE LOAD OF 5 KG-S/HA IN MAJOR RADM SUBREGIONS CONTRIBUT-
                               ING TO DEPOSITION (NOT CONTIGUOUS)


Subregions


Adirondacks
15,13,5,14,7,20,
22,44,10,12,45,17
Sensitive Region
Mid- Appalachians
13,20,15,22,19,28,
27,10,18,31,21,12,
25,14,37,39,29,26,
7,51,34,32,23,9

Southern Blue Ridge
29,31,39,38,37,28,40,
20,51,27,21,30,
22,15,25,41,32,36
                                              88

-------
                                           CHAPTER 4
            POTENTIAL BENEFITS OF AN ACID DEPOSITION STANDARD ON
        VISIBILITY, HUMAN HEALTH, MATERIAL, AND CULTURAL RESOURCES
4.1  INTRODUCTION
Chapter 2 of this report focused on the effects of
acidic deposition on sensitive aquatic: and terres-
trial resources. Acidic deposition and its precursors
affect a broad range of  resources, including human
health, visibility, and materials.  Actions that may
influence sulfur dioxide and  nitrogen oxide emis-
sions  levels might also  affect these other resources.
This chapter discusses the type of ancillary benefits
that further reductions  in sulfur dioxide and nitro-
gen oxide emissions might have on visibility, hu-
man health, and materials. However, the level of
potential benefits  depends heavily  on the types
and level of any such reduction and cannot  easily
be estimated without knowing such  information.

4.2  RELATIONSHIP OF VISIBILITY TO ACIDIC
     DEPOSITION
In certain areas of the  United States, visibility is a
significant environmental  indicator  of  air  qualiiy.
Visibility impairment  and subsequent improve-
ment  is therefore a strong measure of effectiveness
and benefits.  This section  identifies regions in the
United States subject to visibility degradation and
describes how  these areas could benefit from  an
acid deposition standard that, in this case, is de-
signed to protect sensitive aquatic resources.

4.2.1  Visibility Impairment
Visibility refers to the degree  to which the atmos-
phere is transparent to visible light.  Fine particles
in the atmosphere absorb and scatter light, thereby
limiting visual range, decreasing color discrimina-
tion, and obscuring details of distant objects.  Im-
pairment of visibility depends on several  factors,
especially the size and composition of particles in
the viewing path. Some gases absorb visible light
and can impair visibility. Visibility is also  affected
by the angle of sunlight and so varies with time of
day and season.  Humidity can reduce  visibility
when hygroscopic particles absorb  water  and  in-
crease in size;  larger particles scatter more light.
Thus, natural  visibility  in the  humid East is gener-
ally poorer than in the more arid West.

Visibility can be impaired  by natural and anthro-
pogenic sources. Natural sources include fog, pre-
cipitation,  sea mist, windblown  dust,  volcanic
emissions,  and forest fires.  Visibility impairment
from these sources varies by season and  meteoro-
logical condition. Anthropogenic sources include
gaseous and particulate emissions  from stationary
and  mobile sources. Most  visibility impairment
can be traced to one gas,  nitrogen dioxide, and
five particulate substances:  sulfates,  nitrates, or-
ganics, elemental carbon, and soil dust.

The  National  Academy of  Sciences (NAS)  esti-
mated  the  contribution of anthropogenic air pol-
lutants to visibility impairment in three areas of the
country:  the  East (i.e.,  states  east  of the  Missis-
sippi), the Southwest (i.e., California, Nevada, Ari-
zona, New Mexico,  Utah, and Colorado), and the
Northwest   (i.e.,   Oregon,   Washington,  and
Idaho).112 Exhibit 56 summarizes findings for rural
regions in each area based on conditions prior to
the implementation of the Clean Air  Act Amend-
ments.   NAS  also  calculated  that  anthropogenic
sources are responsible for seven-eighths  of  the
visibility impairment in the East, five-eighths in the
Northwest, and three-eights in the Southwest.


           EXHIBIT 56. ANTHROPOGENIC
   CONTRIBUTIONS TO VISIBILITY IMPAIRMENT112
Contaminant
Sulfates
Organics
Elemental Carbon
Suspended Dust
Nitrates
Nitrogen Dioxide
Percent
East
65
14
11
2
5
3
Southwest
39
18
14
15
9
5
Northwest
33
28
15
7
13
4
The  exhibit clearly indicates  that most visibility
impairment in the  East  is  caused  by  sulfates
(transformation products of sulfur dioxide, the ma-
112Committee on Haze in National Parks and Wilder-
   ness Areas. 1993. Protecting  Visibility in National
   Parks  and Wilderness  Areas. National  Research
   Council and National Academy of Sciences, Wash-
   ington, DC
                                                89

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
jority of which is emitted from power plants). No
single  source category  dominates visibility  im-
pairment in the West, although sulfur compounds
play a significant role. In relatively clean areas of
the West, small increases in pollutant levels  can
markedly  degrade  visibility.  Thus,  visibility in
Class I areas in the West is especially sensitive to
increased levels of pollution.

4.2.2  Visibility Protection Laws and Class I
       Areas
Section 169A of the Clean Air Act (CAA) of 1977
established  as a national goal  "the prevention of
any future  and the  remedying of any existing im-
pairment of  visibility in  mandatory Class I  areas
which impairment results from  man-made air  pol-
lution." Class I federal areas are defined in  CAA
Section 162(a) as international parks, national  wil-
derness and  memorial   parks  exceeding  5,000
acres, and national  parks exceeding 6,000 acres in
existence  in 1977.  EPA,  in conjunction  with the
Department of the  Interior (DOI), has designated
158 mandatory Class I areas where visibility is im-
portant.

States  that  either have  Class I  areas or  contain
sources that may contribute to visibility impair-
ment of these areas are required to include in  their
state implementation plans (SIPs) a long-term strat-
egy for making reasonable progress toward reduc-
ing impairment. Major stationary sources reason-
ably expected  to  contribute to visibility impair-
ment in a  Class I area must install best available
retrofit technology.

The  prevention of  significant  deterioration (PSD)
provision  in Sections 160-169 of the CAA  also
applies to visibility protection. The PSD program,
which is directed toward new sources, requires
that  major emitting facilities seeking  to locate in
clean-air areas (i.e., areas meeting the  National
Ambient Air Quality Standard  [NAAQS] for a  par-
ticular pollutant)  use best available control tech-
nology (BACT). The source must also  comply with
air quality increments that specify the maximum
permissible increase in ambient pollutanl levels for
SO2, NO2, and particulate matter. Class I  areas are
further protected by the designation of Air Quality
Related Values (AQRV) for several parameters, in-
cluding visibility.  In addition to complying  with
BACT  and  increment requirements, new sources
must demonstrate that they will not adversely af-
fect an area's AQRV.
When the PSD program was created in 1977, large
national  parks and  wildernesses were  designated
as Class  I areas to provide them  with  special air
quality protection. Other parks and wilderness ar-
eas have been designated Class I  in succeeding
years. About two-thirds of the current Class I areas
are west  of the Mississippi. Nearly one-quarter are
located  in four southwestern states: Utah, Colo-
rado, Arizona, and  New Mexico.  Monitoring visi-
bility conditions at some sites was initiated by the
National   Weather Service in  1978. At approxi-
mately 43 other sites, visibility monitoring began
in 1987 and continues under a multi-agency pro-
gram  called the  Interagency Monitoring of  Pro-
tected  Visual  Environments (IMPROVE).  A rule-
making effort on regional haze protection recently
initiated  by EPA will further examine visibility im-
pacts on  and protection of Class I areas.

4.2.3  Visibility Metrics and the Projected
       Impact of the CAAA on Visibility
No standard or EPA-approved method for measur-
ing optical air  quality exists.  Visibility has been
measured and reported  in several  ways. Standard
visual range is  based on  human  perception  of  a
large black object placed in the sky and is reported
in kilometers or miles. A more scientific measure
of visibility impairment  is light extinction.  An ex-
tinction coefficient  Is proportional to the  attenu-
ation of  light per unit distance due to absorption
and scattering of light by particles or gases. Extinc-
tion coefficients are a function of particle size and
shape and the gaseous chemicals  present. A third,
recently developed  measure is the deciview scale
(analogous to the decibel  scale for sound), which
provides  a haziness  index designed to be linear to
humanly  perceived  changes  in visibility  caused
solely by air quality changes. The deciview (dv)
scale is near zero for pristine atmospheric condi-
tions and increases  as visibility degrades; a 1 dv
change  corresponds to a  10 percent  change  in
light extinction  and  approximates a  minimum,
commonly observable visibility change.

Several  recent  visibility studies have  been  con-
ducted to assess the  impact of the CAAA on visibil-
ity  improvement; some  analyses  specifically as-
sessed the impact of Title IV.

Eastern  United States
A recent analysis compared standard visual range
with and without Title  IV  of the CAAA to assess
                                                90

-------
                                             CHAPTER 4: POTENTIAL BENEFITS OF A STANDARD ON VISIBILITY,
                                                   HUMAN HEALTH, MATERIAL, AND CULTURAL RESOURCES
economic benefits of improvements in visibility.113
The  visual  range maps (Exhibits 57 and 58) illus-
trate  impressive  changes in visibility  associated
with Title IV. Although results are preliminary, the
economic analysis indicates potentially significant
monetary benefits to residential  areas of 31 eastern
states in the United States and to national parks in
the southeastern  United States.   However,  there
has been no analysis of whether there would be
any  monetary benefits  for deeper  and more ex-
pensive emissions reductions beyond the CAAA.

An analysis was conducted  for this study to com-
pare current and future perceptible visibility deg-
radation  in the East, assuming  implementation of
Title IV in the year 2010. The assessment indicated
a noticeable improvement in visibility  across the
eastern United States (Exhibit 59) from the  1980
base year, with most of the change occurring in
the warm seasons.

A 1993 EPA Report to Congress presented visibility
improvements to Class  I areas  that could  be ex-
pected to accompany implementation of the 1990


       EXHIBIT 57: ANNUAL AVERAGE VISUAL
     RANGE (KM) PROJECTED FOR 2010 WITHOUT
       TITLE IV: SOTH-PERCENTILE VISIBILITY
113  Chestnut, L.C., R.L  Dennis,  and  D.A. Latimer.
    1994.  Economic  benefits of  improvements  in
    visibility: Acid rain provisions of the 1990 Clean
    Air Act  Amendments. Presented at  Aerosols and
    Atmospheric Optics: Radiation Balance and Visual
    Air Quality, Air & Waste Management Association
    International  Specialty  Conference,   Snowbird,
    Utah, September 30.
  EXHIBIT 58. ANNUAL AVERAGE VISUAL RANGE (KM)
   PROJECTED FOR 2010 WITH TlTLE IV, INCLUDING
        TRADING: 50TH-PERCENTILE VISIBILITY
   EXHIBIT 59. ANNUAL AVERAGE IMPROVEMENT IN
   50TH-PERCENTILE VISIBILITY (DV) FROM 1980 TO
     2010 WITH TITLE IV, INCLUDING TRADING
CAAA.114 The analysis evaluated impacts of con-
trol provisions for NOX, SO2, and paniculate mat-
ter by assuming implementation of key provisions
of Titles I, II, and IV of the CAAA.  Exhibit 60 lists
specific provisions of each title. The analysis did
114  Office of Air Quality  Planning  and Standards.
    October  1993. Effects of the 1990 Clean Air Act
    Amendments on Visibility in Class I Areas: An EPA
    Report to Congress. U.S. Environmental Protection
    Agency, Washington,  DC.
                                                91

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                         EXHIBIT 60. SPECIFIC PROVISIONS OF TITLES I, II, AND IV
  Title I
* Application of reasonably available control technology (RACT) NOX control for ozone-
  moderate areas (or worse) and ozone transport regions
» Enhanced motor vehicle inspection and maintenance (I&M) for areas with conditions
  classed as ozone serious (or worse) and carbon monoxide-moderate (or worse)
» New Source Performance Standards (NSPSs) for NOX
 Title II
  1995 tailpipe standards for NOX reduction
  Using oxygenated fuels in nonattainment areas for carbon monoxide
 Title IV
• Implementation of Phase I and II SO2 limits, where emissions are based on projections
  from the Coal and Electric Utilities Model for the EPA Regulatory Impact Analysis (RIA)
* Implementation of NOX controls
not assume attainment of  the  particulate matter
(PM) standard in all areas which will also have an
impact on visibility in some regions, but rather as-
sumed across the board PM reductions in nonat-
tainment areas and  estimated PM  levels using size
range distributions for each  source category. Emis-
sions  inventories based on other Clean Air Act
controls such as NOX controls that nonattainment
areas  might impose to achieve the ozone standard
were  unavailable at the time  of  development of
the 1993 Report to Congress.

Because sulfates dominate visibility impairment in
the East, and no single chemical species dominates
in the Southwest, EPA modeled each region sepa-
rately. EPA used the  Regional  Acid  Deposition
Model (RADM) post-processor (EM-VIS) to calcu-
late visibility for each  RADM  cell in 1985  and
2010. The  1985 NAPAP emissions inventory was
used  as the basis for  1985 values.  For  the  year
2010, implementation of Title IV was assumed, but
without SO2 emissions trading, because an emis-
sions  scenario  depicting trading  was unavailable
at that time. (Note, however, that the  analysis of
the environmental  impact  of trading allowances,
described in  Section 3.5.1, found  only minimal
differences in deposition due to  trading.) Reduc-
tions  in Canadian SO2 emissions predicted by En-
vironment Canada as part of the 1990  NAPAP In-
tegrated Assessment were  also used in the  2010
modeling.

Extensive comparisons of percent  change in visual
range were made  using annual  average change
and  90th-percentile  worst   days   (i.e.,   only
10 percent of  days  have  worse visibility)  and
10th-percentile best days (i.e.,  only 10 percent of
days  have better visibility). For  50th-percentile
visibility days (half the days have better and half
                                          worse visibility), the percent  increase in visual
                                          range in the East ranged from 10 to 20 percent in
                                          Florida, New England, and just east of the Missis-
                                          sippi to 30 to 40 percent in the mid-Appalachians
                                          and the Ohio Valley. The largest improvement in
                                          50th-percentile visibility range in Class I areas was
                                          predicted to be in Shenandoah National  Park in
                                          the mid-Appalachians.

                                          Western United States
                                          To illustrate the visibility impact on western Class I
                                          areas in the 1993  Report to Congress, EPA con-
                                          ducted  a comprehensive analysis of  changes in
                                          visual range resulting from implementation of the
                                          1990 CAAA, including  the development of emis-
                                          sions inventories  for  anthropogenic  sources  of
                                          NOX, SOX,   and particulates  for  1988,  a  2005
                                          base-case  scenario,  and a  2005  CAAA-imple-
                                          mented scenario.115 The 1985 NAPAP emissions
                                          inventory served as the basis for the annual inven-
                                          tories. Electric utility emissions estimates in  the
                                          NAPAP  inventory  were  replaced  by emissions
                                          from the more  up-to-date  National  Allowance
                                          Data Base  (NADB).  Emissions estimates  for two
                                          large smelters near the border in Mexico were also
                                          included in  the inventories. These emissions inven-
                                          tories were  used  to  model projected air quality
                                          changes from 1988 to 2005. (EPA's 1993 Visibility
                                          Report to Congress contains a detailed description
                                          of models used and assumptions made.)
                                          115 Visibility modeling described here was conducted
                                             before this  study was initiated;  thus, inventories
                                             used are slightly different from those described in
                                             Chapter 3.  Differences in the inventories should
                                             not  significantly  affect   qualitative  conclusions,
                                             however.
                                                92

-------
                                            CHAPTER 4:  POTENTIAL BENEFITS OF A STANDARD ON VISIBILITY,
                                                  HUMAN HEALTH, MATERIAL, AND CULTURAL RESOURCES
Comparing emissions for SO2, NOX, and
fine  participates revealed only minor dif-
ferences  in the total  emissions  of each
species between 1988 and 2005. Emissions
for the three  scenarios by source category
for SO2  and NOX show that SO2 emissions
are equally distributed among area, utility,
and  other point sources  (e.g.,  smelters,
refineries, and pulp mills). Of the  100,000-
ton decrease in utility emissions between
1988 and  2005, over half is a  result  of
scrubbing  at  the   Navaho station. NOX
emissions  are  primarily  attributable   to
motor vehicles and area sources.  Mobile
sources account for most of the decrease in
NOX emissions, whereas  utility emissions increase
slightly.

Three-hour average  visual range estimates were
developed for  representative  Class  I areas  in  six
geographic regions: Central Coast (California),  Si-
erra,  Southern  California,  Desert  Southwest,
Golden  Circle  (Arizona), and  Rockies.  Exhibit 61
summarizes annual visual range estimates for each
inventory scenario calculated  from the 3-hour av-
erages.

Exhibit 61  indicates that neither the growth  in
emissions between 1988  and 2005 nor implemen-
tation of the 1990 CAAA at sources in the South-
west will have an  appreciable effect on  visual
range in  Class I areas.  The  insensitivity of pre-
dicted visibility changes  between the years 1988
and 2005, even with implementation of the 1990
CAAA in the latter year,  is clearly consistent with
the relatively small changes in SO2/ NOX, and par-
ticulates during this period.
4.2.4  Potential Impact of Further Sulfur
       Dioxide Reductions on Visibility
While  EPA cannot estimate the impact of an acid
deposition standard without first defining the type
and level of the standard, EPA  has studied the im-
pact of further  SO2 reductions on visibility  in the
East. For this study, visibility in the East was  calcu-
lated using the  RADM EM-VIS  model for two SO2
emissions scenarios described  in Chapter 3. Visi-
bility ranges for the post-2010 full implementation
scenario (reflecting the 8.95-million ton  SO2 cap)
and for the additional utility and industrial SO2 re-
duction scenario  (approximately 44 percent  de-
crease in SO2 emissions beyond CAAA reductions)
were  calculated  for 90th-percentile worst days.
Maps  in  Exhibits 62  and 63  show  percentage
changes  in  annual average visibility for these two
EXHIBIT 61. AVERAGE ANNUAL VISUAL RANGE ESTIMATES
FOR REPRESENTATIVE CLASS I AREAS IN THE SOUTHWEST
Geographic Region
Central Coast
Sierra
Southern California
Desert Southwest
Golden Circle
Rockies
Representative
Class I Areas(s)
Pinnacles
Yosemite
San Gorgonio
Chiricahua
Grand Canyon
Arches
Rocky Mountain
Bandelier
Visual Range (km)
1988
96
104
68
118
134
116
121
119
2005
CAAA
94
101
66
115
132
115
120
116
  EXHIBIT 62. PERCENT INCREASE IN VISUAL RANGE FROM
   1985 TO 2010 WITH FULL CAAA IMPLEMENTATION
     EXHIBIT 63. PERCENT INCREASE IN VISUAL RANGE
        FROM 1 985 TO 2010 WITH ADDITIONAL
           SO2 REDUCTION BEYOND CAAA
                                               93

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
scenarios  between  1985  and  2010. To
assess the impact of changes in visibility
due only to decreases in  ambient sulfate
concentration, visibility impairment  from
other ambient species remained constant
in the models.

The greatest improvements in visual range
between 1985 and  the baseline scenario
in 2010  lie in a  band  from  northern
Mississippi  to  southwestern New  York
State.   Improvements   in   visibility  for
Class I areas in  the mid-Atlantic  region,
which includes  the Great Smoky Moun-
tains  and the Shenandoah  Valley, range
between  30 percent  and   more  than
40 percent.  For the  additional  SO2 re-
duction  case,   improvements  in visual
range of greater than a factor of two are
predicted for these key mid-Atlantic areas.
Increases   in  visual   range  of  60 to
100 percent (i.e., the ability to see twice as far) are
predicted for coastal areas and New England.

The study conducted by EPA for  Class I areas in
the Southwest, the region  of the country with the
largest concentration of Class I areas, demonstrates
that  no  single  pollutant  or source category is
responsible for most of the visibility impairment in
that   region. Thus,   while  an   acid  deposition
standard could  reduce ambient sulfate,  nitrate, or
NO2  levels, projection of potential improvements
in visibility for specific Class I areas would require
additional model analysis.

4.3  RELATIONSHIP OF HUMAN HEALTH TO
     ACIDIC DEPOSITION
Exposure  to SO2,  paniculate  matter  (including
sulfate and  nitrate  aerosols, some of which are
acidic), NO2, and  ozone (O3) in ambient air can
cause adverse health effects. (Ozone is a related
concern for acidic emissions and deposition  stan-
dards because  NOX is a  major  precursor in O3
formation.) Possible health effects related to acidic
deposition and  its  precursors are quite complex
because of the variety of pollutants, possible routes
of    exposure,   and    mechanisms  involved
(Exhibit 64).

Current  applicable  standards   include National
Ambient Air Quality Standards  (NAAQS) for  SO2,
NO2, and O3.  These standards  are designed to
protect human health from  all significant known
health  effects   due  to  these  pollutants.     The
standards are subject to revision as new scientific
EXHIBIT 64. RELATIONSHIP OF ACIDIC DEPOSI-
    TION PROCESSES TO HEALTH EFFECTS
       Nitrogen Oxides
                        VOCs
   information  becomes  available.    Further,  the
   CAAA, through  the  State  Implementation  Plan
   process, has the mechanisms in place to bring all
   areas  into attainment.  Several  studies  show  that
   the 10-million  ton  SO2 emissions reduction from
   1980 levels under Title IV is expected to result in
   human health  benefits associated with reduced
   SO2 and fine particulate exposures.  Analyses are
   ongoing to quantify these benefits.

   The following  subsection summarizes  health  ef-
   fects  and potential  risks associated with airborne
   acidic pollutants. Current knowledge on possible
   future  risks to  human  health  associated  with
   changes in  acidic deposition rates is summarized
   in the second subsection.

   Several respiratory problems can be caused or ag-
   gravated  by ambient  air concentrations of  SO2,
   particulate  matter  (including sulfate and  nitrate
   aerosols), NO2, and O3 (separately and in combi-
   nation).  Effects  include  aggravation of  existing
   cases, as well as  new cases, of chronic  bronchitis,
   bronchoconstriction, other pulmonary function im-
   pairments,  chest  discomfort, cough, lung inflam-
   mation, increased incidence of infectious respira-
   tory  disease,  and increased mortality  rates.  The
   elderly, the very  young, and individuals with  pre-
   existing respiratory diseases, such as asthma, are at
   greatest risk and  would benefit most from reduc-
   tions  in the atmospheric concentrations of these
   pollutants.

   Under Sections 108  and 109  of  the CAAA,  EPA
   establishes  primary NAAQS,  which protect the
   most sensitive segments of the population, with an
                                                94

-------
                                             CHAPTER 4:  POTENTIAL BENEFITS OF A STANDARD ON VISIBILITY,
                                                   HUMAN HEALTH, MATERIAL, AND CULTURAL RESOURCES
adequate margin of safety. Under current statutory
deadlines,   all  areas   are  required  to  be  in
attainment with all health-based standards.   If an
acid deposition standard were adopted and to the
extent such deadlines slip, however, the additional
emissions  reductions   achieved   by  an   acid
deposition standard could facilitate the attainment
and  maintenance  of  the  primary  NAAQS es-
tablished under Sections 108 and 109 of the Act.

A  number of recent epidemiological studies  have
associated particulate pollution  with excess  mor-
tality and morbidity at levels below the existing
24-hour particulate matter NAAQS.  For  example,
a decade-long, six-city study provided evidence of
a strong association between fine  particulates, in-
cluding sulfates, and mortality in humans and also
indicated that acidic aerosol concentrations  were
directly associated with increased prevalence of
bronchitis in children.115 Clinical studies suggest
that  asthmatics may exhibit sensitivities  to  short-
term exposures to  acidic aerosols.  As a result, EPA
has initiated the review of the air quality criteria
and  standards for particulate  matter,  including
acidic aerosols. Additional research shows that
sulfate aerosols comprise the majority of acidic
aerosols in ambient air and a large share of total
ambient inhalable  particulate matter in the eastern
United States.117

Based on the available data, many in the  scientific
community believe that if the mortality and  mor-
bidity effects observed  in these studies are causal,
the agent(s) is more likely to be fine particles (<2.5
jim) than coarse particles (2.5 to  10 ^.m). If it is de-
termined, after completion of the ongoing review,
that a new fine particulate standard(s) is appropri-
ate, the associated control strategies will  focus on
the control of the precursors (e.g., SO2, NOX, am-
monia, and condensible hydrocarbons) to  secon-
dary fine particulates. While still  speculative, if the
science eventually bears out this scenario, the
116  D.W. Dockery, C.A.  Pope, X. Xu, J.D. Spengler,
    J.H. Ware, M.E. Fay, B.C. Ferris, and F.E. Speizer.
    1993. An association  between  air pollution  and
    mortality in six U.S. cities. New England Journal of
    Medicine 329:1753-1759.

117  L.G. Chestnut.  September 1994. Human Health
    Effects  Benefits  Assessment  of  the  Acid  Rain
    Provisions of the 1990 Clean Air Act Amendments.
    (Draft final report.)  Prepared for the  Acid  Rain
    Division,  U.S. Environmental Protection Agency,
    Washington,  DC. (Final  report  to be  released
    October 1995.)
reductions needed to  attain a  revised  primary
particulate standard would overlap and reduce the
need for an acid deposition standard.

Insofar as  acidic SO42~ trends roughly parallel total
SO42-trends, NAPAP estimated that  between 2000
and 2020, the  region generally  incorporating the
states of Ohio,  Indiana, West Virginia, Pennsylva-
nia, New  Jersey, Maryland, Virginia,  North Caro-
lina, Kentucky,  Tennessee, and  northern  Georgia
could experience  the greatest decrease in acidic
sulfate  levels   from   implementation   of   the
CAAA.118  The upper Midwest (Michigan and Wis-
consin)  and the upper Northeast (Maine and  New
Hampshire),  which  had  lower  ambient  1-hour
sulfate levels, are estimated  to have only slightly
improved  atmospheric concentrations in the years
2000  and  2020 under this scenario. Several ongo-
ing benefit assessments will  address the extent of
monetary  health benefits associated  with imple-
mentation of Title IV.119

With respect to NO2, no area of the United States
presently exceeds  the NAAQS of 0.053 ppm, an-
nual average. The attainment of the annual  stan-
dard also  significantly  limits the  likelihood and
magnitude of short-term 1-hour  peak NO2 levels.
NO2 and its transformation products, however, are
precursors to O3 formation and to nitrate aerosols,
which are  a  common component of fine  particu-
lates.  Consequently, reductions  in  NO2  or  NOX
emissions  are key components of the particulate
and O3 control  strategies.

4.4  RELATIONSHIP OF MATERIALS DAMAGE AND
     CULTURAL RESOURCES TO ACIDIC
     DEPOSITION
All materials exposed to the  outdoor environment
are subject  to degradation caused by   natural
weathering  processes  involving moisture, heat,
oxygen, solar radiation, bacteria, and fungi. Ad-
verse  effects from these processes can be  acceler-
ated by deposition of wet and dry acidic air pol-
lutants. Several  NAPAP reports, including the State
of the Science and Technology Report No. 19, the
1990  Integrated Assessment Report, and the 1992
Report to Congress, considered the delivery of wet
118  This section is drawn primarily from the National
    Acid Precipitation  Assessment  Program,  1991
    (7990 Integrated Assessment Report. NAPAP Office
    of the Director, Washington, DC).
119  See footnote 117.
                                                95

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
and dry deposition to various types of material sur-
faces, particularly in urban areas.

4.4.1  Acidic Deposition  Effects on Materials
       and Structures
Research  conducted by  NAPAP and other pro-
grams confirms that certain  metals  such as galva-
nized steel are corroded (Exhibit 65), surface coat-
ings  are degraded,  and  carbonate stone is dis-
solved by SO2,  wet-deposited acids, and natural
atmospheric acidity from  dissolved CO2. Spotting
of automotive finishes can also occur from acidic
deposition.

To illustrate the potential role of acidic deposition,
Exhibit 65   shows   that   an   estimated   31 to
78 percent of the dissolution of  galvanized steel
and copper continues to be attributable to wet and
dry acidic deposition.  Acidic deposition  can also
accelerate  deterioration  of  stone  through  three
processes: dissolution  and erosion  of material or
surface  features; blackening of the stone surface;
and  cracking,  splintering,  and  chipping of the
stone surface. One of  the most visible  and docu-
mented forms of pollutant  damage to  limestone
and marble  is the accumulation of dark  gypsum (a
mineral formed from  calcium sulfate and water)
crusts in areas sheltered from rain. Sulfur deposited
onto carbonate  stone  (e.g., marble) reacts with
calcite to form a black crust containing  a mixture
of gypsum,  fly ash  particles, soot,  and  biological
growth, all of which can  cause decay deeper into
the stone.  Laboratory and  field studies show a cor-
relation between dry deposition of SOX  and a thin
black accumulation  on masonry material!;, includ-
ing sandstone, granite,  and brick.  European studies
show that SO42" and NO3~ concentration? in  stone
occur in proportion to atmospheric concentrations
of SO2 and NOX. Erosion rates of stone in Europe,
however, are significantly greater than those found
in North America because of the  higher ambient
concentrations of acidic deposition precursors in
Europe.

Rates  of damage  to  materials  associated  with
acidic  deposition  depend  on  atmospheric  and
structural factors that influence delivery of the pol-
lutant  to a  material's  surface,  i.e., regulate its
"dose." Wet deposition  delivers atmospheric  pol-
lutants to  surfaces of  buildings,  structures,  and
other objects primarily  through  rainfall, snowfall,
fog,  dew,  and frost.  Dry  deposition  provides a
more constant delivery of pollutants as large parti-
cles  fall with gravity and small  particles and trace
gases are delivered by atmospheric  turbulence.
Pollutants delivered by turbulent processes can po-
tentially  damage a  greater proportion  of material
surfaces than can large particles. Also, a damp sur-
face provides a much more effective sink for sol-
uble trace gases (e.g., SO2) than does the same sur-
face when dry. Thus,  dry deposition can often be
intimately linked with the processes by which ma-
terial surfaces are wetted.

The  key  effect of concern for cultural  materials is
physical  damage, often  expressed  in terms of the
time it  takes for the  material to  lose  its  unique
qualities. For construction materials, the key effect
of concern is the expenditure to maintain an ac-
ceptable  level  of  functionality and  appearance
over the  life of the structure.

NAPAP studies reveal that U.S. regions with the
largest number of  cultural  and historical  monu-
ments also often have the highest  levels of acidic
             EXHIBIT 65. PERCENTAGE OF METAL CORROSION ATTRIBUTED TO ATMOSPHERIC FACTORS^
Metal
Study Region
Galvanized Steel
Adirondack Mtns, NY
Washington, DC
Steubenville, OH
Copper
Adirondack Mtns, NY
Washington, DC
Steubenville, OH
Corrosion Rate
(um/yr)
0.62±0.26
1.01 ±0.42
1.47±0.21
0.37±0.14
0.83±0.19
0.88±0.29
Dry Deposition
of Sulfur
6%
52%
56%
10%
38%
57%
Acidity
(hydrogen ion con-
centration)
25%
23%
22%
25%
25%
20%
Other Corrosion
Factors
69%
25%
22%
65%
37%
23%
  a Source: NAPAP. 1 993. 1992 Report to Congress. National Acid Precipitation Assessment Program, Washington,
   DC.
  b Corrosion rates are mean measurements from NAPAP field sites.
                                                 96

-------
                                             CHAPTER 4:  POTENTIAL BENEFITS OF A STANDARD ON VISIBILITY,
                                                   HUMAN HEALTH, MATERIAL, AND CULTURAL RESOURCES
deposition, including areas of the Northeast having
long settlement histories and the greatest number
of pre-Civil War  buildings and tombstones. Most
historic battlefields,  especially those with com-
memorative monuments,  are located east of  the
Mississippi.  Also, material selection has changed
over time;  more durable  materials  have replaced
acid-sensitive  marbles,   sandstones, and  zinc,
which were predominantly used  prior to  large-
scale western expansion and  late 19th-century
population  increases. Distributions of acid-sensi-
tive cultural materials are  therefore expected to be
less dense west of the Mississippi.

4.4.2  Material Life-Cycle and Damage
       Estimates
Potential acidic deposition-related damage to both
function  (e.g., building  material,  bridges,  and
automobile and  other exterior coatings) and cul-
tural value  (e.g., historical monuments and sculp-
tures)  potentially  represent  an  overall  cost  to
society. Acidic deposition  control  can therefore be
linked to  potential  monetary  benefits. Relating
acidic deposition-induced physical damage to the
shortened  usefulness of  materials  remains  an
important area of research. Quantifying changes in
maintenance and replacement cycles attributable
to changes  in acidic deposition  is  necessary for
estimating the economic consequences of physical
deterioration.  The  complexity  of  this  linkage
involves three primary areas:

  *  Extrapolation  of laboratory   findings to
     doses  on large structures caused by ambi-
     ent exposure;

  *  Tolerance to  decay  by  acidic  deposition,
     which varies with the specific function of
     the material within the structure; and

  *  Maintenance   and   replacement  cycles,
     which can be affected by a great many fac-
     tors other than acidic deposition, including
     market factors.

Clearly,  emissions reductions can  minimize  the
need for or frequency of public and private main-
tenance, repair, and replacement. EPA and NAPAP
are working on approaches to estimate the damage
to materials from acidic deposition. Assessing ma-
terial effects present an array of options for valuing
damages, costs,  and benefits of emissions reduc-
tions  in physical  terms  (e.g., corrosion  rates and
reduced service life), market terms (e.g., life-cycle
costing and shifts in material selection and  market
share), and nonmarket terms (e.g.,  heritage valu-
ation of damage  to historical  monuments and
buildings).  Several efforts have  been initiated to
determine  the  material damage  costs associated
with  acidic  deposition; damage to automotive
coatings  are  highlighted here as  one example of
analyses. EPA and NAPAP  have begun to investi-
gate the  costs  of damage  to  automotive finishes
and subsequent savings attributable to CAAA im-
plementation.

Spotting  of automotive finishes  can  occur  from
acidic deposition. This effect is most pronounced
on dark finishes and in  warmer climates, because
the rate of  chemical  reactions causing spotting in-
creases with temperature.  The damage,  typically
appearing as irregularly shaped etched areas, oc-
curs  after  evaporation  of a  moisture   droplet.
Automotive coatings may be damaged by all forms
of acidic deposition,  particularly when dry deposi-
tion is mixed with dew or rain. It has been  diffi-
cult, however, to quantify the specific contribution
of acidic deposition to paint finish damage relative
to other  forms  of  environmental fallout,  the im-
proper application of paint, or deficient paint for-
mulations.

Although the existence  of  damage  to automotive
coatings  has been  well documented, there has
been  little analysis of the economic costs imposed
by this damage. Such an analysis may include in-
vestigation  of issues  such as actions taken to pre-
vent the  negative effects of pollution (i.e., actions
taken  by car and truck manufacturers)  and the
market value of a car or truck damaged by acidic
deposition  (i.e., actions taken by automobile deal-
ers regarding damage which  has occurred). The
scoping exercise conducted by EPA and NAPAP
begins to illustrate the potentially large costs asso-
ciated with this type of  damage and therefore the
potential  benefits of the Acid Rain Program.120 Es-
timates of  the value of potential annual  residual
damage  to  cars  and  trucks   in  the  eastern
United States may range from  $50 million to over
$400  million.   Total  annual  costs   could  be
$96 million  to  $850 million.  While additional
benefits can  be expected for SO2 reductions be-
yond  the CAAA, there  have been no studies to
suggest what those marginal benefits might be.  It
is likely that the marginal cost of additional reduc-
tions will increase.

Additional  materials  damage and pollution reduc-
tion benefit efforts are also underway to determine
120  ICF  Incorporated. September 30, 1994. Acid Rain
    Program Evaluation: Valuing Potential Reductions
    in Automobile Finish Damages-Scoping Study.
                                                97

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
the benefits of acidic deposition control on a func-     value placed on  one-of-a-kind resources.  These
tional item such as steel bridges as well as the po-     and other costs associated with acidic deposition-
tential benefits of control to preserve cultural  re-     induced damage would likely decrease with im-
sources of historical importance. Damage to cul-     plementation of an acid deposition standard, but
tural resources can result in potentially high repair     EPA has no current analyses designed to determine
and maintenance costs, replacement costs, and the     whether these costs are significant.
                                                 98

-------
                                          CHAPTER 5

                                  IMPLEMENTATION ISSUES
5.1  INTRODUCTION
In order to determine the effectiveness of an acid
deposition  standard for  protecting  sensitive  re-
sources, it is necessary to know not only the bene-
fits that would result from a given standard  if it
were established,  but  also  to  know  how  the
standard would be implemented.

In Section 404 of the CAAA, two  requirements that
Congress laid out for this acid deposition standard
feasibility study dealt with  implementation issues.
The statutory language calls for:

   * Description  of the  measures  that would
     need to be  taken to  integrate such  [acid
     deposition] standard or standards with the
     control program required by Title IV of the
     Clean Air Act; and

   » Description  of the  impediments  to imple-
     mentation of such control program [based
     on acid deposition  standard or standards]
     and  the  cost-effectiveness  of  deposition
     standards   compared   to  other  control
     strategies  including  ambient  air  quality
     standards,  new source performance  stan-
     dards and the requirements of Title IV of
     the Clean Air Act.

This chapter describes two basic  approaches to
implementing an  acid deposition standard. Under
the first approach (referred to here as a regional
targeted approach), EPA  would  set a  standard or
standards, either  using existing  authority (if ade-
quate) or seeking further authority from Congress
to set such  standards and  provide deadlines for
their attainment.  Then,  similar  to Title  I, states
would  determine  source-specific  limits  using
source-receptor models and cost analyses,  incor-
porate those limits  in state implementation plans
(SIPs), and enforce them.  If  one or  more states
failed to  do the above,  EPA would promulgate a
Federal Implementation Plan (FIP).

Under the second approach (referred to here as a
national   emissions-based  approach),  Congress
would direct EPA to set  a  deposition  standard or
standards  and to determine  the national  (or  re-
gional) emissions levels for sulfur dioxide and ni-
trogen  oxides  that  would  meet  those standards.
Congress would then set an emissions cap and al-
lowance allocations for  nitrogen  oxides  and, if
necessary, adjust the cap for sulfur dioxide in Title
IV; and provide a  timetable for meeting the new
caps.  EPA would use Title IV provisions to imple-
ment the emissions programs.

For these two basic approaches, this chapter will:

   * Describe how  each would be carried out,
     including any measures that would need to
     be taken to integrate it with Title IV;

   * Describe any  impediments to  implementa-
     tion, including the need for any additional
     statutory authority; and

   * Discuss their relative cost-effectiveness.

To provide a rough comparison of the cost-effec-
tiveness of the two approaches for sulfur reduc-
tions, estimates are  made of the cost of achieving
the same reduction in sulfur deposition at the three
sensitive areas under each approach. The example
used is based on a 50 percent reduction in  na-
tional utility SO2  emissions  beyond that required
by the 1990 CAAA (as described and modeled in
Chapter 3). The  use of this example in  no way
suggests that such a reduction is necessary, appro-
priate, or sufficient, but is merely put forth for il-
lustrative purposes.

There are, of course, many variations on these  ba-
sic approaches, and probably other ways to realize
the goals of an acid deposition standard. The pur-
pose here is not to provide  a complete examina-
tion of this issue,  but  to show how existing  ap-
proaches could be adapted  to implement such a
standard and to compare costs.

Finally, both illustrative approaches would require
monitoring for program effectiveness.  This  chapter
summarizes current monitoring efforts and identi-
fies deposition and effects  (e.g.,  surface water)
monitoring as critical components to development
and implementation of an acid deposition standard
or standards as well as to  efforts to assess the im-
pact of the CAAA.
                                               99

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
5.2  REGIONAL TARGETED APPROACH
5.2.1  Description of Regional Targeted
       Approach
Under new  or existing  authority  (see  Section
5.2.3), EPA would  either set national standards for
sulfur and nitrogen deposition, or set different re-
gional standards for sulfur  and nitrogen based on
the different sensitivities  of different regions  to
sulfur  and  nitrogen. EPA  would  also  establish
deadlines for  the  attainment of such standards,
unless such deadlines  were established  through
new statutory authority.

States  would  determine the form  and  level  of
emissions limits for the sources of sulfur and  nitro-
gen that they determine relevant  to the attainment
of the national acid deposition standards. Their de-
terminations would probably be based on atmos-
pheric models and technical and cost analyses  of
the sources within  their jurisdictions.

The process for making these determinations and
incorporating  such  limits  into   SIPs  would  vary
from state to state. Some states  may wish to pro-
vide flexibility to their sources by allowing various
forms of emissions trading.  EPA would need  to re-
view  each SIP  and determine that it, in  conjunc-
tion  with other relevant SIPs, would attain and
maintain the specific acid  deposition standard  or
standards. States would  be  responsible for collect-
ing emissions  information and enforcing  limits. If
states failed to carry out these responsibilities, EPA
would impose  sanctions and/or  prepare  and im-
plement a FIP.  EPA would  implement such a pro-
gram  through Titles I,  II,  and  V.  The  program
would be evaluated through deposition and effects
monitoring.

5.2.2  Integration with Title IV
Setting an  acid deposition standard or standards
would  not,   by   itself,   directly   affect  other
environmental   programs   unless   it   required
emissions reductions.  The  specific  sources and
level of emissions  reductions would determine the
direct impact on other programs and the usefulness
of coordination and integration.

If no changes were made to Title IV, the allocation
and transfer of allowances would not be  restricted
by state  actions to set  new emissions limits, but
sources would not be able  to emit more SO2 than
their state limits allow, regardless of how  many al-
lowances they might hold. This is the same situ-
ation as currently exists, but if more stringent limits
were imposed on  a large number of sources, the
demand for and price of allowances would de-
cline. In fact, taken to extreme if the aggregate ef-
fect of new source-specific state limits were to re-
duce utility SO2 emissions below the current 8.95
million allowance  cap and the SIPs did  not allow
emissions trading,  the price of allowances should
theoretically drop to zero because they  would be
of no use.

Both in setting the level of the standard  and in its
implementation, an  acid  deposition standard or
standards should  be coordinated and potentially
integrated with several other environmental  pro-
grams, particularly attainment and maintenance of
primary and  secondary  National  Ambient  Air
Quality Standards (NAAQS) for SO2, NO2, O3, and
PM10; visibility protection; new source review; and
new source performance standards under Title I of
the CAAA.  It would also  be useful to coordinate
acid deposition standard-setting and implementa-
tion with water quality programs,  particularly in-
volving eutrophication of estuaries. Substantial re-
ductions  in  SO2  or NOX  could  assist  in the
achievement of the goals of these programs.

5.2.3  Impediments to  Implementation
It may be possible  to set acid deposition standards
under existing statutory authority. At this time, no
definitive determination has been made.  However,
clear direction  from  Congress  in this area would
certainly make implementation more feasible and
effective. For example, implementation  would be
facilitated by explicit authority to set  deposition
standards,  to  set  regionally  different  nitrogen
and/or sulfur standards, to set deadlines  for attain-
ment, to require uniform measurement and report-
ing of emissions for sources not already affected by
Title IV, and to establish  uniform procedures for
interstate trading of NOX  emissions. It may be dif-
ficult for sources  to conduct efficient  interstate
trading of NOX emissions without federal legisla-
tion.  It would also be useful if Congress specified
the degree of protection desired.

Time and resource issues may be significant  if, for
example,  a  regionally  targeted  implementation
approach shared some of the same administrative
complexities as the current SIP process to meet the
NAAQS. For instance, the development of  state-
specific regulations by states for  emissions limita-
tions, followed by EPA review and approval, can
be a resource-intensive and lengthy process. EPA is
currently working  to  streamline  this process. An-
other concern  is the incomplete and  sometimes
inconsistent state  emissions  inventory  data  upon
which compliance and  effectiveness  are  deter-
                                                100

-------
                                                                  CHAPTER 5:  IMPLEMENTATION ISSUES
mined. Furthermore, if a state is not in attainment
of the  NAAQS, it  is commonly a  long period of
time until  state-specific air quality levels  are
achieved.   Compliance  deadlines  for  certain
NAAQS have been refined and revised with each
amendment to the CAA. The uncertainties of this
process are not usually conducive to long-range
planning and  cost-effective compliance by  the
regulated community.

The 1990 CAAA recognized the role of long-range
pollutant transport in ozone formation and author-
ized EPA to create "ozone transport regions" where
nonattainment in  one state may be  the result of
emissions   in  another state. The  newly created
Northeast  Ozone Transport  Region extends from
Northern Virginia to Maine. The Northeast Ozone
Transport   Commission  is  currently  developing
plans to achieve attainment of the ozone standard
by  determining  both local  and  transport-region
wide  limits on  nitrogen oxides  emissions.  The
Commission is also considering market-based  ap-
proaches (e.g.,  NOX trading within the  transport
region) to  achieve maximum protection  at least
cost. This effort represents a possible variation on
the regional targeted approach.

Implementation of a targeted approach for deter-
mining and maintaining acid deposition standards
would  require, at a minimum, an enhancement of
existing deposition  monitoring  and  atmospheric
modeling  abilities.  Program requirements would
need to identify pollutants to be  monitored, and
determine standard procedures for  measuring wet
and dry deposition, spatial resolution, and tempo-
ral  requirements.  Enhanced effects  monitoring
(i.e., surface water monitoring) would be desirable
to evaluate effectiveness of the standard  or stan-
dards.

5.3  NATIONAL EMISSIONS-BASED APPROACH
5.3.1   Description of National Emissions-
       Based Approach
Congress would direct EPA to provide (1) a range
of target loads (standards) and emissions levels of
sulfur and  nitrogen designed to provide a range of
ecosystem protection (and other benefits), (2) lev-
els of  national  and regional  sulfur and nitrogen
emissions that met those target loads, and (3) esti-
mates of the benefits and  costs of meeting those
emissions levels.

Taking this information  into account,  Congress
would  amend Title IV of the CAAA by setting an
allowance  cap  for nitrogen, revising the cap for
SO2 (if necessary), including any new source cate-
gories  (if necessary), allocating allowances, and
providing timetables for the achievement of  the
new  caps.  EPA would implement the standards
through conforming changes to the Title IV rules.
The  program  would  be   evaluated   through
deposition and effects monitoring.

5.3.2  Integration with Title IV
Depending on any statutory changes enacted, EPA
would  make  conforming changes to the  Title IV
implementing regulations, 40 CFR Parts 72-78. Ti-
tle IV  permitting,  allowance  trading,  emissions
monitoring, penalty provisions and data  systems
would  be used to ensure compliance. Since the in-
formation that EPA would provide Congress as in-
put to their deliberations in setting allowance caps
would  include all  human health  and ecosystem
impacts from  sulfur and nitrogen known to EPA at
the time, coordination with standard-setting and
implementation of the other air and  water pro-
grams cited above would be greatly facilitated. If
the number of SO2 allowance were lowered, al-
lowance prices would rise. In certain cases, com-
pliance strategies of affected sources may change.

5.3.3  Impediments to Implementation
New statutory authority would be needed. Title IV
allowance levels for SO2 cannot be changed with-
out Congressional actions, and there is currently
no allowance program for NOX.

Administrative impediments  would be limited to
any difficulty  posed by the  statutory changes. Cur-
rently, Title IV does not appear to have any signifi-
cant  administrative  or  compliance  impediments.
The Title IV implementation process was designed
to address  regional  air pollution problems, espe-
cially those involving long-range transport of pol-
lutants  and  their transformation products. Congress
developed Title IV as a comprehensive program
aimed  at reducing sulfur dioxide emissions across
broad regions to achieve increased protection for
sensitive receptor areas both local to and hundreds
of miles downwind of major point sources of SO2
emissions.   Enhanced effects monitoring (i.e., sur-
face water monitoring) would be desirable to track
the effectiveness of deposition reductions.

5.4  ECONOMIC IMPACTS
If an acid  deposition standard involves emissions
reductions and requires stricter SO2  and NOX point
source  emissions controls, that standard would re-
sult in  direct  cost increases to utilities and indus-
                                               101

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
trial sources. The magnitude of cost increases, and
the sectors most affected would vary depending on
the regulatory approach selected,  the quantity of
emissions reduced, the sources affected, and the
timing of the reductions. For the scenarios consid-
ered in this study, at least two comprehensive sec-
tor models—one for the  electric  utility industry
and one for the industrial sector—would need to
be applied to fully  understand the wide  ranging
impacts that would result from an  acid deposition
standard. While such models  exist, budget and
time constraints did not allow for their use for this
study.  An  expanded economic  analysis would
consider variations in emissions and the timing of
achieving emissions reductions.

While the use of more precise sector models may
become  necessary if further emissions reductions
are pursued,  the scale of the economic analysis
presented here is scoping in nature and  limited in
its degree of detail. Instead, this study uses:

   * Analyses of the electric utility sector that
     have already been conducted by  EPA to
     support Title IV.

   * A spreadsheet-based scoping model to esti-
     mate broadly the range of costs associated
     with different regulatory approaches.

In Chapter 3,  the  impacts of various SO2 control
strategies on  deposition in sensitive regions  were
evaluated for the year 2010. The costs of the key
emissions control strategies described in  Chapter 3
are evaluated below. Costs are estimated for the
following control scenarios for the year 2010.

   * 201OCAAA Scenario,

   * Additional  50 Percent  Utility SO2 Reduc-
     tion Scenario,

   * Additional 50 Percent Utility and Industrial
     SO2 Reduction Scenario, and

   * Geographically Targeted  Utility SO2  Re-
     duction Scenarios.

The  costs of  additional utility and industrial NOX
reductions  are described  qualitatively.  With the
exception of  the  geographically  targeted  utility
SO2  reduction  scenario,  all  cost scenarios are
based on a national emissions-based approach.

5.4.1   2010 CAAA Scenario  (With Trading)
The 2010 scenario with trading described in Chap-
ter 3 is used as the  baseline scenario for compar-
ing costs of alternative emissions reductions sce-
narios. The costs of this scenario were developed
as part of the 1993 EPA Base Case Analysis121 used
to support rulemaking under  Title IV and  have
thus,  been  reviewed extensively.  EPA has esti-
mated that compliance with Title IV of the  1990
CAAA will cost electric utilities about  2.2 billion
dollars in the  year 2010 (Exhibit 66).  SO2 emis-
sions forecasts project that SO2 emissions decrease
by about 9.2 million tons in 2010. This  means that
the average cost of reducing SO2 is about $240 per
ton SO2 removed.  The marginal SO2 removal cost
(i.e., the cost  of  reducing  one additional ton of
SO2)  is forecast  to be  much higher,  however,
about $500 per ton SO2 removed.

With  Title IV fully  implemented,  electric utility
SO2 emissions are  forecast to equal about 9.5 mil-
lion tons in 2010. This is higher than the 8.95 mil-
lion ton SO2 allowance cap that is binding in 2010
because:

It is forecast that about 0.52 million tons of allow-
ances would be "banked" between 1995 and 2009
and used in 2010; and

Units  not affected by Title IV  (i.e.,  those with
nameplate capacity less than 25 megawatts) would
emit about 0.05 million tons of SO2 in 2010.

Exhibit 66 shows utility costs  and  SO2 emissions
forecasts in 2010 by U.S. census region. As can be
seen in the exhibit, the majority of SO2 emissions
reductions and compliance costs (about 60 percent
of the U.S. total) are expected to occur in the cen-
tral United  States  (i.e.,  East  North  Central  and
West South Central Census Regions).

5.4.2  50 Percent Utility SO2 Reduction
       Scenario
For the 50 Percent Utility SO2 Emissions Reduction
Scenario,  the 8.95 million ton electric utility SO2
emissions cap was cut  in  half (i.e.,  set equal to
4.48 million tons SO2). Under this scenario, costs
are estimated to increase to $4.8 billion and SO2
emissions are estimated to decrease by an  addi-
tional  5 million tons in 2010 relative to the CAAA
Scenario (Exhibit 67). The average cost of reducing
SO2 emissions by 5 million tons is about $955 per
ton SO2 removed, which is almost four times the
average  cost of emissions reductions  forecast  in
 121 Economic Analysis of the Title IV Requirements of
   the 1990 Clean Air Act Amendments. 1994.  U.S.
   EPA, Office of Air and Radiation, Acid Rain
   Division, February.
                                                102

-------
                   CHAPTER 5: IMPLEMENTATION ISSUES






o
'C
re
CD
U
00
1
U









CD

P
O
Z
c cu re
CD Qi —
l°!
O 0^ ^
£ <= s
*£ 5 ^
U .2
"re ™ •
3 ^

C &r
^.

c S
~ o
— T- "
p—
UJ m
(X! C
O °v
C T—
D x
Qt ^
rs C
OR
on i—
re x
O c
{2



_g

CD
U


2
1

^







^_
r..
CN


CN




m
CN







g
I"
"ob^
il


en


LO
CN
LO







LO

°1
i—

CN
CN
LO
,— "

^
Tf
ro"


u
c 5
— %
00 >
IS
(Middle Atlantic/Up
(NY,NJ,PA,DE,MD,


R
CN


O







vO
tx.

^


O

T-^

LO
00

r\T







u
lR
k. ^
IS


CO
o
CN


LO
IX,







CO

CN.
CN

o-i
^
tx.
ro"

LO

LO"







!--
LLJ ^


tx,
oo
CN


8
LO







tx.
LO

T—

00
CN
Ix,^
T— "

LO
CO

CN"







S?
C 00
CD -5
U <
»t
(ft —1
ro <
LU C>


CN


LO
CO







IX,
CO
CTi


Oi
O")
00


CN
OO
>— "





D
00
cf
Z
uT
Z
SO
g5.
u z


i
,-^

LO
ro







OO
^•O
o^


ro
ro



O

r-"







"re Z?
^ i
CD ii
U o
00 <"
CD f¥


i
ro"

O
O
1






^
O
^f



fO



8
LO




^
>
D
Z
Z
|§
IS


S
LO
1

LO
CN







O
IX.



IX,
LO



tx
T—








I
^U <"


o


§
CN
— J1






OS

"I
CJ^


o
^
CT>

in
00

oo"
^"






1
'E
D
1
103

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY


o

rri
OJ
u
to
c
_o
-a
0)
c
o
01
E
LU
=
5
1
LU
'c
0)
y
QJ
D.
O
LO
~TO
_O
-D

^O

as
C
Ql
(J
to
<
u
01
Wl f- ^-^
a> c «
C 0) a
QJ ^ —
|of
cu to t
LU C S
j" 0 °2
U ^'~

w
O "B
U .2
11
C C/)
01 J^
p x
LU u,
IN C
O iS
to C


"° o
> °
E -
,S x
rM C
to t

O

^ ,
to t-
— x
^ 01
t


c
o

0) "o?
01 £5
J





LO
en




UO
rxi


0









CN
r-i
T"





, 	
^'
r-i






--v
^
I
— '• ^





LO
LO




o
q_


CO
CO








tx
o
"




LO
^j1
o^
1 —

c >"
— %
^ <

fn U
S.Q
DQ
?2
•— LU
sq
5 rf
< D-
II
5S





LO
LO
a>




o
o
CO


0
0








ro
CO





O
Ix
•^
'~







_y
I R





LO
LO




0
o
•-

a^
o>








o>
^




CO
ro
(N^
CN"







ll
J--
tt —
as —i
LU ~





LO
LO




0
CO
LO


01
LO








CO
o





(X
LO
>— __
' —







ll
§£
u ^
to ^
tt _r
CO <
LU O-





LO
LO
&




LO
LO


CO
•*








LO





tx
CO
01




5
to

d
z
LU"
si
u z





LO
LO




o



o
't
LO








CO
CN





CO
o
o>








aj sx
uq
to 
D

5
-z.
Z
2
cQ
2 O
11





LO
LO




ro
"—


LO









^





O
(x









|
'0 <"





S




o
s


tN
*.
^







|
10




0*1
^
u^
oC







13
-c
0)
<**
"E
D
I
                   ^ o
                                               104

-------
                                                                   CHAPTERS:  IMPLEMENTATION ISSUES
the CAAA Scenario (i.e., $240 per ton SO2  re-
moved).  The marginal cost of reduction is also
much greater than in the CAAA  scenario (i.e.,
about $1,225 per  ton SO2 removed versus $500
per ton SO2 removed).

Costs and unit level SO2  emissions for the 50 per-
cent electric  utility reduction scenario were fore-
cast  using  EPA retrofit scrubber cost assumptions
and unit  level SO2 emissions forecast in the CAAA
Scenario. It was assumed that  SO2 emissions  re-
duction would  be achieved by unscrubbed coal-
fired units  not  already forecast to use low sulfur
coal  in the CAAA  scenario (i.e., unscrubbed units
forecast to have an SO2 emissions rate greater than
0.8 Ibs SO2 per million Btu). All such units would
not necessarily  scrub (e.g., units may fuel switch,
conserve energy,  or use  other  technologies), but
this analysis provides an estimate of the costs units
could face.  Using EPA scrubber cost assumptions
used in the 1993 Base Case Analysis,  costs were
estimated for achieving the 5.0 million  ton SO2
emissions reduction.

5.4.3  50 Percent Utility and Industrial SO2
       Reduction Scenario
Under this scenario,  the 50 percent utility SO2
emissions reduction is no different than described
above. However,  it is supplemented by a 50 per-
cent  reduction from industrial sources. As a result,
costs  are estimated assuming  electric  utility and
industrial sources achieve a total reduction in SO2
emissions of about 7 million tons in 2010 relative
to the CAAA Scenario  (Exhibit  68).  Under the
CAAA Scenario, utility and industrial sources are
estimated to emit about 9.5 and 4.0 million tons of
SO2 in 2010 respectively. Commercial/institutional
sources are expected to emit an additional 0.2 mil-
lion  tons. The  cost of achieving the 50  percent
electric utility and  industrial source SO2 emissions
reduction is  estimated to be  about $6.5  billion
annually. This  corresponds to an  average SO2
emissions reduction  cost  of about  $926 per ton
SO2  removed. As  described in the previous sec-
tion, the average  cost of reductions  for  electric
utilities is about $955 per ton SO2  removed, and
the average cost for industrial sources is estimated
to be about $850 per ton SO2 removed.

5.4.4  Geographically Targeted Reductions
       Scenario
Costs were calculated for both the contiguous and
noncontiguous  geographically targeted  reduction
scenarios corresponding  to the deposition levels
achieved by the 50 percent nationwide SO2 utility
reduction scenario for all three sensitive receptor
regions and for all three regions together. These
scenarios were constructed by sequentially remov-
ing 95 percent of utility SO2 emissions (remaining
after implementation of the Title IV)  from subre-
gions (either contiguously or in order  of contribu-
tion to deposition) until the deposition loads were
achieved.  Ninety-five percent SO2 emissions re-
moval was necessary given the  smaller number of
sources from which to  draw emissions reductions.
Costs were estimated by applying the EPA retrofit
scrubbing  cost assumptions to achieve 95 percent
SO2 removal from utility boilers identified in each
subregion. Exhibit 69a provides  costs for achieving
the target deposition loads for each receptor region
individually and  for all three regions simultane-
ously for  the  contiguous  approach.  Exhibit 69b
provides the  same information for the non-con-
tiguous case.

The total tons removed, about 4.6 million  tons, the
annual costs, about $4.6 billion, and the cost-ef-
fectiveness, about $1,000 per ton removed, are es-
sentially identical for the contiguous and non-con-
tiguous cases.  Thus,   no  increase in efficiency
would be  gained by choosing the more  complex
non-contiguous approach over sequential  contigu-
ous geographical targeting.

Costs were not estimated for the  geographically
targeted utility and  industrial  scenario  because
specific cost functions for SO2 removal from indi-
vidual industrial sources were not available. De-
veloping cost  functions for  individual industrial
sources would require  an extensive cost develop-
ment effort beyond this scoping study.
5.4.5  NOX Reductions
       and Industrial
-50 Percent Utility
EPA is  currently developing regulations  for the
control of NOX emissions from electric utilities af-
fected under Title IV of the CAAA.  Regulations for
Group 1  boilers in  Phase I  and  Phase  II  were
promulgated  on April  13, 1995. Regulations  for
Group 2 boilers are under development and the
costs  and  emissions  reductions expected  from
these regulations were not available for this report.
Preliminary information on the cost of controlling
NOX emissions from various types of electric utility
boilers is  available, however, from a recent EPA
report.122 These costs vary significantly depending
}22 Alternative Control  Techniques  Document—NOX
   Emissions from Utility Boilers, March 1994,  U.S.
   EPA, Office of Air Quality Planning and Standards,
   EPA-453/R-94-023.
                                               105

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                o

                I
<  p
U  u

il

11
I  s
LU  LU
I S
i 1
S z
              Q 3

              15
              g 5

              u y
             11
             00 Q
             0 <
              E
                >

                o
                5


                LU
                U

_g
L.
re
§
V

_g
fj
T3
0)
ai

~o3
. -_
-*-1
I/)
"O
—
T3
C
re
£
^~
V-
£
0)
u
CU
Q.
O
LT>
C

, —
TJ



*
.g
re
0)
CO
<
^^
<
U
is E -^
c $ *
.£ oi "5
D CO Tf
tfc _ OS
UJ g 0^
o ^- "c
u ^ ™


^ .
t/) ^-^
O c
U .2

"re ^^
§2
£ V)'
•^


c o
o o
« q
E x
LU ^n
tN C
O ^
CO t


IS o

o q
E ^
QJ X
^~ C/)
IN C
CO t

o
og O
o q
CO i—
"re x
*- in
0 c
i— o
b


_o
CU "^>
& a»
in ™
C "^
OJ
U





LT)
LO
Oi








LO
IX
^








LO
*~








CN
00





00
^o







i"
"i ^
LLJ ii-
^
S I-"
"7 *>-^


o
ro
Oi








0

•o
•^







'-0
'.O

'^







o
LT)
T~~




cr,
t~^
CN"




c 5
Ci^ ^*
Atlantic/Upper
,PA,DE,MD,DC
0) •=•
-*~
13 >
ss


^J.
ro
Oi








O
00
CT*








IX
00








tN
LO
q_
^~




r—
O
i—"






U
Ig
i— "^T
0 ( "1
JsJ


IX



r.
(N








O
IX
vO








IX
vD








IX
(N
IX





(N
O

r-"





Q
CO_^
-z.
LU"
•z.
SQ
§l
~7 -~
4_l «^
en Jb


O\
00








o
o
IX








S
00








ro
O
CO





g
O
^-"






i ^
CO <
^ — 1



o
LO








o
T—
^








in
LO
CO








^«
^





IX
^"





?
(-"
D
1L
Z.
il
2 c_/
C C^J
12


CTs
(N








ro
^









LT!









^
"~





O
IX








,1
U ?N
.•— Q
'u <


o









o
o

^-







rx
^»_
^O







fs^
s
rC




*
rx
ro"
T~





*
"re
55
"g
x
^J
,^
(Q
^
                                                                         o
                                                                         TO C

                                                                         ~
                                                                         rs
                                                                         u^ 03


                                                                         S.E
                                                                         -C 03
                                                                         *-* 4—'

                                                                         LO E
                                                                         c o
                                                                         O J=

                                                                         « c

                                                                         E .2
                                                           .
                                                           r= CD


                                                           3 ^
                                                           u> O
                                                           C CO
                                                           O —
                                                           *-' ro
                                                           -C 'C


                                                           « 1

                                                           3-0
                                                           o .E
                                                                         LO U
                                                                         in tt
                                                                         0) O
                                                                         -a c

                                                                         JE o
                                                                         u TJ
                                           106

-------
                                                                   CHAPTER 5:  IMPLEMENTATION ISSUES
              EXHIBIT 69A: ANNUAL COSTS OF GEOGRAPHICALLY TARGETED REDUCTIONS EQUIVALENT TO
                 NATIONWIDE 50% UTILITY SO2 REDUCTION (CONTIGUOUS RADM SUBREGIONS)

Emissions Reduction (tons)
Cost ($ million)
Cost-Effectiveness ($/ton)
Sensitive Region
Adirondacks
3,018,000
3,131
1,037
Mid- Appalachi-
ans
1,952,000
2,127
1,089
Southern
Blue Ridge
1,508,000
1,610
1,068
All Three
Receptor Regions
4,526,000
4,741
1,048
      EXHIBIT 69s: ANNUAL COSTS OF GEOGRAPHICALLY TARGETED REDUCTIONS EQUIVALENT TO NATIONWIDE 50%
       UTILITY SO2 REDUCTION: MAJOR RADM SUBRECIONS CONTRIBUTING TO DEPOSITION (NOT CONTIGUOUS)

Emissions Reduction (tons)
Cost ($ million)
Cost- Effectiveness ($/ton)
Sensitive Region
Adirondacks
3,160,000
3,151
997
Mid- Appalachi-
ans
2,079,000
2,193
1,054
Southern
Blue Ridge
2,081,000
1,952
938
All Three
Receptor Regions
4,658,000
4,523
971
on the type of technology applied,  NOX control
efficiency, and boiler specific parameters. Some of
the results presented in that  report are described
below.

Title IV of the CAAA requires EPA to set emissions
limits for Group 1 boilers (i.e., dry  bottom wall-
fired and tangentially fired boilers) based on  the
application of low-NOx burners (LNB) at affected
electric utility  units.  In the RIA, which  covers
Group 1  boilers, EPA  estimated that  NOX  emis-
sions would  decrease  by about 1.5 million tons
annually at an average cost of about $200 per  ton
NOX removed.123 The  RIA considers a variety of
NOX control technologies.  Applied to pre-New
Source Performance Standard  (NSPS)  coal-fired
electric utility boilers,  these  technologies are  ca-
pable of  achieving NOX emissions reductions of
about 10 to 60 percent.  For a boiler operating in
baseload, these technologies are estimated to cost
from about $100 to $1,000 per ton NOX removed.

To reduce electric utility NOX emissions by more
than required by Title IV, it would be necessary to
apply technologies with NOX removal  efficiencies
and  costs greater than  LNB  technology such  as
123 Regulatory Impact Analysis of NOX Regulations,
   February 1994, U.S. EPA Office of Atmospheric and
   Indoor Air Programs, Acid Rain Division.
selective catalytic  reduction  (SCR)  or  selective
non-catalytic reduction (SNCR). SCR can achieve
NOX removal  efficiencies ranging  from 75 to 85
percent.  In combination  with LNB technologies,
SCR can reach removal  efficiencies of 85 to 95
percent.  EPA  estimates that  a stand  alone  SCR
would  cost from $810 to $2,490 per ton NOX re-
moved at a  coal-fired unit operating in baseload.
In combination with LNB technology,  SCR appli-
cation  could cost about $1,300 to  $2,490 per ton
NOX removed. These costs would  be expected  to
decline with the wide-scale application of  SCR
throughout  the electric utility industry, based on
economies of scale.

As with utility boilers, a wide variety of NOX  con-
trol technologies are applicable to industrial  boil-
ers. These  include LNB,  SNCR,  and  SCR.  It  is
usually more  cost-effective to  apply these  tech-
nologies to electric utility boilers than to industrial
boilers  because  electric  utility boilers generally
provide more suitable operating  conditions and
economies of scale.

5.4.6  Summary of Economic Impacts
Exhibit 70 summarizes the total costs and costs per
ton of SO2 removed for each reduction  strategy for
which   costs  were  developed.  The  additional
50 percent   nationwide  SO2  utility   reduction
                                               107

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                EXHIBIT 70. SUMMARY OF COSTS OF VARIOUS EMISSIONS REDUCTIONS SCENARIOS
Scenario
CAAA baseline
50% utilitySO2 removal
Targeted utility SO2 removal
(contiguous)
Targeted utility SO2 removal
(not contiguous)
50% utility and industry SO2
removal
SO2 Removed
(Tons x 1000)
9,166
5,047
4,526
4,658
7,047
Annual Cost
($ Billion)
2.2
4.8
4.7
4.5
6.5
Cost- Effectiveness
($/Ton SO2 Rem.
in 1994 dollars)
240
955
1,048
971
926
scenario is a factor of four less cost-effective than
the CAAA baseline. This is not surprising since the
allowance  trading  program  was  designed  to
achieve maximum cost-effectiveness,  and,  thus,
utility  sulfur  dioxide  reductions  beyond  that
required  by the  Act would necessarily be  less
economically   attractive.   Additional  emissions
reductions  would likely  impact  the emissions
trading program  and may limit the  compliance
flexibility  inherent in the  current  program. From
Exhibit 70 it is apparent that the difference in costs
between   the   nationwide  and   geographically
targeted SO2  emissions  reductions  strategies to
achieve the same level of deposition  are not sig-
nificant. Therefore, there does not  appear to be a
significant  cost  advantage  to adopting a geo-
graphically targeted  approach to  achieving the
deposition  levels attained by  the  nationwide 50
percent SO2 utility reduction scenario. It may be
appropriate to assume that some level of  cost
savings associated with an  unrestricted  national
trading program (as assessed for implementation of
the Acid  Rain  Program under Title IV) could  also
result in reduced costs of  compliance with broad
emission  reductions  beyond the current program.
This could widen the  cost difference  between a
geographically targeted and  national  emissions
reduction  strategies.  Interestingly,  the  50 percent
utility and industrial reduction scenario  is about
equal in cost-effectiveness to the 50 percent utility
reduction  scenario. This indicates that emissions
reductions from major  industrial sources would be
as cost-effective as additional utility reductions.

The costs presented in this  economic analysis were
based  on  conventional SO2  scrubbing  and are
scoping in nature. These costs should be viewed as
conservative (i.e., over-stated)  in  that within the
time frame in which  acid deposition standards
could be  developed, newer innovative technolo-
gies  may  become available which may be  less
costly  than  conventional  scrubbing,  including
clean coal technologies,  repowering with natural
gas, or increased use of pollution prevention tech-
nologies such as renewal energy and conservation.

If further emissions reductions  were mandated by
Congress in order to implement an acid deposition
standard, there may  be a greater impetus for the
commercial deployment  of clean coal  technolo-
gies in the U.S. market place as well  as other re-
powering  or pollution  prevention  technologies.
Furthermore,  development of  an  acid deposition
standard  or standards would  require additional
analysis of costs and benefits to determine the
level of incremental  benefits in a range  of effect
areas  as  compared to the  deposition reductions
necessary to meet a  range of standard levels and
the costs associated with these levels.

5.5  MONITORING PROGRAM EFFECTIVENESS
To assess the adequacy  and  effectiveness of an
acid deposition standard or standards, a process
for measuring the impact of the program must be
created.   Attainment of ambient air quality stan-
dards is determined  by measuring ambient levels
of criteria pollutants at sites which can reasonably
be expected to demonstrate violations of the stan-
dard.   An acid deposition  standard or standards
could  be  structured similarly.    Two types  of
monitoring programs would be needed: a deposi-
tion monitoring program  to measure the effect on
deposition and to assess progress towards the goal,
and  an ecosystem monitoring program to deter-
mine the  environmental impacts of the standards.
Depending on the intensity of  the monitoring and
the geographic coverage, the costs of such moni-
toring  may be in  the range of  $10-20 million per
year.   (Some of the monitoring already exists as a
                                               108

-------
                                                                   CHAPTER 5:  IMPLEMENTATION ISSUES
result of programs begun  under NAPAP in the
1980s and in response to the CAAA).

To implement an acid deposition standard, depo-
sition monitoring  would have to match  the  re-
quirements of the standard.  The principal species
of interest for deposition monitoring are wet and
dry sulfate and nitrate.  Wet and dry ammonia is a
significant contributor to total nitrogen deposition;
however, few data are available on deposition of
this pollutant. In the 1980s a number of federal,
state, and private  sector monitoring efforts  were
initiated to gather information on acidic deposition
and  trends.    The  monitoring  networks  were
designed  to  collect  information   on  acidic
deposition and ozone.

Currently, the largest  continuing wet  deposition
monitoring network  is the National  Atmospheric
Deposition   Program   (NADP)/National  Trends
Network which  operates  about 200 sites in the
United States. Data from this network, which was
initiated in 1978,  have been used extensively by
NAPAP for assessment and to produce maps of wet
sulfate and nitrate. Measurement of dry deposition
is resource intensive, and thus the number of dry
deposition sites is  limited.  The National Oceanic
and Atmospheric Administration (NOAA) operates
a small network of intensive dry deposition sites
and a larger number of indirect measurement sites.
EPA operates a 50-station National Dry Deposition
Network (NDDN).  Both networks use models to
infer dry deposition  from  meteorological  and  at-
mospheric measurements.

EPA  established  the  Clean  Air  Status  and Trends
Network (CASTNET) which is a multi-agency ap-
proach to monitoring and developing assessment
tools.  Participants include federal and state agen-
cies  and universities.   Data from the NADP, the
NDDN, and  the NOAA wet deposition network
provides  weekly sulfur and  nitrogen  deposition
data  on  the coarse spatial  scale of  deposition
across the country.

Deposition monitoring as part of the implementa-
tion  of an acid  deposition standard  or standards
could become part of CASTNET.  The purposes of
a monitoring program to accommodate monitoring
for an acid deposition  standard would need to be
established and data quality objectives would have
to be  developed.    The data  quality  objectives
would  depend  on  whether  deposition values
would  be used in a regulatory sense to determine
specific violations  or to assess trends and progress
toward  environmental  goals.   Program require-
ments  would  need to  identify  pollutants  to be
monitored, standard procedures for determining
wet and dry deposition, spatial resolution, tempo-
ral requirements,  and approaches to  establish ref-
erence standards.

Implementation of  an  acid  deposition  standard
would require effects  monitoring  (i.e.,  surface
water) to  determine  the  effectiveness  of  the
standard.   Key aquatic monitoring  indicators for
surface water chemistry include changes in ANC,
pH,  inorganic monomeric aluminum,  and acid-
sensitive  aquatic  species,  especially  fish. Several
ongoing  monitoring  and  assessment  programs
provide    useful    monitoring    approaches.
Continuation   of  these programs would  likely
address many important  monitoring   concerns.
Two examples of these program are highlighted
here.   First,  the Long Term Monitoring  (LTM)
Project collects data from 80 lakes  and streams
located in mostly  acid sensitive portions of Maine,
Vermont,  New   York,  Pennsylvania,  Michigan,
Wisconsin, and Colorado.  This project provides
the best  available data to date  on  patterns  and
trends  in surface  water acidification  at individual
study sites sensitive to acidic deposition.

Assessing continuing benefits from the CAAA on
aquatic ecosystems  is the goal of both the LTM
and the Temporally Integrated Monitoring of Eco-
systems (TIME) projects. The TIME project began
monitoring a  statistically representative sample of
lakes in the Northeast in 1991 and a similar selec-
tion of streams in the mid-Appalachians  in 1993.
Central to the objectives of this project is the abil-
ity to  detect  statistically  significant changes in
chronic acidification  trends  across  the  regional
populations of lakes and streams. These sites are
selected  to  represent  regional   subpopulations.
Continuation  of   effects monitoring  would be a
necessary tool to evaluate the environmental ef-
fectiveness  of  a standard  or  standards  to protect
sensitive ecosystems.

5.6  CONCLUSIONS
In this  chapter, the discussion has been focused on
implementation issues associated  with  an acid
deposition standard  or standards. Assuming that a
decision were made to reduce emissions of sulfur
dioxide and/or nitrogen oxides beyond the current
Clean  Air  Act to  address the  acidification of
surface  waters   and/or  the   multiple  effects
associated  these  pollutants,  it  is  feasible  to
implement  such an  approach. There are different
approaches that could be taken and various factors
                                               109

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
to  be  considered   (administrative  complexity,
resource demands on the government  and regu-
lated  industry,  costs,  interactions with  other pro-
grams).  Based  on the multiple effects of acidic
deposition and  its sulfur and nitrogen precursors, it
is   recommended   that   if   further  emissions
reductions  are  pursued,  they  be as  broad  as
possible; either a national approach or  a  regional
approach  that  incorporates a  large  area of the
country (e.g.,  east of the Mississippi River). Fur-
thermore, Title IV is  an administratively  efficient
way to achieve  emissions reductions, with the
basic  infrastructure already set up  under  Title IV
being well-suited to incorporate further sulfur di-
oxide and  nitrogen oxides  emissions reductions.
Further emissions reductions characterized in this
report  could lead to  costs that are  more  than
double  those  of the  current  acid rain control
program, but the timing of those reductions would
affect the costs and the benefits. Compliance  costs
could be significantly  lessened by the timing of
any  further  emissions  reductions.   Reductions
required  later rather than  earlier  may  cost  less
based on  cost-saving technologies demonstrated
through  clean   coal  technology   and  pollution
prevention efforts and based on the replacement of
existing sources  by new, lower emitting sources.
The  benefits would  likely  be  in multiple effects
areas.
                                                110

-------
                                         CHAPTER 6

                          INTEGRATION AND CONCLUSIONS
6.1  INTRODUCTION
Section 404 Appendix B of the Clean  Air Act di-
rects  EPA  to  assess the feasibility  and environ-
mental effectiveness of an acid deposition standard
or standards to protect sensitive ecosystems. This
chapter summarizes  and integrates the findings
from the previous chapters showing how scenarios
for potentially  reducing acidic  deposition  de-
scribed in Chapter  3  address environmental goals
defined in  Chapter  2. Chapter 4 described poten-
tial benefits to visibility, human health, material,
and  cultural  resources accompanying  additional
reductions  in  acidic deposition.  Chapter 5  ad-
dressed feasibility and effectiveness of implement-
ing a deposition standard.

Section 6.2 highlights the conclusions that are key
to establishing effective environmental  goals and
efficient deposition  control strategies. Section  6.3
then integrates these findings to show the potential
surface water  benefits from a selection of alterna-
tive emissions reduction scenarios, considering  a
range of times to watershed nitrogen saturation.
Section 6.4 highlights the merits of understanding
the inherent ecological processes  of sensitive re-
gions and the usefulness of developing resource-
specific goals  to  provide a  basis  upon which  to
measure program effectiveness.  Section 6.5 sum-
marizes the range of feasibility  and effectiveness
issues regarding developing and implementing an
acid deposition standard or standards.

6.2  DETERMINING ENVIRONMENTAL GOALS
Determining  environmental  goals  for  protecting
ecosystems from acidic deposition requires charac-
terizing potential  environmental  effects  and bene-
fits over a range of  sulfur and nitrogen deposition
loadings and then deciding on the  desired level of
protection  for  the  ecosystem.  Resource-specific
goals  can be  used  to determine what emissions
and deposition reductions would likely be needed.
A regionally specific acid deposition standard can
be used to achieve effective  and  efficient envi-
ronmental protection  of those resources and eco-
systems most sensitive to adverse effects and most
likely to benefit from acidic deposition control.
Establishing environmental goals requires selection
of appropriate ecological endpoint  criteria  and
indicator measures. Such measures  must provide
information to accurately judge how successfully
the key ecosystems and  resources of concern are
being protected and at what time benefits may be
anticipated (e.g., rate of recovery). The applicabil-
ity of these measures varies among regions and, in
some cases, among individual systems (e.g., water-
sheds,  lakes,  or streams). Although  the  analysis
presented in this chapter focuses on changes in
surface water quality reflected by two ANC meas-
ures within these waters,  other endpoints may be
equally or  more appropriate for protecting sensi-
tive  resources of local  interest, such as  individual
stands of red spruce forests or populations of listed
threatened   or endangered  species.  Potentially
useful endpoints can include the "most sensitive"
systems or species in a region  or some defined in-
dex of ecological structure. An index can also be
designed to address a  region  as a whole and re-
fined to address natural acidity issues.

Several major points and conclusions from the re-
search  and analyses  presented in  the  previous
chapters  are key to the process of integrating our
understanding  of  environmental   effects   and
source-receptor relationships  related to  determin-
ing appropriate environmental  goals and setting an
effective acid deposition standard or standards.
The  following points highlight the processes  and
relationships described  earlier in this report.

  *  CHEMICAL   EMISSIONS AND   ATMOSPHERIC
     PROCESSES AFFECTING  DEPOSITION ACIDITY:
     The principal acids in deposition are sulfu-
     ric (H2SO4) and nitric (HNO3) acids. Thus,
     emissions of compounds  containing  sulfur
     and nitrogen  have been the primary focus
     in  acidic  deposition  control   strategies.
     Volatile organic  compounds (VOCs) and
     their oxidation products are also important
     because they often control  reactions that
     produce the oxidizing species that lead  to
     formation of sulfuric and  nitric acids  in the
     atmosphere and therefore affect the loca-
     tion and form in which sulfur and nitrogen
     are  deposited. In  total,  production  of at-
                                                111

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
     mospheric acids is a complex process, in-
     volving  140  known reactions among 60
     chemical species, 40 of which are organic
     compounds.

  * NATURAL EMISSIONS SOURCES: Natural emis-
     sions  of acid  precursor species,  organic
     matter,  and  alkaline  materials  (dust)  are
     generated by vegetative matter, soil and
     saltwater microbes, geochemical  activity,
     lightning,  and natural  combustion (e.g.,
     forest fires). Emissions of SO2,  sulfates, and
     nitrogen  oxides are  significantly  smaller
     from  natural  than  from   anthropogenic
     sources  by a factor of ten  or  more on an
     annual basis.

  * ANTHROPOGENIC EMISSIONS SOURCES: Acidic
     deposition precursor species and  reactive
     atmospheric  chemicals are generated by
     energy  production,  industrial  processes,
     mobile sources, and waste disposal.  Cur-
     rent estimates indicate that of  the total an-
     thropogenic emissions of SO2,  electric util-
     ity fuel combustion contributes 70  percent,
     industrial  fuel  combustion contributes 14
     percent, and  the balance comes from other
     sources.  For   anthropogenic  emissions  of
     NOX,  both electric utility fuel combustion
     and highway vehicles each are estimated
     to contribute 32  percent,  industrial fuel
     combustion provides  15 percent,  and  off-
     highway vehicles produce 12 percent, with
     the balance  coming  from other  sources.
     The potential need to effectively and effi-
     ciently   further   reduce   anthropogenic
     source  emissions  and  deposition would
     likely focus primarily on limiting emissions
     from these major sources.

  * CAUSES OF ACIDIC  DEPOSITION  EFFECTS:  Ac-
     cumulating scientific evidence verifies that
     deposition of acid-forming sulfur and nitro-
     gen compounds,  both  independently and
     synergistically, can be significant causes of
     surface  water acidification  effects.  Al-
     though  past  research  and  control efforts
     have  primarily focused on the  control of
     sulfur emissions and deposition, recent re-
     search indicates  that  nitrogen  deposition
     often  may be an  equally  and  sometimes
     more important  cause  of some  surface
     water acidification effects. For example,
     considerable evidence indicates that nitro-
     gen deposition is generally a greater acidi-
     fication  concern  in  the  western  United
  States and that nitrogen deposition as well
  as sulfur  deposition can be  a  significant
  contributor  to  episodic  acidification  of
  surface waters in the East.

* WATERSHED  NITROGEN SATURATION:  There
  are limits to the amount of nitrogen  that
  can be sequestered (e.g., in organic matter)
  in watersheds.  As these systems approach
  saturation, nitrogen losses from watersheds
  will  increase leaching of nitrate. This can
  lead  to  acidification  of surface  waters.
  Times to  nitrogen  saturation  vary among
  regions due to differences in  temperature,
  moisture,  length of growing  season,  soil
  fertility,  forest  age,  history  of  nitrogen
  deposition, and other variables.  Significant
  variability and uncertainties remain  in de-
  termining the time to  nitrogen  saturation
  for specific  watersheds across and  within
  regions. Nitrogen saturation  is a  potentially
  significant concern that  contributes  to the
  acidification  process, even  if total  satura-
  tion  never occurs.

* MOST  SENSITIVE REGIONS AT  RISK:  Based
  on the  NAPAP National  Surface  Water
  Survey, six  regions contain 95  percent of
  the  lakes and  84  percent  of the  stream
  reaches that  were chronically acidic [i.e.,
  having  an   acid  neutralization  capacity
  (ANC) of  0 ueq/l or less] due to inorganic
  ions, predominantly SO42~,  NO3~, and CK
  These areas include the southwest Adiron-
  dack Mountains in New York,  New Eng-
  land,  mid-Appalachian   Region, Atlantic
  Coastal Plain, northern Florida  Highlands,
  and  low-silica lakes in the upper Midwest.
  Compiled evidence indicates that  acidic
  deposition most likely  caused  significant
  acidification  of surface waters in the Adi-
  rondacks, the  Pocono/Catskill  subregion,
  mid-Appalachians,  eastern upper Midwest,
  the  New Jersey Pine Barrens,  and, to  a
  lesser extent, northeastern  Florida. These
  regions,   therefore,  require  the greatest
  consideration when determining the need
  for protection from future acidic deposition
  loadings.

* MOST SENSITIVE RESOURCES AT RISK: An  acid
  deposition  standard  or  standards  could
  provide adequate protection  for the most
  sensitive  resources at  greatest  risk.  The
  predominant natural resources that  appear
  to be both most sensitive to and  at greatest
                                                112

-------
                                                       CHAPTER 6: INTEGRATION AND CONCLUSIONS
potential  risk from  acidic deposition are
aquatic systems  and, potentially,  certain
high-elevation red spruce forests.

NATURALLY ACIDIC SURFACE WATERS:  Evalu-
ation of acidic deposition should include
the  realization  that many  regions  hold
naturally  acidic surface waters or  surface
waters with low ANC. For example, about
40 to 50  percent of the target population
surface waters in  the Adirondacks with
ANC of 50 ueq/l  or  less (i.e., sensitive) are
likely to persist even with  complete elimi-
nation of  anthropogenic acidic deposition.
Certain biota evolve to live in naturally
acidic systems.   Management  and  policy
decisions  should recognize the existence of
these  systems and  consider  protecting
populations  and  communities  that  have
naturally  evolved as part of  these ecosys-
tems.

ENVIRONMENTAL GOALS TO PROTECT SENSITIVE
AQUATIC RESOURCES: The  assessment end-
points of an environmental goal are formal
expressions  of the environmental value(s)
thresholds   for     harmful     conditions
(commonly  some ecological condition of
concern) that a standard would attempt to
prevent.   Assessment endpoints should be
biologically  relevant, operationally  defin-
able, scientifically predictable and measur-
able, and  sensitive to the pollutants of con-
cern.  From a policy perspective, assess-
ment endpoints  should  also be socially
relevant (mutually understood  and valued
by the public and decision makers).   The
biological  effects associated  with  acidic
deposition are minimized  as the level of
acidic deposition  is decreased and pH and
ANC levels  in sensitive  waters are kept
relatively  high.   For example,  based on
laboratory and field studies of sensitive
aquatic species, one general goal may be
to maintain surface  water pH  above  6.0.
Greatest protection of sensitive aquatic re-
sources occurs in surface waters where
ANC is  generally  maintained  above  50
ueq/l. Another way to state this goal  is that
there  would  be   no  deposition-driven
chronically acidic lakes or streams and no
episodically acidic lakes or streams. Other
assessment endpoints may also be appro-
priate.

EPISODIC ACIDIFICATION: Short term acute ef-
fects occur  when pulses  of  acidic waters
enter lakes  or streams with storm  runoff
  and snowmelt. The resulting  potentially
  acutely  toxic  changes  in  surface water
  chemistry often occur at the most  biologi-
  cally significant time of year. The projected
  number of systems at risk of episodic acidi-
  fication  is  substantially  higher than  the
  number of chronically acidic systems.

* CRITICAL  AND TARGET LOADS: A critical  load
  is a quantitative pollutant  loading below
  which no significant harmful effects occur
  to ecological processes; a critical load de-
  pends solely on inherent ecological proper-
  ties. A target load may incorporate social,
  policy, economic, and other considerations
  along with the scientific observations.  Pro-
  tection   approaches  emphasizing  critical
  and target  loads  are currently  being as-
  sessed and  in  certain  cases  utilized  in
  European countries and Canada. A critical
  and/or target load approach is conceptually
  similar to the deposition standard approach
  discussed in this report for determining the
  most appropriate level of protection.

  This report does not develop or set critical
  or target loads  or deposition standards. It
  does,  however, provide the   foundation
  upon which to  determine critical loads or
  deposition standards. The scientific uncer-
  tainty regarding watershed nitrogen  satura-
  tion makes determining critical or target
  loads or  a standard difficult at this time.

* MONITORING TO ASSESS EFFECTIVENESS  AND
  BENEFITS  OF CONTROLS: Although the analy-
  ses  presented  in  this  report focused on
  deposition standards appropriate for reduc-
  ing  regional proportions of  surface waters
  with ANC below 0  ueq/l and maintaining
  surface water ANC above 50 ueq/l, moni-
  toring to assess the  actual effectiveness of
  any emissions or deposition controls should
  assess not only the  potential  benefits  of
  controls  on surface water ANC, but also on
  other resources  of concern.  Such concerns
  include possible changes in the stand  con-
  dition within red spruce forests at potential
  risk,  visibility  impairment  in  National
  Parks, and  degradation of  materials  and
  cultural resources.

* REGIONALLY BASED  RESEARCH:  Outside the
  context of an  acid deposition standard, re-
  gionally  based  ecological  knowledge can
  be used to help guide efforts to improve or
  monitor  the ecological health  of sensitive
  areas. Other research efforts have focused
  on  the role of atmospheric nitrogen depo-
                                           113

-------
ACID DEPOSITION STANDARD FEASIBILITY STL DY
     sition  in producing nutrient enrichment of
     surface  waters  (e.g.  estuarine and  near
     coastal)  leading to nuisance algae blooms
     low oxygen conditions, and other problems
     associated  with  the  widely  recognized
     water  quality problem of eutrophication.
     For example, EPA has reported that from
     25  to 40  percent  of the total nitrogen
     loading   entering  the  Chesapeake   Bay
     comes  from  atmospheric  deposition.124
     Additional  scientific studies to determine
     the potential impact of acidic deposition
     on climate  change may also be relevant to
     fully characterize ecological effects

6.3  PROJECTED ENVIRONMENTAL CONSE-
     QUENCES OF ACIDIC DEPOSITION
     REDUCTION SCENARIOS
From the above  summary of findings, the surface
waters in the United States at greatest apparent
continuing  risk from acidic deposition extend from
the  Adirondacks south  along the  Appalachian
chain into northern Florida. Past research has pri-
marily focused on understanding acidic deposition
relationships within three representative case-study
regions along this area: lakes in New York's Adir-
ondack  Mountains  and  acid-sensitive  stream
reaches in  the mid-Appalachian Region (portions
of New  York, New Jersey, Pennsylvania,  Mary-
land, West Virginia, and  Virginia) and  in  the
Southern Blue Ridge Province (portions of North
Carolina, South  Carolina,  Tennessee, and Geor-
gia). These representative  regions  receive fairly
high levels  of acidic deposition, have the best his-
torical data, and  are best understood by scientists.

Potential benefits of additional sulfur and  nitrogen
deposition reductions to the three sensitive aquatic
resource regions were  projected using relation-
ships  defined through  the  Nitrogen  Bounding
Study (NBS) discussed in Chapter 2  and shown  in
Appendix B.  Specifically, the scenarios presented
in Exhibits 40 and 49 in Chapter 3 that provide for
the maximum  reduction in acidic deposition levels
by  the year 2010 represent approximately a  44
percent decrease in SO2 emissions from all sources
beyond those achieved by the CAAA and a  24
percent decrease in projected NOX emissions from
1990  levels.   These  deposition  reductions were
produced by  reducing  both  sulfur  and  nitrogen
emissions from  utility  and industrial  combustion
sources by  50 percent. The NBS projections were
then used to relate the resulting sulfur and nitrogen
124  U.S. Environmental Protection Agency. May 1994.
    Deposition of Air Pollutants to the Great Waters.
    First Report to Congress. EPA-453/R-93-055.
deposition levels  to probable proportions of sur-
face waters  in the two  ANC groupings.  Because
times to watershed  nitrogen saturation  are not
known for these three (or any other) regions, pro-
jections using four possible times for watersheds to
reach nitrogen saturation were modeled by NBS.

Exhibits 71-73 present actual and modeled pro-
portions of lakes and streams for two ANC group-
ings (ANC<0 ueq/l and ANC<50 ueq/l). Projected
proportions in both of these ANC categories are
shown for each of the three study regions under
each of four possible times to watershed  nitrogen
saturation for the following six scenarios:

  * Actual 1984 or 1985,

  * NBS projections  of surface waters  in the
     year 2040 if the 1990 CAAA had not been
     implemented,

  * NBS projections  of surface waters  in the
     year 2040 with full implementation  of the
     1990 CAAA.

  » NBS projections  of surface waters  in the
     year 2040 with  additional  reductions  in
     utility and industrial emissions of sulfur be-
     yond the CAAA (see Exhibit 40).

  * NBS projections  of surface waters  in the
     year 2040 with  additional  reductions  in
     utility and industrial emissions of  nitrogen
     beyond the CAAA (see Exhibit 48).

  » NBS projections  of surface waters  in the
     year 2040 with  additional  reductions  in
     utility and industrial emissions of both sul-
     fur and nitrogen beyond the CAAA (see  Ex-
     hibits 40 and 48).

The following points highlight the major  relation-
ships shown in the projections for the year 2040.
In reviewing these projections,  it is important to
remember that these projections are for  changes in
proportions  representing the NBS modeled  sub-
populations of the "most sensitive surface waters"
within each  of these regions; these proportions do
not apply to all the surface  waters or some  other
subset of surface waters in these  regions.

  * The modeling projects that  the 1990  CAAA
     would  reduce the proportions of surface
     waters  projected to be acidic (i.e.,  ANC<
     0 ueq/l) by 2040 in all  three regions, rela-
     tive to  conditions projected without  its
     implementation. This  is seen  by comparing
     the second and third vertical plotted bars
     shown for each region and across  each  as-
     sessed time  to watershed  nitrogen satura-
     tion.
                                               114

-------
                                         CHAPTER 6:  INTEGRATION AND CONCLUSIONS
    EXHIBIT 71. YEAR 2040 NBS PROJECTIONS FOR ADIRONDACK LAKES
   50 Yr     100Yr   250 Yr   Never     50 Yr    100Yr
                    Time to Watershed Nitrogen Saturation
250 Yr
Never
• 1 984 Base
I I w/o CAAA
• 1990 CAAA
• CAAA-additional S
[~~l CAAA-additional N
• CAAA-additional S+N
   EXHIBIT 72. YEAR 2040 NBS PROJECTIONS FOR MID-APPALACHIAN STREAMS
50 Yr    100Yr    250 Yr    Never     50 Yr    100Yr    250 Yr
                 Time to Watershed Nitrogen Saturation
        Never
                               115

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
           EXHIBIT 73. YEAR 2040 NBS PROJECTIONS FOR SOUTHERN BLUE RIDGE PROVINCE STREAMS
           30
        _c
        u
        ra
        o
        4)
        oc

        0)
        a
        01
        a.
20


15


10


 5
                Note: Maximum range is 30%
                           ANC<0
                         n
                                                     ANC<50 ueq/l
1
                 50 Yr     100Yr    250 Yr    Never     50 Yr     lOOYr    250 Yr
                                   Time to Watershed Nitrogen Saturation
                                                                        Never
• 1 985 Base
I I w/o CAAA
• 1 990 CAAA
WM CAAA-additional S
I I CAAA-additional N
• CAAA-additional S+N
  * The modeling projects that the 1990 CAAA
     would  reduce the  proportions of  stream
     reaches in the mid-Appalachians projected
     to become increasingly sensitive to poten-
     tial effects from episodic acidification (i.e.,
     ANC<50 ueq/l) by 2040. This benefit is
     projected to be lesser in magnitude for the
     lakes   in  the  Adirondacks  and   stream
     reaches in the Southern Blue Ridge.

  * The modeling  indicates that  sensitivities of
     target aquatic resources and  their potential
     responses  to changes in acidic deposition
     clearly differ among the modeled regions.

  * The modeling  indicates that  the benefits to
     sensitive surface waters from sulfur deposi-
     tion  reductions  mandated  by the 1990
     CAAA  may be lessened due to future in-
     creases in nitrogen  leaching  caused  by
     continuing nitrogen deposition and satura-
     tion of watersheds with deposited nitrogen.
     This  is shown by the projected increasing
     proportion  of  ANC<0 ueq/l and ANC<50
     ueq/l surface waters at shorter times to wa-
     tershed nitrogen saturation.
                                           » Uncertainty related to actual times to wa-
                                             tershed nitrogen saturation within these re-
                                             gions causes projections  of relationships
                                             between  deposition and ANC responses to
                                             range by factors of about two  or greater.
                                             Longer times to nitrogen saturation lead to
                                             fewer  projected  acidic   and  sensitive
                                             aquatic systems.
                                           * Significant  uncertainty continues to accom-
                                             pany the rate of watershed nitrogen satura-
                                             tion, contributing to uncertainty in project-
                                             ing the impact  of  additional  reductions in
                                             sulfur and nitrogen deposition.

                                           * Any  reductions  in  nitrogen   deposition
                                             would not  only reduce total acidic deposi-
                                             tion,  but would also increase   the actual
                                             times to watershed nitrogen saturation. This
                                             is similar  to sulfur deposition   reductions
                                             that are  now  likely  extending times for
                                             watershed sulfur saturation.

                                         6.4 SELECTING DEPOSITION GOALS
                                         Data and analyses presented in  this report indicate
                                         that the sulfur reductions in the Clean Air Act will
                                         provide significant  benefits  to sensitive ecosys-
                                                116

-------
                                                            CHAPTER 6: INTEGRATION AND CONCLUSIONS
terns. The data also suggest that adverse impact of
nitrogen deposition on  watersheds can be signifi-
cant.  Available research  indicates that nitrogen
plays a  significant  role  in episodic  (short-term
acutely toxic) acidification.  The  most recent re-
search recognizes the  importance of nitrogen in
long-term chronic acidification as well.  Reducing
atmospheric  loading rates  for nitrogen deposition
to watersheds would lead to three  different but
related potential  environmental  benefits.  First, for
many of the  most sensitive watersheds, such red-
uctions would markedly lessen the potential direct
influence of nitrogen in causing gradual long-term
chronic effects. Second, for many sensitive water-
sheds, such reductions would also reduce the po-
tential for harmful episodic short-term acute ef-
fects. Third,  for these watersheds, such reductions
would reduce the frequency and lengthen  times
necessary to  reach nitrogen  saturation.   With re-
gional times to  watershed  nitrogen saturation
lengthened, the potential for adverse effects pro-
duced by the remaining sulfur and nitrogen depo-
sition  together  would   significantly  decrease.
(Nitrogen saturation occurs when an affected wa-
tershed  accumulates  deposited  nitrogen   com-
pounds in excess of the  growth requirements for its
resident plant and animal populations. This excess
nitrogen then becomes available to acidify surface
and ground waters draining from the watershed.)

Environmental resources have ranges  of  sensitivi-
ties and risks to potential effects  caused by acidic
deposition.  Resources having equivalent  sensitivi-
ties have different risk potentials for harmful effects
that  depend on how much acidic deposition they
receive. For lakes and streams, biological effects of
acidic deposition  are minimized as the level of
deposition is  decreased  and pH and ANC levels in
sensitive waters are kept relatively high.  A general
goal and one used to guide numerous acid deposi-
tion  standard and critical load efforts in the United
States and in other countries may be to  maintain
pH  in sensitive  lakes  above 6.0  (i.e.,  the  goal
strives for no deposition-driven chronically acidic
lakes or streams).  To minimize effects to aquatic
resources  from   acute  episodic  acidification,
surface water ANC should be maintained above
50 ueq/l  (i.e.,  goal  strives  for   no,  or  reduced,
deposition-driven  episodic events).    This  report
identifies  several  scientifically  credible  resource
sensitivity  endpoints  but  no policy  decision  is
made to choose among  the range of options at this
time. Future efforts could  focus on characterizing
resource-specific   sensitivities   and   appropriate
environmental  indicators on additional  sensitive
areas  in  the  United States.     Further   effects
research,  particularly concerning the extent and
rate of nitrogen  impacts on watersheds and moni-
toring  of  appropriate  environmental  indicators
would facilitate this effort.

Aquatic modeling results presented in  this report
indicate that additional  reductions in sulfur and/or
nitrogen  would  reduce regional  proportions  of
chronically acidic surface  waters and proportions
of surface waters most sensitive to episodic effects.
The magnitude  of benefits varies by region.   Al-
though the model results  presented are the  best
now obtainable, considerable uncertainty  accom-
panies these projections. Despite these uncertain-
ties, the model results presented in this report indi-
cate,  as  consistent  with  theory,  that reducing
deposition loadings  for both  sulfur and  nitrogen
compounds would lessen  the  number of  surface
waters adversely affected  by deposition  of both
compounds within all three regions modeled.

Scientific  uncertainties  make  setting  an  acid
deposition standard  or  standards at a particular
deposition level difficult at this  time.  Even when
the uncertainties have been resolved or reduced,
setting a  single,  uniform national standard  may be
an inappropriate approach  in view  of the differing
sensitivities and risks associated with resources in
different  regions of the country.  Some have ar-
gued, however, that acidic  deposition goals (rather
than standards)  established through consideration
and analysis of resource sensitivity and  risk would
provide  useful information and guidance.   Such
goals would provide reference points to assess the
effectiveness of pollution control efforts and would
serve as  guides for environmental  policy  makers
until such time  as the scientific uncertainties that
inhibit setting  an acid deposition standard or stan-
dards are better understood.  Because such depo-
sition goals are intended only as reference points,
they should not be construed as standards within
the meaning of the CAA.

In developing deposition goals, it is important  to
note that the available scientific data indicate that
individual watersheds in certain regions of North
America  are progressing toward and, in a  few in-
stances,  have reached  nitrogen saturation.   Re-
maining  times to nitrogen saturation  for  water-
sheds,  however, can only be  roughly approxi-
mated.  This is true on both national and  regional
scales. Thus, establishing interim deposition goals
requires making expert  assumptions regarding not
only existing times to nitrogen  saturation for wa-
tersheds  over  regional scales, but  also the effect
                                                117

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
that reducing nitrogen deposition would have on
future times to nitrogen saturation.

Reducing both  sulfur  and  nitrogen  deposition
would lessen  proportions  of  surface waters ad-
versely affected  by acidification for the most ex-
tensively impacted region  studied,  the Adirond-
acks.  Reducing  nitrogen deposition is especially
critical, if the time to nitrogen saturation is as short
as some reasonably contend (e.g., 50-100 years or
less).   If, however, these times are much longer
(e.g.,  100-250 years or more),  as others also rea-
sonably  contend,  then  efforts  to  reduce sulfur
deposition would be  a  higher priority  than nitro-
gen deposition reduction.   The modeling results
presented in  this  report can provide rough  esti-
mates  of deposition  reductions that  would be
needed to achieve a range of environmental goals.
It is important to recognize that these estimates are
highly dependent  on  assumptions made with re-
spect  to time to  nitrogen saturation.  Furthermore,
no guidance  has been  provided by Congress re-
garding the degree of  protection desired by a stan-
dard or standards.  An  acid deposition  standard
can be designed to achieve a variety of environ-
mental goals such as (1) maintenance ol  specific
conditions  as observed  at  a  particular  point  in
time;  (2) return to pre-industrial conditions; or (3) a
level that balances effects, costs, and other societal
values.

Within the  limitations  of  significant uncertainty
associated with modeling projections and times to
watershed nitrogen saturation,  several  illustrateve
examples are  provided.  If the time to nitrogen
saturation in the Adirondacks is assumed to be 50
years, model  projections suggest that significant or
even  complete  elimination of  sulfur  deposition,
without  a   significant  reduction  in  nitrogen
deposition,  would   provide   at   most   a  few
percentage  points of  change in  the   affected
sensitive  lakes at year 2040.   To  maintain the
proportion of Adirondack lakes with ANC<0 ueq/l
at levels approximating those  found  during the
NSWS in 1984 (19  percent),  model  projections
indicate that nitrogen deposition may need to be
reduced   to   approximately   6.5  to   7  kg
N/hectare/year from the current level of 9.5 kg
N/hectare/yr.  Such a reduction likely would also
extend the projected time  for  nitrogen saturation
over the  Adirondacks.  If this reduction results in
an increase in the regional time to saturation  to
100  years, then the model  projects  that  sulfur
deposition  may  need  to  be reduced  to ap-
proximately 5.5 to 6 kg S/hectare/year.  Deposition
values of 6.5 to 7 kg N/hectare/year and 5.5 to 6
kg S/hectare/year in the Adirondacks correspond
approximately to a  50 percent reduction in  nitro-
gen emissions from utility,  industrial, and mobile
sources from  1990 levels  and an  additional 50
percent reduction  in  utility sulfur  emissions  be-
yond those mandated by the CAAA.

If instead,  the time to nitrogen  saturation  in  the
Adirondacks is assumed to be 250  years, deposi-
tion values of 9.5 kg N/hectare/year and 6.9 kg
S/hectare/year   (reflecting  1990 NOX  emissions
and the SO2 emissions  that are  projected in this
report  to  result  from  full implementation of  the
CAAA) are projected  to reduce the proportion of
Adirondack lakes with ANC<0 ueq/l below 1984
levels.   If all  anthropogenic NOX and SO2  emis-
sions  were eliminated, background  deposition
levels may approximate  4 kg N/hectare/year and 1
kg S/hectare/year, and the  model projects that in
this  case, only naturally  acidic   lakes would
remain.

Moving to more southerly regions, watersheds tend
to have longer times to  nitrogen  saturation due to
warmer temperatures and  differences  in forestry
practices.  For the Mid-Appalachians, if the time to
nitrogen saturation  is  100 years  or  longer,  reduc-
tion of deposition  levels to approximately 8.3 kg
N/hectare/year  and    8.1    kg   S/hectare/year
(reflecting approximately a 50 percent reduction in
nitrogen  emissions from  utility, industrial, and
mobile sources from 1990 levels  and an  additional
50 percent reduction in  utility sulfur emissions be-
yond those mandated by the CAAA) are projected
to reduce the  percent  of streams with  ANC<
50 ueq/l  below  the 27 percent observed  in  the
1985  NSWS.  If all  anthropogenic  NOX and SO2
emissions were eliminated,  the model projects that
the number of sensitive mid-Appalachian streams
could be reduced to less than 5 percent.

For the Southern Blue Ridge, if the time to nitrogen
saturation  is 250 years or longer,   reduction of
deposition  levels  to   approximately  7.9  kg
N/hectare/year  and   6.8   kg   S/hectare/year
(reflecting the 50 percent NOX and  SO2 emissions
reduction scenarios described  above) is projected
to reduce the  percent  of streams with  ANC<
50 ueq/l  to  slightly  more than the  6  percent
observed in the  1985  NSWS.  If  all  anthropogenic
NOX  and  SO2  emissions  were eliminated,  the
model  projects that  the  number of  sensitive
Southern Blue Ridge streams could  be reduced to
less  than  5  percent.   Approximate  ranges  of
environmental goals in each case study region are
subject to high levels of uncertainty.
                                                118

-------
                                                           CHAPTER 6:  INTEGRATION AND CONCLUSIONS
Actual environmental effects should continue to be
assessed through ongoing and new research efforts
to evaluate the reliability of these and other pro-
jections presented in the report.  This  research is
also needed to support development and  refine-
ment of models for projecting watershed responses
to the simultaneous deposition of sulfur and nitro-
gen  with  increased certainty,  and to determine
whether future  restrictions on deposition  of either
is warranted.

Several ongoing and potential regulatory efforts are
likely to  result  in significant reductions  in acidic
deposition levels. The Northeast Ozone Transport
Commission under Title I has efforts underway  to
reduce NOX emissions in order  to facilitate attain-
ment of the existing national ambient air quality
standards (NAAQS) for ozone.  EPA is also under-
taking reviews of the ozone and particulate matter
NAAQS.  Adoption of  revised ozone NAAQS and
a new fine particle NAAQS could likely result  in
further NOX and SO2 emissions reduction as part of
the associated control strategies.  In assessing pro-
gress toward the goals described above, it will be
important to integrate the nitrogen and sulfur red-
uctions that are  likely to be achieved  by these
other programs.

This report recognizes  that there is merit and im-
portance to understanding the inherent ecological
processes of sensitive regions and monitoring the
health and changes of those ecosystems. Develop-
ing resource-specific goals would provide a guide
to assessing  whether existing programs are effec-
tively protecting the environment.  Much  uncer-
tainty remains to be addressed if a formal standard
were to be pursued.

6.5  FEASIBILITY OF ESTABLISHING AND IM-
     PLEMENTING AN ACID DEPOSITION
     STANDARD
A  variety of factors could affect the implementa-
tion  of an acid deposition standard. The effect  of
these factors on implementation would depend on
the approach selected. To  be successful, an im-
plementation approach must be  clear and unambi-
guous, and provide certainty as to the responsibili-
ties of the regulated community,  EPA,  and states.
Two basic approaches were evaluated in this re-
port:

   * First, a regional  targeted approach  would
     involve  establishing deposition  standards
     for specific geographic regions, and require
     geographically   targeted  reductions   in
     emissions  of SO2  and/or NOX. EPA  would
     either set a standard or standards using ex-
     isting authority (if adequate) or would seek
     additional  Congressional  authority  and
     timetables. Source-specific limits would be
     determined using source-receptor models,
     and limits would be incorporated into State
     Implementation Plans (SIPs)  and enforced
     by states. The regional approach would be
     similar to the SIP program used to  imple-
     ment Title I of the Act regarding attainment
     of National Ambient  Air Quality Standards
     (NAAQS).

  * Second,  a  national,  emissions-based ap-
     proach which would  involve congressional
     direction to EPA to set an acid deposition
     standard  or standards  and  to determine
     emissions levels for SO2 and NOX  needed
     to  meet  the standards  within a specified
     time frame. An emissions cap and  allow-
     ance allocations would have to be  set for
     NOX and, as appropriate,  adjusted for SO2.
     The national approach would be similar to
     the current Acid Rain  Program.

Two emissions reductions scenarios were  devel-
oped to  compare national  and  targeted  reduc-
tions—a  nationwide 50 percent reduction beyond
the  CAAA of  SO2 emissions from utilities  and a
geographically targeted strategy  which   removed
95  percent  of utility  SO2  emissions from geo-
graphically constrained regions. Both approaches
were analyzed for emissions reduction efficiency
and  cost-effectiveness.  The following points high-
light  the  findings and conclusions  from these
analyses.

  * Geographic  targeting  can  be  used  to
     achieve target  loads  in  each receptor re-
     gion and  in all  three  regions  simultane-
     ously.  Delineation of  targeted emissions
     reduction areas  is made  complex, how-
     ever, by potentially large target areas, sig-
     nificant source-specific  emissions   reduc-
     tions requirements (i.e., 95 percent SO2 re-
     ductions), and impairment to the cost  sav-
     ings of the current Acid Rain Program due
     to regional  restrictions on  allowance trad-
     ing.

  * The total emissions reductions required to
     achieve the deposition loads in all three
     receptor regions simultaneously did  not dif-
     fer  significantly  between  the nationwide
     and targeted approaches. Costs of  control
     and  cost-effectiveness for the nationwide
     emissions reduction scenario and the geo-
                                               119

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
     graphically  targeted  scenario  were  ap-
     proximately  equal.  (To  assess the  ap-
     proaches,  each scenario was designed to
     achieve the same  environmental  goal.)
     There  is no environmental  or  economic
     gain  by geographically targeting source re-
     gions.
     Title IV will produce the largest emissions
     reductions  in the highest  source  regions
     (i.e., Ohio,  Indiana, West Virginia,  and
     western Pennsylvania).  Based on the mul-
     tiple effects  of acidic deposition  and its
     sulfur and nitrogen precursors, it is recom-
     mended that if further emissions reductions
     are pursued, they be as broad as possible;
     either  a national approach or  a  regional
     approach that incorporated a large area of
     the country  (e.g., east of  the  Mississippi
     River).
     The cost of additional emissions reductions
     addressed  in this report are, at a minimum,
     double the  cost of the current  Acid Rain
     Program. The benefits, although not quan-
     tified here, would be in multiple effects ar-
     eas such as  human  health, visibility,  and
     materials,  as well as aquatic systems.

     Finally, additional research as well  as con-
     tinued   and   enhanced   environmental
     monitoring)  (i.e.,  deposition  and effects
     monitoring) would be  necessary to evalu-
     ate the effectiveness of current emission
     reduction  efforts, to  determine  the appro-
     priate level of a standard or standards, and
     to  assess  the adequacy of that standard.
     Further research on the long-term effects of
     nitrogen on watersheds, including the ex-
     tent and rate  of  impact,  is  necessary to
     more fully understand watershed dynamics
     and the simultaneous impacts of sulfur and
     nitrogen.  Refinement of the  available at-
     mospheric and watershed models is critical
     for adequate nitrogen deposition and nitro-
     gen retention/cycling dynamics to provide
     regional-scale   reformation   across   the
     country.
This report concludes that establishing acid depo-
sition standards for sulfur and nitrogen  deposition
may at some point in  the future  be technically
feasible although appropriate deposition loads for
these acidifying chemicals cannot  now be defined
with  reasonable certainty.   Major  scientific  un-
knowns,  particularly regarding  watershed process
leading to nitrogen acidification  and  remaining
times to  watershed saturation with nitrogen, limit
the current  ability to recommend an appropriate
standard for any region. Furthermore,  policy de-
cisions regarding appropriate or desired goals for
protecting  sensitive aquatic  and terrestrial   re-
sources are needed to help guide the  Agency in
continued analyses and decisions  regarding possi-
ble  establishment of acid  deposition  standards.
This   includes   Congressional   guidance   and
continuing  efforts  to   address   social  science
uncertainties related to the public's desired level
of  protection of  standards  and  the  costs  and
benefits associated with meeting  such  standards.
Therefore, an acid deposition standard  is not  rec-
ommended at this time.
                                                120

-------
          APPENDIX A




SUMMARY OF SELECTED NAPAP REPORTS

-------
                                       APPENDIX A

                      SUMMARY OF SELECTED NAPAP REPORTS
INTRODUCTION
This appendix presents a synopsis of major conclu-
sions from  10 of the NAPAP State of Science and
Technology Reviews.  In a few of these summa-
ries, where knowing how the study was designed
and conducted can help the reader better appreci-
ate the study's key results, background information
on  the study's methods is presented.  Most infor-
mation summarized  in this appendix came from
NAPAP's Summary Report of the U.S. National
Acid Precipitation Assessment Program.1   Major
conclusions from  these NAPAP studies  that influ-
ence identifying key relationships between acidic
deposition  and  receptor effects are presented  in
the order that they were presented by the individ-
ual SOS/T report summaries.  Also, not all conclu-
sions from those studies are  presented in the fol-
lowing sections. In some case,  similar key conclu-
sions appear in  more than one report,  conse-
quently, they are  also repeated in more than one
of the case summaries below.

SOS/T REPORT 9: CURRENT STATUS OF
SURFACE WATER ACID-BASE CHEMISTRY
Phase  I of  the National  Surface Water  Survey
(NSWS) consisted of three  major surveys.  The
1984 Eastern Lake  Survey  (ELS)  sampled 1,592
lake of 4 ha and  larger that represented an esti-
mated population  of 17,953 lakes  in the Northeast,
Upper  Midwest, the  Southern  Blue Ridge region,
and Florida.  The 1985 Western Lake Survey (WLS)
sampled 719 lakes of 1 ha and larger that repre-
sented  an estimated population  of  10,393 lake
throughout the Sierra Nevada, Cascade, and Rocky
Mountain  ranges.    Finally, the National  Stream
Survey (NSS)  sampled 500 stream reaches in 1986
that represented an estimated  population of 56,000
stream  reaches (200,000 km) throughout much  of
the Mid-Atlantic Coastal Plain, mid-Appalachian,
Poconos/Catskills,  Interior Southeast,  and Florida
Regions.  A Phase II  lake survey (ELS-II) was con-
  Irving, P.M.  1991.  Acidic Deposition:  State of
  Science and Technology.  Summary Report of the
  National Acid  Precipitation Assessment Program.
  Office of the Director, National Acid Precipitation
  Assessment Program, Washington, DC.
ducted  in  1986 to  evaluate seasonal chemical
variability in Northeast Lakes.  Thus, seven major
subregions were sampled during the NSWS:

  * Northeast Subregion

  * Mid-Appalachian Subregion

  * Mid-Atlantic Coastal Plain Subregion

  * Interior Southeast Subregion

  * Florida Subregion

  » Upper Midwest Subregion (northern Michi-
     gan, Wisconsin, and Minnesota)

  * West Subregion

Several key findings from the NSWS include:

  1.  An estimated 4.2% (1,181) of the  National
     Surface Water Survey (NSWS) lakes were
     acidic, defined as having ANC<0 peq/l;
     nearly all were in the East.   These  lakes
     had pH levels in the range 5.0 to 5.5.

  2.  Acid  lakes tend to be smaller than non-
     acidic lakes. In the East acidic lakes aver-
     age 12  ha versus 17  ha  for non-acidic
     lakes.

  3.  Nearly 20% of the NSWS  lakes  had  an
     ANC of 50 ueo/l or less.

  4.  About 3.0% of the NSWS lakes had total
     inorganic  monomeric  aluminum (Al,m)
     concentrations of less than 50  ug/l.

  5.  No sampled lakes were acidic in  the Inte-
     rior Southeast or Minnesota, and only one,
     a  geothermal spring, was  acidic in the
     West. These results lead to the conclusion
     that there are virtually no acidic lakes in
     these  subregions, within the subpopulation
     of surface waters sampled.

  6.  Based on total stream length with the NSS
     target population, 2.7% (5,506 km) were
     acidic (ANC<0 ueq/l)  and  12%  (23,595
     km) had ANC<50 ueq/l, excluding reaches
     affected by acid mine drainage.
                                              A-1

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
  7.   For  8%  (16,780  km) of the  NSS  stream
      length, the pH was 5.5 of less and for 1 8%
      (35,771 km) the pH was 6.0 or less.

  8.   A greater percentage of streams reaches
      were acidic in  their upstream  reaches
      (6.1%)  compared to  their downstream
      reaches (2.3%)

  9.   Most of  the acidic stream reaches in the
      NSWS target population occurred in the
      mid-Appalachian     and    Mid-Atlantic
      Coastal Plain regions, with smaller length
      totals  in  the  Poconos/Catski 1 1  subregion
      and  Florida.

  10.  Acidic and low pH  streams  in  the mid-
      Appalachian region  and Poconos/Catski 1 1
      subregions were  restricted to watersheds
      smaller than 30 km^ and were  generally
      found  at  elevations  higher than  300  m.
      These were clearwater systems with a me-
      dian DOC of 1.1 mg/l and Al,m of 202
Based on results from this survey six "high-interest
regions" were identified that contained most of the
acidic lakes and streams identified in the NSWS.
While the  combined  high-interest lakes  popula-
tions included only 26% of the  all  NSWS lakes,
they included  95% of all  inorganic acidic lakes.
Similarly, 37% of all NSS  upstream  reaches were
in the high-interest population, but 84% of all in-
organic  acidic NSS upstream reach ends were in-
cluded.  The six areas targeted were

  » SOUTHWEST   ADIRONDACK    MOUNTAINS:
     Within this  subpopulation,  38%  of the
     lakes were acidic (ANC<0 ueq/l) and 51%
     had closed-system pH of 5.5 or  less. These
     acidic lakes  are  typically  rapidly  flushed
     drainage lakes in which SO42- is the domi-
     nant   anion   and  DOC   concentrations
     ranged 3  to 4  mg/l.   Many  had high
     concentrations of inorganic monomeric Al,
     with  36% having Alim greater than  100
     NEW ENGLAND: Within  this subpopulation
     4.7% of the NSWS lakes had ANC<0 ueq/l
     and 6.8% had pH<5.5.   In  the Seaboard
     Lowlands area 7.7% of the NSWS lakes
     were acidic (ANC<0 ueq/l).  The majority
     of  the  acidic  lakes in  this region were
     dominated by  inorganic acids, but about
     one-fifth,  mostly in the Seaboard Lowlands,
     were dominated by organic acids.
FORESTED MID-ATLANTIC HIGHLANDS (A.K.A.,
MID-APPALACHIAN  REGION):  Within   the
stream populations, 11.5% were acidic and
16.7% had closed-system pH<5.5.  Among
lakes of this subpopulation,  10% had ANC
<0  ueq/l  and  9%  had  pH<5.5.  Acidic
surface waters of this subpopulation  typi-
cally had  low DOC (stream mean of 1.5
mg/l,  lake mean of 2.6 mg/l) and  high
SO42~  concentrations (stream mean of 148
ug/l, lake  mean of  122 ug/l).   All acidic
lakes and streams were dominated by inor-
ganic acids, with SO42~  being the dominant
anion with relatively low NO3- concentra-
tions (mean < 10 ueq/l). Acid streams had
the highest concentrations  of inorganic
monomeric Al  of all  high  interest areas,
with a mean Alim of 202 ug/l. Acid lakes
had a mean Alim of 77 ug/l.

ATLANTIC COASTAL PLAIN: At their upstream
ends, 14% of the NSS stream reaches were
acidic  (ANC<0  ueq/l) and 17% had pH of
5.5 or less. Both mineral and organic acids
provided  important contributions  in  the
acid-base  status in these streams.   Among
the acidic streams,  65%  were dominated
by  organic  acids,  while  32%  were
dominated by inorganic acids. Most acidic
streams in the New Jersey Pine  Barrens
were dominated by inorganic acids,  but
with considerable  influence by  organic
acids.   In the  rest of the region,  NSS
streams were dominated by organic acids.
Twenty acidic  lakes  identified on Cape
Code were dominated by chloride  and had
salt water SO42' concentrations of 110 to
175 ueq/l.

NORTHERN FLORIDA HIGHLANDS: This subre-
gion includes  the  Central  Lake   District
(Trail Ridge) north of 29°N latitude and the
Panhandle. For this subregion, 63% of the
lakes   were acidic  (ANC<0  ueq/l)  and
52.6% had closed system pH of 5.5 or less.
Most   (90%)  of the  acidic  lakes were
seepage lakes.  Of  the acidic  lakes,  80%
were dominated by inorganic  acids,  with
median SO42-  concentrations of 83 ueq/l,
median DOC of 2.4 mg/l, and mean Alim
concentrations of 39 ug/l.   Acid streams
had low ionic concentrations (mean  base
cation  of 21 ug/l, SO4* of 16 ueq/l). Three-
quarters   of   the   acid   streams  were
dominated by inorganic acids, with DOC
of less  that 2 mg/l.
                                              A-2

-------
                                                            APPENDIX A: SUMMARY OF NAPAP REPORTS
  * LOW-SILICA  LAKES IN  THE EASTERN  UPPER
     MIDWEST: This subregion  includes lakes in
     northeastern  Wisconsin  and  the  Upper
     Peninsula   of  Michigan  having  silica
     concentrations of less than  1  mg/l.   Of
     these  lakes, 15.9% were acidic  (ANC<0
     ueq/l) and 19.3% had  closed-system pH of
     5.5  or less.   Of these acidic  lakes, 80%
     were  seepage   lakes,  with  low  silica
     concentrations reflecting  the lack of well-
     buffered ground  water inputs.   Inorganics
     dominated  the  acidity in 87% of these
     lakes.  SO42~ was generally  the dominant
     anion,  with  a mean concentration  of  69
     ug/l.  DOC concentrations were relatively
     high  (mean   of  3.9   mg/l),  indicating
     substantial influences by organic acids.  Al
     concentrations were low.

SOS/T REPORT 10: WATERSHED  AND LAKE
PROCESSES AFFECTING SURFACE WATER ACID-
BASE CHEMISTRY
  1.  Atmospheric deposition is often an impor-
     tant, yet highly uncertain,  component of
     the acid-base budget  in many watersheds
     and in lakes and streams having low ANC.
     Greatest uncertainty is in regions holding
     high elevation or rough terrain and for the
     processes of paniculate and gaseous depo-
     sition. The uncertainty in nitrogen deposi-
     tion is less important to acid-base budgets
     compared with internal fluxes. The uncer-
     tainty in base cation deposition may cause
     large uncertainty in acid-base budgets in
     some watersheds.

  2.  Watersheds having a greater proportion of
     their water flowing through shallow, more
     acidic soils tend to have more acidic and
     lower  ANC  surface  waters  than water-
     sheds in which  a large proportion of the
     water   flows  through  deeper,   more
     weatherable materials. This conclusion is
     generally valid for baseflow chemistry and
     for  episodic changes in acid-base chemis-
     try  for surface  waters.   The major path-
     ways for movement of water through wa-
     tersheds are subsurface flows, even during
     most extreme  flow events.    Generally,
     overland flow is rarely observed.

  3.  Even without acidic deposition, the natu-
     ral  sequence in watershed development is
     one of soil acidification associated with
     base cation accumulation in biomass (i.e.,
    in vegetation and humus), an increase in
    soil  cation  exchange capacity,  and  in-
    creased  leaching  of  soluble or weather-
    able materials from upper soil horizons.

4.  Naturally acidic  lakes and streams domi-
    nated  by organic anions occur predomi-
    nantly in areas where bedrock or uncon-
    solidated sediments are highly resistant to
    weathering.   Most are  in regions with
    large buildups of organic matter (e.g., ar-
    eas  with spodic soils and in  wetlands),
    often in  more northern and coastal plain
    regions having low-relief terrain and rela-
    tively poor drainage.

5.  Adsorption-desorption properties of soils
    in the watershed  regulate  export  of  at-
    mospherically deposited SO42~ into surface
    waters.   Vegetation  has only a  limited
    capacity to immobilize SO42~, even under
    optimal conditions. Great uncertainty still
    exists  regarding the reversibility of SO42'
    adsorption and critical factors controlling
    desorption.   But,  at least one documented
    study that examined adsorption-desorption
    across a  variety of soils showed that most
    soils exhibited some degree of irreversible
    adsorption.

6.  The  extent to which atmospheric nitrogen
    deposition affects the export of  NO3- from
    watersheds  depends on  biological  rather
    than geochemical  process in the water-
    shed.  Most forest soils (including those
    that  are not currently affected by nitrogen
    deposition leading to increased nitrogen
    accumulation  and  possible   saturation)
    have the potential  to produce  and leach
    NO3~ to surface waters.

7.  Sulfate deposition often  causes an anion
    shift from  drainage  water  composition
    dominated by HCO3~ and organic anions
    to one dominated by SO42".

8.  The  relative mix of cations in soil solution
    depends  on the mix of cations on soil  ex-
    change sites  (i.e., soil base saturation) and
    cation selectivities  of the exchange sites.
    In extremely acid soils (i.e., less than 10%
    base saturation), Al and  H+ dominate  the
    cations in solution, resulting  in reduced
    ANC concentrations  in  drainage  waters
    from the watershed; in less acid soils (i.e.,
    greater than 20%  base  saturation), base
    cations dominate and ANC remains unal-
                                               A-3

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
     tered.  Data on  soil chemistry and miner-
     alogy,  bedrock  mineralogy,  and!  water
     flow paths are all necessary to predict the
     chemistry of waters draining terrestrial sys-
     tems.

  9.  Water quality changes that occur when
     soil water rich in  CO2 flows into surface
     waters depend on ANC in both waters.  If
     SO42~ and NO3~ from  deposition produce
     negative ANC in the soil solution, the pH
     effects observed within the soil solution
     may  be minimal  but the  consequential
     change produced by discharge of this low
     pH and high Al soil water to surface water
     may be substantial.

  10. Many  seepage  lakes  (lakes  having  no
     stream input) that receive even 5% of their
     total  water input from groundwater are
     seldom  likely to  be  affected  by  acidic
     deposition to their surfaces.  That is, ANC
     supplied by a small proportional volume
     of ground water is often sufficient to neu-
     tralize incident acidic precipitation.  The
     chemistry  of  seepage  lakes that  derive
     greater than 95% of their input  from pre-
     cipitation are greatly influenced by incom-
     ing precipitation.

  11. Numerous processes modify the  acid-base
     chemistry  of  water  entering lakes  and
     streams, including base cation production
     and SO42- and NO3~  retention.   These
     process occur in all systems, but rates for
     these processes  vary among systems. For
     example, those  systems having  long resi-
     dence times and relatively deep  mean soil
     depths tend to have the greatest influence
     on acid-base chemistry in  their drainage
     waters.  Further, in some lakes, more ANC
     is derived from  in-lake production than
     that generated within the terrestrial water-
     shed.  The most important in-lake proc-
     esses generating ANC are likely to occur
     in the sediment  rather than  the water-col-
     umn of acid-sensitive lakes.

SOS/T REPORT 11: HISTORICAL CHANCES IN
SURFACE WATER ACID-BASE CHEMISTRY IN
RESPONSE To ACIDIC DEPOSITION
  1.  The role of acidic deposition as a cause of
     acidic surface waters  is supported by nu-
     merous  lines  of evidence,  including the
     current  chemical  composition  of acidic
     surface waters,  extensive paleolirnnologi-
    cal analyses of bottom sediment deposits,
    the worldwide distribution of acidic sur-
    face waters,  experimental  studies,  moni-
    toring and re-survey data,  and principles
    of geochemistry.

2.   Acidification of low pH and low ANC (<
    50 ueq/l) Adirondack lakes has been less
    than previously believed because of con-
    siderable watershed and in-lake neutrali-
    zation of acidic inputs via  enhanced base
    cation mobilization.

3.   Outside  the  Adirondack  Mountains, the
    chemistry of drainage lakes and streams in
    several areas  is  consistent with  the hy-
    pothesis  of acidification of sensitive sys-
    tems  by  acidic deposition.   This is most
    notable  for  the  Poconos/Catskill  subre-
    gion,   the Mid-Atlantic  Highlands  (i.e.,
    mid-Appalachians), the eastern portion of
    the Upper Midwest, and the New Jersey
    Pine Barrens.

4.   The chronic acid-base character of lakes
    in Maine has been generally unaffected by
    acidic deposition.   Chemistry of  acidic
    streams in the  Mid-Atlantic Coastal Plain,
    outside the New Jersey Pine Barrens, in
    most  cases, suggest acidification  due to
    organic acidity and not acid deposition ef-
    fects.  Chronic  acidification of  western
    lakes from acidic deposition has not ap-
    parently  occurred.

5.   The acid-base character of acidic streams
    in the Florida Panhandle can be ascribed
    to a combination of organic acidity, ma-
    rine  cation  retention,  and near  zero
    weathering inputs to some systems.  For
    some waters,  acidic deposition may also
    have provided minor contributions.

6.   In the Upper Midwest, the  chemistry of
    sensitive  (ANC<50  ueq/l,  SiO2<1  mg/l)
    seepage  lakes exhibit both increasing sen-
    sitivity (lower base cation  concentrations)
    and increasing effects from acidic  deposi-
    tion across a  longitudinal  gradient from
    west  to  east.   In the Upper Peninsula of
    Michigan, 15% of the lakes are of this
    type  and  many of  these  are  currently
    acidic because of  high SO42- relative to
    base  cation concentrations.   These lakes
    have probably been acidified by  acidic
    deposition. Throughout most of the Upper
    Midwest,  however,  substantial  regional
                                               A-4

-------
                                                           APPENDIX A: SUMMARY OF NAPAP REPORTS
     acidification  of lake water  from acidic
     deposition has not occurred.

  7.  Most lakes  and  streams in the  United
     States, especially those that have  current
     ANC greater than  about 50 ueq/l,  have
     probably  not had declines in pH or  ANC
     within their recent histories.

SOS/T REPORT 12: EPISODIC ACIDIFICATION
OF SURFACE WATER ACID-BASE CHEMISTRY
  1.  Episodic  acidification  is  the process  by
     which lakes  and streams develop short-
     term  decreases in ANC  and pH,  usually
     during hydrological events and over  time
     scales of  hours to  weeks.  Episodes are
     stochastic or  probabilistic in  nature,  in
     terms of occurrence, frequency, intensity,
     duration,    and,    to    some    extent,
     composition.

  2.  Episodic  acidification  is  practically   a
     ubiquitous process  in streams and drain-
     age  lakes. Presently, data are  not avail-
     able that  allow rigorous population  esti-
     mates of  episodic  acidification  in  the
     United  States to be completed.    Most
     states plus southeastern Canada, however,
     where monitoring  data  have  been  col-
     lected, display characteristics of episodic
     acidification.

  3.  Episodic acidification is controlled by  a
     combination of natural and anthropogenic
     factors.  The  relative importance of these
     factors varies among  regions and  among
     watersheds within  regions.   There  are
     three primary natural process that  can
     produce episodes:   (1) dilution, (2) nitrifi-
     cation, and (3) organic acid production.

  4.  Episodic acidification is  not symptomatic
     of human  caused chronic acidification.

  5.  The  severity (minimum ANC and pH  or
     highest dissolved aluminum) of episodes is
     increased  by acidic  deposition in some ar-
     eas.

  6.  While improvements  in  water  chemistry
     during episodes in some lakes and streams
     would be  expected  with reduced loading
     by  acidic deposition,  especially  in the
     Northeast  and Mid-Atlantic, this issue has
     not been addressed  by scientific investiga-
     tions.  The  roles of nitrogen  and sulfur
     deposition and of organic acids  in causing
     episodic acidification all need to be exam-
     ined.

  7.  Modeling episodic acidification in surface
     waters has been only moderately success-
     ful,,  primarily because of a  lack of data
     and a lack of understanding of  important
     hydrological   flow   and  biogeochemical
     process.

SOS/T REPORT 13: BIOLOGICAL EFFECTS OF
CHANCES IN SURFACE WATER ACID-BASE
CHEMISTRY
  1.  The most important chemical properties of
     surface waters influencing  biological  re-
     sponses  to acid-base chemistry  are pH,
     inorganic monomeric aluminum, and cal-
     cium.  Decreases in pH  (particularly be-
     low 6.0-6.5) and increases in the concen-
     tration of inorganic  monomeric aluminum
     (above 30-50 ug/l for  the  most  sensitive
     organisms) can increasingly  cause adverse
     biological effects.  Small changes in cal-
     cium are  particularly  important at low
     calcium concentrations (< 100-150 ueq/l).

  2.  A  number  of  the species that commonly
     occur  in surface  waters  susceptible  to
     acidic deposition cannot  survive,  repro-
     duce, or compete in acidic waters.  Thus,
     with increasing acidity, these "acid-sensi-
     tive" species  are eliminated and species
     richness (the number of species living in a
     given  lake or stream) declines.   These
     changes  in aquatic  community structure
     are found to begin in many  surface water
     systems as chronic pH  levels drop  below
     the range of about 6.0 to 6.5.  Acid-sensi-
     tive species occur in all major groups of
     aquatic organisms.  Both chronic and epi-
     sodic acidification can affect aquatic or-
     ganisms, with chronic acidification per-
     haps the primary cause of  continuing  ef-
     fects in acidified lakes and episodic acidi-
     fication being particularly important case
     of effects in streams.

  3.  System level processes, such as decompo-
     sition,  nutrient cycling, and productivity,
     are fairly robust and affected only at rela-
     tively high levels of acidity  (e.g., chronic
     pH<5.0-5.5).

  4.  Relatively  few studies  have  been  con-
     ducted  on  the recovery   of  biological
     communities  following reduction of acid
                                              A-5

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
      inputs.  Based on our current understand-
      ing  of  the  processes  of  biological  re-
      sponse,  decreases  in  acidity  would  be
      predicted to likely  allow  acid sensitive
      species  and  species  richness both  to
      increase in acid affected surface waters.

  5.   Laboratory toxicity experiments and field
      surveys  provide  an  adequate  basis  for
      quantifying the relationship, on a regional
      scale, between changes in pH,  aluminum,
      and  calcium  and acidity-induced stress on
      fish  populations.    Thus,  toxicity-based
      models,  field based models, and  models
      that  combine laboratory and field data can
      be used to evaluate  the biological  signifi-
      cance  of projected changes in  acid-base
      chemistry given alternative  deposition and
      emissions scenarios.

  6.   The  loss  of  fish  populations  and/or ab-
      sence of fish species as a  result of  acid-
      base chemistry changes has been docu-
      mented for some lakes and  streams in sev-
      eral  regions of the United States. Applica-
      tion  of  fish  response models suggest that
      the percentage of NSWS waters with  acid-
      base chemistry unsuitable for the survival
      of acid-sensitive fish species range  from
      less  than  5% in areas such as  the Upper
      Midwest to  near 60%  for  upper  stream
      reaches in the Mid-Atlantic Coastal Plain.
      An estimated 23% of the Adirondack lakes
      and  18% of the mid-Appalachian  streams
      classified  as  potential brook trout  habitat
      currently have acid-base chemistry unsuit-
      able for brook trout survival.

SOS/T REPORT 14: METHODS FOR PROJECTING
FUTURE CHANCES IN SURFACE WATER ACID-
BASE CHEMISTRY
  1.   The   Direct/Delayed  Response  Project
      (DDRP) approach  utilized  the  best avail-
      able procedures for  projecting  the effects
      of sulfur deposition  on  future changes in
      surface water acid-base chemistry  for tar-
      get populations of lakes in the Northeast,
      Upper Peninsula of Michigan,  and Florida
      and  streams  in the mid-Appalachians and
      Southern Blue Ridge Province.

  2.   Several models  provide credible  projec-
      tions of selected  subpopulations of target
      lakes.
  3.  Measurement,   parameter,   input,   and
     population  extrapolation  error  can  be
     quantitatively estimated for model projec-
     tions, but aggregation and model assump-
     tion error can be estimated only qualita-
     tively.  Results from individual watershed
     projects  can be extrapolated through the
     probability sampling frame for regional es-
     timates of population attributes.

  4.  Although there are remaining uncertainties
     with  respect  to structural  error, aggre-
     gation, and long-term projection confirma-
     tion,  model  projections  are  the  only
     feasible  approach   for  comparing  the
     effects of  different  illustrative emissions
     control scenarios on future changes in sur-
     face water acid-base chemistry.

SOS/T REPORT 15: LIMING ACIDIC SURFACE
WATERS
  1.  Liming  can effectively  mitigate many of
     the adverse ecological  effects  of surface
     water acidification  independent of reduc-
     tion of acidifying emissions.

  2.  Conventional whole-lake liming is a  more
     established mitigation alternative than lim-
     ing running waters and watersheds.

  3.  Liming surface waters commonly results in
     significant    positive    and    predictable
     physiochemical  changes in  aquatic  eco-
     systems.

  4.  Liming  generally  increases  nutrient cy-
     cling,  decomposition,  and  primary  pro-
     ductivity and results in  positive responses
     in fish and other aquatic biota.

SOS/T REPORT 16: CHANGES IN FOREST
HEALTH AND PRODUCTIVITY IN THE UNITED
STATES AND CAN ADA
  1.  The vast majority of forests  in Jhe United
     States are not affected by decline.

  2.  There is experimental evidence that  acid
     deposition and  associated pollutants can
     alter the resistance of red spruce to winter
     injury;  through this  mechanism, acidic
     deposition may have contributed to red
     spruce decline  at high  elevations in the
     northern Appalachians.   Evidence of red
     spruce decline and pollutant involvement
     in the southern Appalachians is  less sub-
     stantial.
                                               A-6

-------
                                                          APPENDIX A:  SUMMARY OF NAPAP REPORTS
3.  Most sugar maple trees and stands in the
    United States and Canada are not affected
    by decline.  Sugar maple declines are sig-
    nificant problems, however, in Quebec
    and in some parts of Ontario, Vermont,
    and Massachusetts.   Natural stresses, in-
    cluding nutrient  deficiencies and defolia-
    tion by insects,  are  important  factors  in
    these  declines.   Involvement  of  acidic
    deposition  and/or ozone  as  significant
    contribution or predisposing factors have
    not been demonstrated, but such involve-
    ment cannot be  ruled out on the basis of
    available information.

4.  A regional  decline of southern  pines has
    not been demonstrated.    The  reported
    growth rate reductions in certain classes of
    trees and stands  are not fully understood,
    however, and require further investigation.
    The  occurrence  of  reduction  in  tree
    growth rates in natural pine stands is an
    expected consequence of  historical  land
    use patterns, increases in  stand  age and
    competition, and other non-pollutant fac-
    tors.   Available  information  is not ade-
    quate to determine whether the magnitude
    of reported growth reduction is greater or
    less than would  be expected  in the ab-
    sence  of acid deposition and associated
    pollutants.   Results  of exposure-response
    experiments indicate that ozone  at ambi-
    ent  concentrations can  alter the growth
    and physiological responses of  southern
    pines  seedlings.   This  justifies  concern
    about  adverse effects of  ozone  on the
    health and  productivity of  southern pine
    forests.

5.  Compared to ozone and many non-pollut-
    ant  stress factors,  acidic deposition ap-
    pears to be a relatively minor factor affect-
    ing the current health and productivity of
    most forest  in the United States and Can-
    ada.   Most of these forests are receiving
    acidic depositions at doses that have not
    had a serious  impact on health and  pro-
    ductivity.   The  possibility  of  long-term
    (several decades) adverse effects on some
    soils appears realistic.  Sulfate deposition
    increases leaching losses of nutrient cat-
    ions from many  different forest soils and
    over the long term may reduce the fertility
    of  soils with low buffering  capacities or
    low mineral weathering rates.
SOS/T REPORT 17:  DEVELOPMENT AND USE OF
TREE AND FOREST RESPONSE MODELS
  1.  The models  presented in this report are
     preliminary,  and they emphasize our lack
     of knowledge about fundamental tree and
     forest processes.  Nonetheless, considera-
     tion of the dynamics implied by what we
     do know of the processes indicate that
     considerable caution is needed in project-
     ing long-term effects from acidic deposi-
     tion and ozone.  In particular, long-term
     dynamics  generated  by  synergies,  and
     compensations between mechanisms, can
     produce threshold effects. The possible ex-
     istence  of  these threshold  effects implies
     that simple projections will not  be ade-
     quate to capture long-term effects of acid
     deposition. Therefore, the null hypotheses
     of no long-term effect should not be ac-
     cepted without caution, even if it appears
     warranted by the current data and theory.

SOS/T REPORT 18:  RESPONSE OF VEGETATION
To ATMOSPHERIC  DEPOSITION AND AIR
POLLUTION
  1.  Based on crop-effects research  conducted
     by NAPAP and other research programs,
     acidic precipitation at ambient levels  in
     the United States has not been shown  to
     be responsible for regional crop yield re-
     duction.

  2.  Ambient fog acidity concentrations are not
     great  enough to reduce the yield of agri-
     cultural crops, but under certain localized
     conditions may  occasionally  be  high
     enough  to cause visible injury to plant tis-
     sue and thereby reduce the marketability
     of sensitive crops.

  3.  Ambient  SO2  concentrations  by  them-
     selves  are not responsible for regional-
     scale crop yield reductions in the United
     States.

  4.  Nitrogen  dioxide  at ambient  concentra-
     tions  is not  a  direct  source of regional-
     scale growth  or yield reduction in U.S. ag-
     ricultural crops.

  5.  Although  pollutant mixtures (e.g., SO2  +
     O3, or SO2 +  NO2) are  of undetermined
     importance on a national scale, at least  in
     some regions  (e.g., Ohio River Valley),
     ambient air  quality monitoring  suggests
                                             A-7

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
     the potential for effects from mixed expo-
     sures.

  6.  There  is evidence that acidic cloud water
     in combination with other stresses affects
     some  high elevation spruce forests  in the
     eastern United States.

  7.  Long-term changes in the  chemistry of
     some  sensitive  soils is  expected  from
     acidic  deposition,  but  it  is  uncertain
     whether this will  result  in reduced forest
     health, how  effects will be manifest, how
     much  of  the forest  resources will  be im-
     pacted, or how long it will  take for such
     effects to  occur.

  8.  There   is   no  conclusive evidence  that
     acidic precipitation  is a major causal fac-
9.
tor in sugar maple decline, but in limited
areas where nutrient deficiency symptoms
are  currently  evident,  acidic  deposition
could further exacerbate their expression.

Ambient SO2  concentrations  are  not re-
sponsible for regional-scale growth reduc-
tions in the United States.
10. Nitrogen  dioxide at ambient concentra-
    tions is  not a direct source  of  regional-
    scale growth reduction in forests of the
    United States.

11. The same concern about  possible effects
    from pollutant  mixtures   discussed  for
    crops  (Conclusion 5,  above) applies to
    forests.
                                                A-8

-------
      APPENDIX B

    PLOTS FROM EPA's
NITROGEN BOUNDING STUDY

-------
                                       APPENDIX B

                 PLOTS FROM EPA'S NITROGEN BOUNDING STUDY
INTRODUCTION
The Nitrogen  Bounding  Study (NBS)1 is one of
several  recent and ongoing studies conducted by
the U.S. Environmental Protection Agency (EPA) to
investigate aquatic and terrestrial  effects of acidic
deposition.  This study was initiated to address a
major issue that arose during EPA's completed Di-
rect/Delayed Response Project (DDRP), namely to
investigate the  role of nitrogen compounds in the
soil water and  surface water acidification within
forested watersheds.   Models and  analyses used
during the DDRP focused on sulfur deposition and
its effects on water chemistry, but effects due to ni-
trogen cycling received much less attention.  That
difference in focus was due to the general lack of
quantitative knowledge regarding  nitrogen trans-
formation  processes in soils. In addition, evidence
available when the DDRP was designed and initi-
ated primarily  indicated  that  most deposited  at-
mospheric nitrogen is taken up and held by biota,
thus making little contribution  to acidification.
More recent evidence suggests,  however,  that
some forest catchments   in  the  eastern  United
States, for  example, can become nitrogen saturated
and that nitrogen leaching from these systems can
contribute substantially to lake and stream acidifi-
cation,  particular during  runoff episodes.   While
several  long-term studies intended to  address the
role of nitrogen deposition  in surface water acidi-
fication  are under way, the NBS was  intended to
provide   near-term  information  on  which  to
"bound" likely relationships for nitrogen and sulfur
deposition  on   surface-water  acidification   re-
sponses.

The NBS  evaluated target  populations of surface
waters in three  sensitive geographic regions: lakes
in the Adirondack Region  and stream reaches in
the Mid Appalachian  Region and the  Southern
Blue Ridge Province.   Model projections com-
pleted during the NBS  used a  modification of the
Model of  Acidification of Groundwater in Catch-
ments (MAGIC), the  model of current choice  for
1  Van Sickle,  J., M.R. Church.   1995.   Methods for
  Estimating the Relative Effects of Sulfur and Nitrogen
  Deposition  on Surface  Water  Chemistry.    U.S.
  Environmental Research Laboratory, Corvallis, OR.
assessing  many watershed  processes  associated
with acidic deposition.  It  includes a minimum
number  of critical  chemical  and  hydrological
processes occurring in watersheds to simulate soil
solution and surface water chemistry, and to pro-
ject average monthly or annual concentrations of
acid-base chemistry in surface water.    Primary
input data  for its use in NBS came from the  Na-
tional Surface Water Survey (NSWS), the DDRP,
and updated deposition information from EPA at-
mospheric   modeling   studies   discussed   in
Chapter 3 of this report.

Nitrogen uptake  parameters in the  model were
used  to  provide  simple  surrogates  for complex
processes within the nitrogen cycle. That is, these
parameters were adjusted to  yield "best  case" (i.e.,
maximum nitrogen retention in the biota within a
watershed)  and "worst  case" (i.e.,  complete nitro-
gen saturation  in the biota within a watershed) ap-
proximations  to estimate the resulting  combined
effects by nitrogen  and sulfur  deposition on lake
and stream acidity.  Thus,  model  results provide
upper and lower bounds on the levels of acidifica-
tion  that more realistic models (currently under
development) would likely project.

The NBS projected surface water chemistry for two
target years (years 2015 and 2040) using the as-
sumption that  emissions reductions of  10 million
tons SO2 and 2 million  tons NOX mandated by the
1990 Clean Air Act Amendments of (CAAA) were
fully implemented.  As such, deposition  rates for
sulfur and  nitrogen were assumed  to equal those
projected by  atmospheric models  to accompany
emissions reductions with full CAAA implementa-
tion at the  year 2010  and that any further reduc-
tion would  be attained by the year 2020.

After  the year 2010,  the NBS defined  different
deposition  scenario projections for years between
the years 2010 and 2020 using  S and  N  deposition
rates that decline linearly from the common year
2010 rates to a selection of different modeled year
2020 deposition rates for each scenario modeled.
For example, some modeled scenarios maintained
the 2010 deposition rates through the year 2020,
while  some  alternative  modeled  scenarios  de-
                                               B-1

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
creased the year 2010 rate to background deposi-
tion  rates only by the year 2020.  Rates for still
other modeled scenarios decreased to levels be-
tween these  extremes.  (Background deposition
rates  are those materials that originate from only
natural, agricultural  fertilizer,  and domestic live-
stock sources.)  Each modeled  deposition rate was
then assumed to remain constant at  the  specific
modeled 2020 rate until the year 2040, the end  of
the model projection period.  The selected plots
from the NBS contained in  this appendix include
model projections for the years 2015 and 2040.
The 2040 plots were primarily depended upon for
the review and conclusions from the  NBS  pre-
sented in Chapter 2 of this report.

In examining the NBS plots it  is first important  to
recognize that the plotted response surfaces repre-
sent only projected proportions for the target sensi-
tive surface waters of the NBS; they do not repre-
sent  responses for either all surface waters or for
all NSWS sampled surface waters in the modeled
regions.  Each page holds four NBS plots  display-
ing projected response surfaces over ranges of pos-
sible  sulfur and  nitrogen deposition  rates at the
year 2040.  Each  of the four plots represent one of
the four possible durations used to bound likely al-
terative times to watershed nitrogen saturation:  50
years,  100 years, 250 years, and  never (i.e.,  as-
sumes nitrogen uptake remains constant  into the
future at recently estimated rates).

Sets  of pages are grouped within the three mod-
eled regions,  (1) Adirondacks Region, (2) mid-Ap-
palachian Region, and (3)  Southern  Blue Ridge
Province.  Within the 10 pages for each region,
one  each is presented for projections of relative
proportions of lakes  with ANC<0  ueq/l,  ANC<50
ueq/l,  pH<5.0,  pH<5.5,  and  pH<6.0  over the
ranges of sulfur and  nitrogen depositions assessed
for each region and for each of the four bounding
times  to watershed  nitrogen  saturation,  for the
years 2015 and 2040.

INTERPRETING THE NBS PLOTS
To illustrate interpretation  of  the  NBS  plots, the
first  page of  year  2040 plots  shows  projected
percentages of NBS target population lakes in the
Adirondack Mountains having ANC of 0  ueq/l  or
less, under assumptions of four bounding times  to
watershed nitrogen  saturation  equal 50, 100, and
250  years, and  never (i.e., constant).  Response
contours for each plot show how percentages  of
target waters are projected to relate to  possible
differences in  total sulfur and  nitrogen deposition
loading rates throughout the modeled ranges for
these depositions.   For these plots,  deposition
ranges for  the  year  2040  begin at  projected
background deposition rates for sulfur and nitrogen
(1  kg S/ha/yr and 4 kg N/ha/yr) and extend to their
maximum modeled concentrations (7.5 kg S/ha/yr
and  11.3  kg N/ha/yr), which are the rates  pro-
jected  to  accompany full implementation  of the
1990CAAA.

Thus,  the  plots  on  page   B-10  shows   NBS
projections for proportions  of Adirondack  lakes
having ANC<0 ueq/l.   The upper  right plot shows
projections for an assumed 100 years to watershed
nitrogen  saturation.    Here,  in  the  upper  right
corner of  this  upper right plot, model projections
estimate  that  approximately  26  percent of  the
target Adirondack lakes  may be acidic (ANC<0
ueq/l)  in  the year 2040 for  modeled sulfur and
nitrogen deposition rates projected to accompany
implementation of the 1990 CAAA.   With  only
"background" deposition of sulfur  and  nitrogen, as
shown in  the lower left corner of the upper  right
plot, 3.4 percent of these target lakes are projected
to be acidic in 2040.

In reviewing these plots,  it is helpful to recognize
that  several relationships generally apply to all of
them:

   *  The slopes of contour lines in  each of the
     NBS  response plots  reflect the relative im-
     portance  of sulfur and nitrogen in causing
     the    projected   response   relationships.
     Nearly vertically plotted response contours
     indicate that the projected ANC response is
     attributable primarily to sulfur deposition.
     Nearly   horizontal    plotted    response
     contours indicate the plotted  ANC response
     is  attributable   primarily   to   nitrogen
     deposition.   A forty-five degree diagonal
     contour indicates equal  contributions by
     both  sulfur and nitrogen depositions.

   *  Changes in the spacing between individual
     response  contours within each plot appears
     to be attributable to  patterns  in  sample
     weighting during model projections, rather
     than  due to some intrinsic character of the
     deposition-response relationships.

   *  The density of response contours across the
     modeled deposition ranges for each plot di-
     rectly relates to the potential  average re-
     sponsiveness by  target  waterbodies to po-
     tential changes in sulfur and nitrogen depo-
     sition rates on the specified water quality
                                               B-2

-------
                                                                             APPENDIX B: NBS PLOTS
     classification variable modeled (e.g., ANC<
     0  ueq/l).   Therefore, plots  with a  high
     density of contour lines depict a high  level
     of  responsiveness   to   future   possible
     changes in deposition rates.

   * In general, modeled ANC responses to re-
     ductions  in sulfur deposition found during
     the NBS  appear  to be linear and propor-
     tionally equivalent  across the  ranges of
     modeled  sulfur reductions.  Additional  in-
     vestigation may help to determine whether
     this relationship is due to actual  environ-
     mental  functions  or to  some artifact
     inherent in the model's application.

In considering possible individual  extrapolation of
the results presented with these plots, beyond that
presented in the preceding chapters, care must be
taken to ensure that these  results are  not over ex-
trapolated. That is, in applying the NBS model  re-
sults, as is the case when applying any simulation
modeling results, it is  important that  the assump-
tions underlying the modeling  be  understood and
carefully considered relative to additional  condi-
tions or systems to which they are to be applied.
In doing this, the modeled processes should be de-
scribed and evaluated to determine how well they
correspond  to  the system (e.g.,  watershed) for
which the application is  intended.   In general,
models should only be applied to (i.e., constrained
to) systems,  conditions, and assumptions that fall
within or very near the boundaries  of those  used to
develop the model. Whenever  models are applied
outside these   boundaries,  the  consequences  of
knowingly violating the model's constraints should
be assessed as part of the model analysis. Unfortu-
nately, violation of model  assumptions cannot  al-
ways be readily known or easily assessed.  Never-
theless, when model constraints are not met by the
natural processes modeled or by the data collected
for model input,  model projections typically will
deviate from reality.  The  magnitude of such de-
viations contribute  markedly to what is generally
called model   error.    Estimates  of  the possible
magnitude of model error are termed model uncer-
tainty.  Reasons why watershed models, including
MAGIC, are particular difficult to  design and test
include the following.

   * Processes  controlling watershed  functions
     are very difficult to observe either in nature
     or in  any laboratory experiment.  Thus,
     these processes may be either virtually un-
     known  or inaccurately represented in the
     model, i.e., the model might not be a "true"
     model.

   * Actual conditions within individual water-
     sheds that determine processing or transfer
     rates   may   be  unknown   or   poorly
     quantified.  Consequently, the model might
     be poorly parameterized, i.e., the modeled
     parameters  may be  poorly  adjusted  or
     calibrated   to  approximate  parameters
     actually  occurring within   the  modeled
     system.

   * Inputs to the models can be poorly known
     or unable to be accurately predicted (e.g.,
     dry deposition loads to a watershed).

   * Models often are difficult to test.  That is,
     models   may  be   largely   "unverified,"
     "unconfirmed," or "unvalidated." In fact, it
     is often argued that a model can never be
     confirmed to be true, it can  only  be falsi-
     fied by failing to accurately project some
     outcome.  Further, when a model does ac-
     curately predict an outcome, its validity is
     not proven,  because the  "right" result may
     have been projected for the wrong reason.

Further, it is  useful  to  remember  that environ-
mental monitoring  and simulation modeling have
complementary  environmental  assessment  roles.
Effects monitoring (i.e., surface water chemistry) is
necessary  to determine the actual  effects of acidic
deposition rates on environmental resources and to
provide data  needed to develop and test models.
In turn, simulation modeling  is useful  to project
potential differences among  future  deposition or
environmental scenarios. This is particularly  true
for projections of acidic deposition effects because
no set of  monitoring records of sufficient length
exist  that  allow establishing  a  clear  statistical
relationship linking changing historic  ecological
responses  to  changing  acidic  deposition  input
rates. Without this historic record it is not possible
statistically to  project  future  changes, i.e., we
cannot predict the future directly from the  past
because we do not know the past.

Simulation models,  such as  the  MAGIC model,
generally do not require well quantified historical
relationships   to   provide   potentially    useful
projections of future conditions.  But  watershed
simulation  modeling,  including  the NBS,  often
continues  to  include significant  uncertainties, as
noted in the above paragraphs and  in Chapter 2.
This is especially the case regarding the modeling
of nitrogen cycling  within watersheds and the po-
                                                B-3

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
tential acidifying effects of nitrogen deposition on
soils, watersheds, and associated surface  waters.
Additional uncertainties also remain regarding ex-
actly which processes to include in;such  models
and  about how such processes should be linked
within these models.  Further,  there  is almost a
complete  dearth  of  monitoring  and   survey
information on the regional distributions of water-
shed characteristics that would allow such  models
to be calibrated  and applied  to project future
effects. Improving capabilities to model these rela-
tionships will lead to better projections of potential
future environmental effects from both  sulfur and
nitrogen  deposition.  Present knowledge of nitro-
gen cycling and early steps toward nitrogen mod-
eling provides a solid foundation for more impor-
tant and  productive advances  in this field.  These
advances, particularly, would  lead to significantly
reduced  uncertainty in  potential future effects of
nitrogen.

-------
                                                             APPENDIX B:  NBS PLOTS
Net annual N uptake <= 5% in 50 yr Net annual N uptake <= 5% in 100 yr
12-
IT
| 10-
?>
.*,
g
'55
8. 6-
0)
Q
c
0) 4-
o
1 2-
CO
£
0-

^NN^R
X^1^,
22 ^x.
\ x
\
20.8 %







12-
"S*
=c 10-
O)
^ 8-
g
O c
Q. o-
o>
Q
c
(D 4-
0) *
o
Z 2_
CO
I—
0-


V
^\^
• o \^
\
1 1 .9 %





















1 1 1 1 i i i i i i i i
02468 10 02468 10
Total Sulfur Deposition (kg S/ha/yr) Total Sulfur Deposition (kg S/ha/yr)
Net annual N uptake <= 5% in 250 yr Net annual N uptake unchanging
12-
«T
| 10-
z
05
^

g
'55
O c
Q. D~
0)
Q
c
S, 4-
O)
o
l_
1 2-
2
"Q
1-
o-




s.
\
\0

8.7 %











12-
"^
^ 10-
z
o>
V
' — ' 8—
g
'co
o c
Q. o
CD
Q
c

-------
ACID DEPOSITION FEASIBILITY STUDY
Net annual N uptake <= 5% in 50 yr Net annual N uptake <= 5% in 100 yr
12-
1
1 10-
g>
"? 8~
g
'55
a e-
 4-
o
E 2-
l—
0-

t
\
\
\
\
\
\
53.5 %























I
0
I
2
                             8
10
i
0
       Total Sulfur Deposition (kg S/ha/yr)
    2     4     6     8    10
Total Sulfur Deposition (kg S/ha/yr)
Exhibit B2.  Percent of  target  population  lakes with  ANC <>  50 meq/L  for  the
             Adirondacks region at Year 2015. Deposition equals median deposition
             at Year 2015.
                                        B-6

-------
                                                           APPENDIX B:  NBS PLOTS
Net annual N uptake <= 5% in 50 yr Net annual N uptake <= 5% in 100 yr
12-
ir
| 10-
O)
? 8-
.g
'55
8. 6-
o>
O
c

c" 8~
.0
Q. 6-
0)
Q
c
S, 4-
Q
1 2-
o
1-
0-

\Vl
) \ t<)
( \
A i
\
3.3 %








12-
=c 10-
z
o>
"c 8^
g
O c
Q. (3-1
0)
Q
c
co 4^
I
^
^.
lo
0
t-
0-

1 \
\
)
1 *
\
3.5 %























i i i i i i i i i i i i
02468 10 02468 10
      Total Sulfur Deposition (kg S/ha/yr)
Total Sulfur Deposition (kg S/ha/yr)
Exhibit B3.  Percent of target population lakes with pH < 5.0 for the Adirondacks
            region at Year 2015. Deposition equals median deposition at year 2015.
            pH is estimated from the empirical pH-ANC model.

-------
ACID DEPOSITION FEASIBILITY STUDY
Net annual N uptake <= 5% in 50 yr Net annual N uptake <= 5% in 1 00 yr
12-
k.
is
€ 10-
z
2
c" 8~
0
'55
1 6"
Q
C
0) 4-
8"
w
•*•*
? 2-
o

0-





•
L

2

N. N.
X^v N.
\\\. 3
g\ jQ \p
,
\

LJ


i



25.4 %









i i i
024









i
6









12-
13
=c 10-
z
S
c" 8~
g


\
.5%









i
6

3




3












12-
"5l
"t3
=c 10-
Z
0)
f" 8-
g
55
O c
Q. 0-
0)
Q
c

P 1^ 14
\
\
9.6 %




























ii i i i i i i
8 10 0 2 4 6 8 10
       Total Sulfur Deposition (kg S/ha/yr)
Total Sulfur Deposition (kg S/ha/yr)
Exhibit B4.  Percent of target population lakes with pH < 5.5 for the Adirondacks
            region at Year 2015. Deposition equals median deposition at year 2015.
            pH is estimated from the empirical pH-ANC model.
                                       B-8

-------
                                                           APPENDIX B: NBS PLOTS
Net annual N uptake <= 5% in 50 yr Net annual N uptake <= 5% in 100 yr
12-
>»
73
| 10-
a>
1= 8~
o
'53
O c
Q. 6-
Q
c

|
z
Is
.0
0-

\
- — \
X\
\\>
4ll4-
T
A
38.1 %







12-
>,
73
^ 10-
2
o»
c" 8~
g
"55
O ^
Q. D~
Q
C
0) 4-
0) H
2
5 2-
,2
o-

\ \
\\
//
L)
3^83
T
A
30.3 %























1 i i i i i i i i i i i
02468 10 02468 10
Total Sulfur Deposition (kg S/ha/yr) Total Sulfur Deposition (kg S/ha/yr)
Net annual N uptake <= 5% in 250 yr Net annual N uptake unchanging
12-
15
=£ 10-
2
O)
-X
~c 8~
;|
8. 6-
0)
Q
c
& 4-
I
2 2_
3
o
h-
0-





^
30.2 %










12-
7B~
=c 10-
~?
at
o
55
8. 6-
0)
Q
c

o
z
~j} ^
"3

0-

I)
/ /
( (
in T/P Tbl
J^J v^f^. CfM-
\
29.7% .



























    0     2    4     6     8    10
      Total Sulfur Deposition (kg S/ha/yr)
r
0
    2     4     6     8    10
Total Sulfur Deposition (kg S/ha/yr)
Exhibit B5.  Percent of target population lakes with pH < 6.0 for the Adirondack^
            region at Year 2015. Deposition equals median deposition at year 2015.
            pH is estimated from the empirical pH-ANC model.
                                       B-9

-------
ACID DEPOSITION FEASIBILITY STUDY
      Net annual N uptake <= 5% in 50 yr
  12-
w


1 10-





I 8-



I-
O

§  4-

i

E  2-
JO
'o

   0-
         6%
      I

      0
I

2
                   i

                   4
I

8
 i

10
        Total Sulfur Deposition (kg S/ha/yr)
                                     Net annual N uptake <= 5% in 100 yr
                                               12-

                                             t.



                                            1 10-





                                            ?  8-
                                            .o

                                            '«

                                             8.  eH
                                             o>
                                            O


                                             o>  4-
                                             c»  H
                                             o
                                             to
                                             o
                                                0-
                                        3.4 %
i

0
i

4
F

6
8
 i

10
                                        Total Sulfur Deposition (kg S/ha/yr)
     Net annual N uptake <= 5% in 250 yr
  12-

i—
>s


l.OH
c
o
   8-
a 6^
0)
Q
c

o>
o
   0-
         0%
      0     2     4     6     8     10


        Total Sulfur Deposition (kg S/ha/yr)
                                       Net annual N uptake unchanging
                                   12-
                                    ,
                                   1QH
                                             O)



                                             g

                                             "to

                                             a
                                             0)
                                             0
                                   8-
                                   4-
                                   0-
                                         0%
                                      0     2     4     6     8     10


                                        Total Sulfur Deposition (kg S/ha/yr)
Exhibit B6.  Percent of  target  population lakes  with  ANC  < 0  meq/L  for  the

             Adirondacks region at Year 2040. Deposition equals median deposition

             at Year 2020.
                                         B-10

-------
                                                               APPENDIX B:  NBS PLOTS
     Net annual N uptake <= 5% in 50 yr
  12-
  10-
o
8s
   8-
   6-
   4-
   0-
        53.2 %
      I
      0
                  i
                  4
                         I
                        6
                       i
                       8
 i
10
       Total Sulfur Deposition (kg S/ha/yr)
                                                  Net annual N uptake <= 5% in 100 yr
                                               12-
                                             w

                                             £ 10-



                                             t  8-
                                             .g
                                             '55

                                             I  6"
                                             Q
                                             c
                                             O)
                                             o
                                                4-
                                               2-
                                                0-
                                                     50.9 %
1
6
i
8
024

  Total Sulfur Deposition (kg S/ha/yr)
 i
10
     Net annual N uptake <= 5% in 250 yr
  12-


f 10-
   8-
O)


,|
'35
a e-
CD
Q

I. 4-
   0-
                           -54
        44.1 %
                  i
                  4
                         i
                        6
                               i
                              8
    2     4     6     8     10

Total Sulfur Deposition (kg S/ha/yr)
                                                   Net annual N uptake unchanging
                                               12-
                                             o

                                            '35
                                             o
                                             Q.
                                             0)
                                            Q
                                             c
                                             CD
                                             O)
                                             o
                                             O
                                                8-


                                                6-
                                                0-
                                                                        -54
                                                       4S4B50
                                                     44%
               0     2     4      6     8     10

                Total Sulfur Deposition (kg S/ha/yr)
Exhibit B7.  Percent of  target  population  lakes with ANC  < 50  weq/L for  the
             Adirondacks region at Year 2040. Deposition equals median deposition
             at Year 2020.
                                        B-ll

-------
 ACID DEPOSITION FEASIBILITY STUDY
     Net annual N uptake <= 5% in 50 yr
  12-



^ 10-






f 8"

I 6-
0)
Q


I <
|

Z  ,

ffl  '



   o-
        4.1 %
           ~
            2
                        i
                        6
                              i
                              8
          4     6     8    10

Total Sulfur Deposition (kg S/ha/yr)
                                                Net annual N uptake <= 5% in 100 yr
                                             12-


                                           lio-
                                           01
                                              8-
                                              6-
                                           g

                                           
                                            cn
                                            o
12-
10-
8-
6-

0-


i

\
\
|\

; i
\
3%




i i i i i i
0 2 4 6 8 10
        Total Sulfur Deposition (kg S/ha/yr)
                                                   Total Sulfur Deposition (kg S/ha/yr)
Exhibit B8.  Percent of target population lakes with pH < 5.0 for the Adirondacks

             region at Year 2040.  Deposition equals median deposition at year 2020.

             pH is estimated from the empirical pH-ANC model.
                                        B-12

-------
                                                               APPENDIX B: NBS PLOTS
      Net annual N uptake <= 5% in 50 yr
  12-
  10-
o
   8-
  4-
                                            O)  H
                                            o
                                               o-
                                                    6.1 %
                                                              i

                                                              4
02468


  Total Sulfur Deposition (kg S/ha/yr)
                                 \

                                10
     Net annual N uptake <= 5% in 250 yr
  12-
1 10-
Ol
g
   8-
O  c_
CD

Q



CD

O)
O
   0-
        3.5 %
      0     2     4     6     8     10


        Total Sulfur Deposition (kg S/ha/yr)
                                                  Net annual N uptake unchanging
                                              12-
                                              10-
                                            o>
                                            C  "
                                            g

                                            '55

                                            a  e-
                                            Q


                                            
-------
 ACID DEPOSITION FEASIBILITY STUDY
     Net annual N uptake <= 5% in 50 yr
  12-
  10-
o>
.1
•53
   8-
a eH
CD  4-
81   H
   0-
        33.4 %
      0     2     4     6     8     10

        Total Sulfur Deposition (kg S/ha/yr)
                                                Net annual N uptake <= 5% in 100 yr
                                              12-
                                              10-
                                           .1

                                           a
                                           CD
                                           Q
                                              8-
                                               4-
                                              o-
                                                   21.8%
                                                        I
                                                       2
                                                                         i
                                                                         8
0     2     4    6     8     10

  Total Sulfur Deposition (kg S/ha/yr)
     Net annual N uptake <= 5% in 250 yr
                                                  Net annual N uptake unchanging
  12-
  10-
- 8H
g
'55
a e-
CD
Q
C
CD 4-
O) ^
O


I *
   0-
         10%
      0     2     4     6     8     10

        Total Sulfur Deposition (kg S/ha/yr)
                                              12-
                                              10-
                                            0)
                                            c
                                            g
                                               8-
                                            a  e-
                                            CD 4-
                                            0) H
                                            2


                                           I 2-
                                               0-
                                                 0     2     4     6     8    10

                                                   Total Sulfur Deposition (kg S/ha/yr)
 Exhibit BIO. Percent of target population lakes with pH < 6.0 for the Adirondack^
              region at Year 2040.  Deposition equals median deposition at year 2020.
              pH is estimated from the empirical pH-ANC model.
                                         B-14

-------
                                                            APPENDIX B: NBS PLOTS
Net annual N uptake <= 5% in 50 yr
14-
u.
| 12-
2
Deposition (kg
O> CD O
i i i
c
1 4-
2
"Jo 2-
o
0-

il
(





i i i
024

"^
X,
te *-•
)%





i i i
6 8 10
Total Sulfur Deposition (kg S/ha/yr)
1
14-
2 10-
0
1 8-
s.
S 6-
c
0>
O) .
o 4-
z
S 2-
0-








I
14-
"w*
c8 1 9—
i
g
~ 8—
o
Q.
8 6-
1
Q 4-
2
"S 2-
£
0-
Met annual N uptake <= 5% in 100 yr

A
(





i i i
024

K X

)%





I I I
6 8 10















Total Sulfur Deposition (kg S/ha/yr)
Met annual N uptake <= 5% in 250 yr

A
(





v/\^
^
3%










14-
^ 12-
Deposition (kg N
o> co o
i i i
0)
I5 4-
2
« 2-
0-
Net annual N uptake unchanging

(







}%
















                I
                4
\
6
i
8
    2    4    6    8    10

Total Sulfur Deposition (kg S/ha/yr)
                                                 Total Sulfur Deposition (kg S/ha/yr)
Exhibit Bll. Percent of target  population streams  with ANC <  0 meq/L for the
             Mid-Appalachian  region at Year  2015.   Deposition equals median
             deposition at Year 2015.
                                       B-15

-------
ACID DEPOSITION FEASIBILITY STUDY
Net annual N uptake <= 5% in 50 yr Net annual N uptake <= 5% in 100 yr
14-
£
| 12-
gio-
^
0
i s-
a
& 6-
c

g> 4_
£
^p
« 2-
£
0-

^SsN^3^-'
^XX^4-
^^*'
r
20.7 %








14-
^.
>,
^ 12-
5 10-
c
o
1 8-
a
S 6-
c
o>
? 4-
k.
3 2~
J2
0-






\ \,

^^
\

Q
O
i

17.2 %








1 1 i i i i ill
02468 10 024








I I I
6 8 10
























Total Sulfur Deposition (kg S/ha/yr) Total Sulfur Deposition (kg S/ha/yr)
Net annual N uptake <= 5% in 250 yr Net annual N uptake unchanging
14-
k_
>.
|i 12-
z
U 10-
^
1 8^
a
a e-
c

f 4-
Z
3 2-
o
H-
0-

>d "2ii
\.
\. \v
\V N.
"fl Tl/lX X.
\ ~
17.1 %









14-
>,
^ 12J
^
5 10-
c
o
I 8-
o
a
& 6-
c
CD
a> .
o 4—
Z
3 2~
Q
f—
0-







ie









1 1 1 1 1 1 III
02468 10 024

N\XS\XN?
^v\
N^Sc
>€

\
5.9 %









i i i
6 8 10

p


































Total Sulfur Deposition (kg S/ha/yr)
                                                  Total Sulfur Deposition (kg S/ha/yr)
Exhibit B12. Percent of target population streams with ANC <  50 weq/L for the
             Mid-Appalachian region  at  Year 2015.   Deposition equals  median
             deposition at Year 2015.
                                       B-16

-------
                                                           APPENDIX B: NBS PLOTS
Net annual N uptake <= 5% in 50 yr Net annual N uptake <= 5% in 100 yr
14-
£
| 12-
§
!io-
£2
O
1 8-
1
Q 6—
0)
o 4-
2
3 2-
,2
o-











^1
0%







i i i i i i
0 2 4 6 8 10
Total Sulfur Deposition (kg S/ha/yr)















14-
>.
J 12-
2
O) An
^ 10-
--^'
C
1 8-
§.
& 6-
§
S 4~
2
2 2-
(2
o-

X




A
0%






















1 i 1 1 1 I
0 2 4 6 8 10
Total Sulfur Depos tion (kg S/ha/yr)
Net annual N uptake <= 5% in 250 yr Net annual N uptake unchanging
14-
u.
>>
|j 12-
JMo-
^
o
! 8~
o
Q.
Q Q~
c
.
| 12-
C" m
J£_ 10-
c
o
1 8-
o
QL
® o
Q 6-
C
0)
1 4-
2
* 2-
i°
0-





\
0%























     I
     0
I
2
i
4
i
6
\
8
10
I
0
8
10
       Total Sulfur Deposition (kg S/ha/yr)
                                     Total Sulfur Deposition (kg S/ha/yr)
Exhibit B13. Percent  of  target  population  streams  with  pH  <  5.0  for  the
            Mid-Appalachian  region at Year 2015.   Deposition equals  median
            deposition at year 2015.  pH is estimated from the empirical pH-ANC
            model.
                                     B-17

-------
ACID DEPOSITION FEASIBILITY STUDY
Net annual N uptake <= 5% in 50 yr Net annual N uptake <= 5% in 100 yr
14-
]j| 12-
Deposition (kg t
en CD o
t i i
§
I1 4-
Z
ro 2-
^
0-

XN^
X^x
3.8 %




i i i i i
0 2 4 6 8 10
14-
| 12-
^
Deposition (kg I
en co o
i - i I
0)
O) .,
g 4-
15 2-
o
0-

^\
X ^
\
0.5 %




i i i i i i
0 2 4 6 8 10






Total Sulfur Deposition (kg S/ha/yr) Total Sulfur Depos tion (kg S/ha/yr)
t
14-
| 12-
Deposition (kg 1
co co o
I i I
CD
I ^
Z
15 2-
0-
Met annual N uptake <= 5% in 250 yr

^^
\^ N
0%



14-
>.
« 12-
^
Deposition (kg I
»^
CO CD O
I 1 1
CD
O) .
o 4-
« 2-
0-
Net annual N uptake unchanging

^

A
0%








          2     4    6    8    10

       Total Sulfur Deposition (kg S/ha/yr)
    i
    2
i
6
\
8
 \
10
Total Suliur Deposition (kg S/ha/yr)
Exhibit B14. Percent  of  target  population  streams  with  pH  <  5.5  for  the
            Mid-Appalachian region  at  Year 2015.   Deposition equals  median
            deposition at year 2015. pH  is estimated from the empirical pH-ANC
            model.
                                      B-18

-------
                                                           APPENDIX B: NBS PLOTS
Net annual N uptake <= 5% in 50 yr
14-
]j 12-
^
Deposition (kg I
en oo o
I I I
CD
§> 4-
Z
15 2-
jO
0-

i.
5





i i i
024

^§^i
"^^?
6%





i i i
6 8 10
>







14-
>»
J 12-
Deposition (kg
en oo o
_L L 1
C
0)
8> 4-
Z
To 2-
o
0-
Met annual N uptake <= 5% in 100 yr

^|^,
\^^>
\
4.4 %





I I i I i I
0 2 4 6 8 10







Total Sulfur Deposition (kg S/ha/yr) Total Sulfur Deposition (kg S/ha/yr)
I
14-
^ 12-
Deposition (kg N
en co o
i i i
0)
O) .,
Q 4-
z
15 2-
£
0-
^Jet annual N uptake <= 5% in 250 yr

4
4





"^^^
^^
3%










14-
1 12-
Deposition (kg ^
en CD o
i i i
0)
en ,
Q 4-
Z
75 2-
o
0-
Net annual N uptake unchanging

—v
\^
3.5 %










                               I
                               10
                   8
 i
10
       Total Sulfur Deposition (kg S/ha/yr)
Total Sulfur Deposition (kg S/ha/yr)
Exhibit B15. Percent  of  target  population  streams  with  pH  <  6.0  for  the
            Mid-Appalachian region at Year  2015.   Deposition  equals median
            deposition at year 2015. pH is estimated from the empirical pH-ANC
            model.
                                      B-19

-------
ACID DEPOSITION FEASIBILITY STUDY
Total Nitrogen Deposition (kg N/ha/yr)
£ oMAcnoooro*.
1 _ 1 1 ! I 1 1 1 1
Total Nitrogen Deposition (kg N/ha/yr)
o to *>. en co o fo
1 1 1 ! 1 I I
Net annual N uptake <= 5% in 50 yr
4
(

\ XX
N^"^

)%

Total Nitrogen Deposition (kg N/ha/yr)
oro^cnoooroA
i i i i i i i 1 _,.
I I I I I I
0 2 4 6 8 10
Total Sulfur Deposition (kg S/ha/yr)
^Jet annual N uptake <= 5% in 250 yr


\ ^
D%

Total Nitrogen Deposition (kg N/ha/yr)
oioAcnoooMA
i i i i i i i i
slet annual N uptake <= 5% in 100 yr
(

X
}%

i i i i i i
0 2 4 6 8 10
Total Sulfur Deposition (kg S/ha/yr)
Net annual N uptake unchanging



*
3%


            I
           2
i
4
i
6
                    8    10

Total Sulfur Deposition (kg S/ha/yr)
\
6
i
8
 i
10
   2     4

Total Sulfur Deposition (kg S/ha/yr)
Exhibit B16. Percent of target population streams with  ANC  < 0 weq/L for  the
             Mid-Appalachian region at Year 2040.  Deposition  equals median
             deposition at Year 2020.
                                       B-20

-------
                                                               APPENDIX B: NBS PLOTS
      Net annual N uptake <= 5% in 50 yr
  14-
  12-
o

:f  8-
o
Q.



I  6"
 /I
I  4~

iz
—  O-_


H

   0-
                                 I
                                 10
      02468

        Total Sulfur Deposition (kg S/ha/yr)
                                              14-
                                              12H
                                            c
                                            o
                                            o
                                            o.

                                            3  6H

                                            0)
                                            O)
                                            o
4-
                                               2-
                                               0-
                                                    3.9 %
   0    2     4     6    8    10

    Total Sulfur Deposition (kg S/ha/yr)
 Exhibit B17. Percent of target population streams  with ANC  < 50 meq/L  for  the

              Mid-Appalachian region at Year 2040.   Deposition equals median

              deposition at Year 2020.
                                         B-21

-------
ACID DEPOSITION FEASIBILITY STUDY
Net annual N uptake <= 5% in 50 yr Net annual N uptake <= 5% in 1 00 yr
14-
>,
1 12-
Z
5 10-
c
o
'« 8~
o
Q.
Q 6"~
«
o* 4-
Z'~~
-3; /5
CO c.
J2
0-










" — ^\ ^X
NV ^
N.
>v
^2^
X











\
0%


14-
^| 12-
^10-
c
o
1 8~
a
3 6-
c
O) .*
o 4-
L_
Z._
— — o«.
t°
0-









^1
X
















\
0%
















1 1 I I I I 1 1 1 1 ! I
02468 10 02468 10
Total Sulfur Deposition (kg S/ha/yr) Total Sulfur Deposition (kg S/ha/yr)
Net annual N uptake <= 5% in 250 yr Net annual N uptake unchanging
14-
1 12-
z
I! 10-
c
o
1 8~
a
3 6-
c
CD
ITl
? 4-
i_
Z
nj 2~"
"Q
I —
0-










L



















\
0%




14-
"C"
j| 12-
.§10-
c
o
1 8-
o
Q.
3 eJ
c
CD
O) A
O 4-
u.
iz
"co 2—
o
V-
0-










*



















\
0%





















0    2    4     6     8    10

  Total Sulfur Deposition (kg S/ha/yr)
                                                i
                                                0
i
2
r
4
r
6
8
 i
10
                                                 Total Sulfur Deposition (kg S/ha/yr)
Exhibit B18. Percent  of  target  population  streams  with pH   <  5.0  for  the
             Mid-Appalachian  region at Year  2040.   Deposition equals  median
             deposition at year 2020.  pH is estimated from the empirical pH-ANC
             model.
                                      B-22

-------
                                                             APPENDIX B: NBS PLOTS
     Net annual N uptake <= 5% in 50 yr
O)
^


.i

'w

8.
Q)

8s
I
  14-



  12-



  10-



   8-



   6-



   4-



   2-



   0-
                 \
                 4
                           i
                           8
                         i
                        10
0246

  Total Sulfur Deposition (kg S/ha/yr)
                                                Net annual N uptake <= 5% in 100 yr
                                             14-
                                             12-
                                   g

                                   '55
                                   o
                                   Q.
                                   0)  -
                                   Q  6-


                                   0)
                                   o> *
                                   o  4-
                                           «  2H
                                              0-
                                                   o%
      T

      0
I

2
i

4
I

8
 I

10
                                                  Total Sulfur Deposition (kg S/ha/yr)
     Net annual N uptake <= 5% in 250 yr
  14-

*CT*


^ 12-
gio-

.0

OT  8~

I
O  6"*

0)


I  ^
2
——  rt^__


.2

   0-
 0%
                                i

                                10
    2468

Total Sulfur Deposition (kg S/ha/yr)
                                                 Net annual N uptake unchanging
                                             14-
                                             12-
                                           c
                                           o
1  
-------
 ACID DEPOSITION FEASIBILITY STUDY
     Net annual N uptake <= 5% in 50 yr
  14-
  12-
O>
o
  10-
&  6H
I  4H
z
«  2H
   0-
      i
      0
I      I
4    6
_I     |
 8     10
        Total Sulfur Deposition (kg S/ha/yr)
                                                Net annual N uptake <= 5% in 100 yr
                                              14-
                                              12-
o
S  8-
o
CL
<§  6-
c
o> .,
p  4-
                                              2-
                                              0-
                                                   1.6%
                                                 i
                                                 0
           l
           2
i
4
i
8
 I
10
                                                   Total Sulfur Deposition (kg S/ha/yr)
     Net annual N uptake <= 5% in 250 yr
                                                  Net annual N uptake unchanging
  14-
  12-
£.10-

o
1  8H
a
   6-
0

? 4-^
—  O
   0-
        1.2 %
      0    2    4     6    8    10

       Total Sulfur Deposition (kg S/ha/yr)
                                           I
                                           2
                                           O)
                                           c
                                           g
                                           "w
                                           a
                                           0)
                                           Q
                                           (U
                                           §>
  14-


  12-


  10-


   8-


   6-


   4-


   2-


   0-
                                                    0%
                                                  i     i     i     i      i     i
                                                 0     2     4     6     8    10
                                                   Total Sulfur Deposition (kg S/ha/yr)
 Exhibit B20. Percent  of  target  population  streams  with  pH  ^  6.0  for  the
              Mid-Appalachian region  at  Year 2040.   Deposition equals  median
              deposition at year 2020. pH  is estimated from the empirical pH-ANC
              model.
                                        B-24

-------
                                                             APPENDIX B:  NBS PLOTS
Net annual N uptake <= 5% in 50 yr Net annual N uptake <= 5% in 100 yr
12-
|p
^ 10-
z
en
^ 8-j
g
•'uj
O o
Q. 0

Q
c
o> 4-
cn
Z 2_
to
o
0-







a


(













*
\

~) °/o





















12-
i"
^ 10-
z
O)
^ 8-
g
OJ
O c
CL 6
0)
Q


-------
 ACID DEPOSITION FEASIBILITY STUDY
       Net annual N uptake <= 5% in 50 yr
Net annual N uptake <= 5% in 1 00 yr
12-
>.
1 10-
y
o>
c" 8-
g
55
8. 6-
o>
O
1 4-
2
z
1 2"
o
0-





£





*





14.6%





i i i
024





I I
6 8





12-
!£
1 10-
z
O)
c" 8-
.g
"35
0 R
CL 0
0)
Q
I 4"
t_
iz
^~ rt
"<5
,2
0-



•4 ><
N.
X. N.
N.
\^ >\.
^so ^*w^
\
6.7 %





i i i i i i i
10 0 2 4 6 8 10
         Total Sulfur Deposition (kg S/ha/yr)
   Total Sulfur Deposition (kg S/ha/yr)
      Net annual N uptake <= 5% in 250 yr
  Net annual N uptake unchanging
12-
|io-
z
O)
c" 8~
g
1 6-
0)
Q
0) 4-
cn H
9
—
^ 2-
!§
0-





J
N- • - - 	



b
\
5.9 %





1 1 1 1 1
02468


-


}







12-
|,0-
0)
c" 8~
g
'55
a e-
0)
Q
v
N ^

\
3.7 %





1 i i i i i i
10 0 2 4 6 8 10
         Total Sulfur Deposition (kg S/ha/yr)
   Total Sulfur Deposition (kg S/ha/yr)
Exhibit B22. Percent of  target population streams with  ANC <. 50 weq/L for the
             Southern Blue Ridge region at Year  2015.  Deposition equals median
             deposition at Year 2015.
                                        B-26

-------
                                                            APPENDIX B: NBS PLOTS
Net annual N uptake <= 5% in 50 yr Net annual N uptake <= 5% in 100 yr
12-
f,0-

O)
_^£
^ 8-
g
'tn
I 6-
o>
Q

Q
H 4-
O)
o
•^
1 2-
o
o-







L









>
\









0%






1 1 1 1 1 1 1 1 1
02468 10 024






i i
6 8






















i
10
Total Sulfur Deposition (kg S/ha/yr) Total Sulfur Deposition (kg S/ha/yr)
Net annual N uptake <= 5% in 250 yr Net annual N uptake unchanging
12-
I
| 10-
2
O)
V*
^ 8~
l_
g
55
O c
Q. O~
0)
Q
QJ 4-
0)
2
1 2-
,2
0-
-









4*
\
0%




12-
1
=c 10-
2
O)
j<:
^ 8-
g
55
8. 6-
0)
Q
o> 4-
cn *
o
^»
1 2'
o
H-
o-









A











>
\











0%




























    2     4    6     8    10
Total Sulfur Deposition (kg S/ha/yr)
    2468
Total Sulfur Deposition (kg S/ha/yr)
                                                                            10
Exhibit B23. Percent of target population streams with pH < 5.0 for the Southern Blue
            Ridge region at Year 2015.  Deposition equals median deposition at year
            2015. pH is estimated from the empirical pH-ANC model.
                                      B-27

-------
ACID DEPOSITION FEASIBILITY STUDY
osition (kg N/ha/yr)
> co o ro
i i i
Total Nitrogen Deposition (kg N/ha/yr) Total Nitrogen Dep
oroAoioooro 01x3*^0
i i i i i i i _ i i i i
Net annual N uptake <= 5% in 50 yr


\
0%
12-
1,0-
Z
O)
c" 8~
g
'55
0 R
Q. o



            I
            2
I
4
0     2    4     6     8     10

  Total Sulfur Deposition (kg S/ha/yr)
i
8
          4     6     8    10

Total Sulfur Deposition (kg S/ha/yr)
Exhibit B24. Percent of target population streams with pH < 5.5 for the Southern Blue
            Ridge region at Year 2015. Deposition equals median deposition at year
            2015. pH is estimated from the empirical pH-ANC model.
                                      B-28

-------
                                                           APPENDIX B: NBS PLOTS
Net annual N uptake <= 5% in 50 yr Net annual N uptake <= 5% in 100 yr
12-
I
I 10-
O)
? 8~
g
5
O c
Q. °~
0>
O
o> 4-
s
E 2-
 4-
o>
o
Ih.
1 2-
(0
,2
0-









A











V












o'%


















12-
>%
73
| 10-
z
O)
j^;
^ 8-
g
'55
a e-
0)
Q

-------
 ACID DEPOSITION FEASIBILITY STUDY
Total Nitrogen Deposition (kg N/ha/yr) Total Nitrogen Deposition (kg N/ha/yr)
ofu-p^oioooro o iv> A C5 03 o ro
i i i i i i i i i i i i i i
Net annual N uptake <= 5% in 50 yr
I

\. "4s
>
D%

12-
"S
=c 10-
2
CD
"c" 8-
0
55
O ft_
Q. O-1
CD
Q
o> 4-
0)
2
o
0-
i i i i i i
0 2 4 6 8 10
Total Sulfur Deposition (kg S/ha/yr)
vJet annual N uptake <= 5% in 250 yr
A
(


V
3%

tiii i
0 2 4 6 8 10
Irogen Deposition (kg N/ha/yr)
*. O> 00 O IV3
i i i i i
2 2_

-------
                                                               APPENDIX B: NBS PLOTS
      Net annual N uptake <= 5% in 50 yr
                                                 Net annual N uptake <= 5% in 100 yr
 I

 o

 '55
 o
 ex
 a>
 Q

 0)

 81
 CO
 o
12-



10-



 8-



 6-



 4-



 2-



 0-
         4.4 %
            ~T
             2
                 ~T
                  4
                        6     8

        Total Sulfur Deposition (kg S/ha/yr)
                                               12-
                                               10-
                                            c
                                            o
                                               8-
                                            8. 6-
                                            CD
                                            Q

                                            c
                                            Q>
                                            O)
                                            O
                                            OJ

                                            O
                                               4-



                                               2-



                                               0-
       \
      3.8 %
          i      i     I      I
    0     2     4     6     8     10

     Total Sulfur Deposition (kg S/ha/yr)
      Net annual N uptake <= 5% in 250 yr
                                                   Net annual N uptake unchanging
  12-



  10-



   8-
O  /~_

0)
Q



I4
£
o
    0-
         T
         3.4 %
      0     2     4     6     8     10

        Total Sulfur Deposition (kg S/ha/yr)
                                             o>
                                             c
                                             o
                                             O
                                             CL
                                             0>
                                             Q

                                             c
                                             CD
                                             O)
                                             O
12-



10-



 8-



 6-



 4-



 2-



 0-
                                                     A
                                                     2%
                                                        2     4     6     8     10

                                                    Total Sulfur Deposition (kg S/ha/yr)
Exhibit B27. Percent of  target population streams with  ANC < 50 meq/L for the
             Southern Blue Ridge region at Year 2040.  Deposition equals median
             deposition at Year 2020.
                                        B-31

-------
 ACID DEPOSITION FEASIBILITY STUDY
rogen Deposition (kg N/ha/yr) Total Nitrogen Deposition (kg N/ha/yr)
*> o> CD o N> o ro .&. o> oo o ro
i i i i i _ i i i i i i i
_ 2-
o
0-
Net annual N uptake <= 5% in 50 yr
t.


k

trogen Deposition (kg N/ha/yr)
— fc —A
*>• O> CO O IN3
CO
0
0-
i i i i i i
•0 2 4 6 8 10
Total Sulfur Deposition (kg S/ha/yr)
Met annual N uptake <= 5% in 250 yr
A
(


v.

12-
cr>
? 8~
g
o c
Q. 6-
0)
Q
c
IB 4-
o
I 2-
O
o-
Met annual N uptake <= 5% in 1 00 yr
L


h

i i i i i i
0 2 4 6 8 10
Total Sulfur Deposition (kg S/ha/yr)
Net annual N uptake unchanging
£
(


:*


      0     2     4    6     8    10

       Total Sulfur Deposition (kg S/ha/yr)
i
2
i
6
         4     6     8    10

Total Sulfur Deposition (kg S/ha/yr)
Exhibit B28. Percent of target population streams with pH < 5.0 for the Southern Blue
            Ridge region at Year 2040. Deposition equals median deposition at year
            2020. pH is estimated from the empirical pH-ANC model.
                                     B-32

-------
                                                            APPENDIX B: NBS PLOTS
Net annual N uptake <= 5% in 50 yr Net annual N uptake <= 5% in 100 yr
rogen Deposition (kg N/ha/yr)
*>• o> co o ro
i i i i i
Total Nitrogen Deposition (kg N/ha/yr) Total Nit
oho^cncDoro orvj
i

^
h
D%
i
Total Nitrogen Deposition (kg N/ha/yr)
_* _L
O I\D -li. O) CD O fO
1 1 1 1 1 1 1
I I I I I I
0 2 4 6 8 10
Total Sulfur Deposition (kg S/ha/yr)
Met annual N uptake <= 5% in 250 yr
4
C


^ •

rogen Deposition (kg N/ha/yr)
~-A _fc
*•. CD CD o ro
i i | i i
Z 2_
o
o-
L


v
\
3%

i i i i i i
0 2 4 6 8 10
Total Sulfur Deposition (kg S/ha/yr)
Net annual N uptake unchanging
A
(


V
]%


       i
       0
i
2
i
4
i
6
 i
10
         Total Sulfur Deposition (kg S/ha/yr)
02468

  Total Sulfur Deposition (kg S/ha/yr)
 i
10
Exhibit B29. Percent of target population streams with pH < 5.5 for the Southern Blue
            Ridge region at Year 2040. Deposition equals median deposition at year
            2020. pH is estimated from the empirical pH-ANC model.
                                      B-33

-------
 ACID DEPOSITION FEASIBILITY STUDY
12-
f,0-
Cn
S 8-
g
'55
a e-
 4-
0) H
2
1 2-
.2
0^
Net annual N uptake <= 5% in 50 yr



^

^jj^E^
^\^
x--a-
k
D%




i i i i i i
0 2 4 6 8 10
Net annual N uptake <= 5% in 100 yr
12-
f,0-
O>
.g
8 6-
Q. 0
CD
Q
 4-
O) ^
2
B
0-
Net annual N uptake unchanging



^
(




h
3%









      0     2     4     6     8     10
       Total Sulfur Deposition (kg S/ha/yr)
0    2     4    6     8    10
 Total Sulfur Deposition (kg S/ha/yr)
Exhibit B30. Percent  of  target  population  streams  with  pH  <  6.0  for  the
            Mid-Appalachian region  at  Year 2040.   Deposition equals  median
            deposition at year 2020. pH  is estimated from the empirical pH-ANC
            model.
                                      B-34

-------
        APPENDIX C

RANGE OF INFLUENCE OF EMISSIONS
FROM RADM TAGGED SUBREGIONS

-------
                                       APPENDIX C

                         RANGE OF INFLUENCE OF EMISSIONS
                         FROM RADM TAGGED SUBREGIONS
This report  includes the first extensive use of a
Eulerian  model to study source-receptor relation-
ships.  Source-receptor relationships  are  used in
this report to analyze changes in sources of deposi-
tion from implementation of Title IV of the Clean
Air Act Amendments of 1990 and  to  investigate
the effectiveness of several geographically targeted
emissions reductions strategies to achieving target
loads in sensitive receptor  regions.  The Tagged
Species Engineering Model1 was developed under
NAPAP to study such relationships. The Tagged
Species Model gives the Eulerian RADM modeling
system the  capability to identify, for  assessment
purposes, the concentration  and deposition fields
attributable  to specified SO2 emissions  sources in
the presence of  the full concentration  fields.  The
Tagged Model preserves the oxidant competition
across  space and time.  A tagging concept  is ap-
plied in which additional, identical mass conserva-
tion equations are solved for a portion of the sulfur
concentration  field  that  originates  from  specific
geographical locations within  the full modeling
domain.  This allows tagged concentration  fields
and tagged wet and dry deposition to be identified
and tracked  in the model separate from,  yet as por-
tions of, the  total sulfur chemical environment  that
is nonlinear  and that produces the complete con-
centration and deposition fields.

Calculations from the Tagged Species Model illus-
trate the  distances over which  an SO2 emissions
source can have an influence. The results from this
model permit the visualization of source  attribu-
tion. Emissions from a subregion (see Exhibit 19 for
the 53 tagged RADM subregions)  have a range of
influence is more than 1,000 kilometers. Typically,
the range  of influence of a subregion  extends out
to between 500 and 1,200 kilometers. The  dif-
ference in scale of influence is primarily  due to
meteorology.  A number of meteorological factors
influence the existence of dominant transport di-
rections and determine how sources of SO2 emis-
sions influence nearby regions. Key factors  are the
position of the jet  stream, which moves  storms
across the upper  Mid-West; the influence of the
Appalachian Mountains on winds  and  rainfall pat-
terns; the  Bermuda highs (stagnation) that move
Ohio River Valley emissions  in a counter-clock-
wise  direction; and the  ocean  and  Gulf Coast
weather that  produces  lighter  winds  and more
convective conditions, including a typically large
proportion of convective  clouds across the south-
ern states. Thus, the patterns and ranges of source
influence can  vary.  Models, such  as those in the
RADM system, help to interpret and  explain the
deposition at receptors of interest.

This  appendix contains maps of which show the
proportion of total annual sulfur  deposition  con-
tributed by each of the  53 tagged RADM subre-
gions in 1985 and projected for 2010 with imple-
mentation  of Title IV.
1  McHenry,  J.N.,  F.S.  Binkowski,  R.L. Dennis,  J.S.
  Chang,  and D. Hopkins.  1992. The tagged species
  engineering model (TSEM). Atmospheric Environment
  26A(8):1427-1443.
                                              C-1

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                                                C-2

-------
          APPENDIX C: RANGE OF INFLUENCE OF EMISSIONS
C-3

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                        o
~^
to
	 1
£
p
r—
e
o
CD
C£
s
O
O
o
ce
u
0.
k.
g§
o ^
(/I ^

-------
          APPENDIX C: RANGE OF INFLUENCE OF EMISSIONS
C-5

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                                                C-6

-------
          APPENDIX C: RANGE OF INFLUENCE OF EMISSIONS
C-7

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                                                C-8

-------
          APPENDIX C: RANGE OF INFLUENCE OF EMISSIONS
C-9

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                                      21
£
o
/«
1985 EMISSIOI
§
UJ
O)'
o
i
03
                                               C-10

-------
          APPENDIX C:  RANGE OF INFLUENCE OF EMISSIONS
C-11

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                                               C-12

-------
                  c  RANGE OF INFLUENCE OF EMISSIONS
C-13

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                                                                                                ^x
                                               C-14

-------
              NDIX C:  RANGE OF INFLUENCE OF EMISSIONS
C-15

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                                               C-16

-------
           APPENDIX C:  RANGE OF INFLUENCE OF EMISSIONS
C-17

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                                               C-18

-------
                  C:  RANGE OF INFLUENCE OF EMISSIONS
C-19

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                                               C-20

-------
           APPENDIX C:  RANGE OF INFLUENCE OF EMISSIONS
C-21

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                                               C-22

-------
           APPENDIX C:  RANGE OF INFLUENCE OF EMISSIONS
C-23

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                                               C-24

-------
          APPENDIX C: RANGE OF INFLUENCE OF EMISSIONS
C-25

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                                               C-26

-------
          APPENDIX C:  RANGE OF INFLUENCE OF EMISSIONS
C-27

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                                               C-28

-------
           APPENDIX O  RANGE OF INFLUENCE OF EMISSIONS
C-29

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDV
                                               C-30

-------
           APPENDIX C:  RANGE OF INFLUENCE OF EMISSIONS
C-31

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                                               C-32

-------
           APPENDIX C:  RANGE OF INFLUENCE OF EMISSIONS
C-33

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                                               C-34

-------
          APPENDIX C:  RANGE OF INFLUENCE OF EMISSIONS
C-35

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                                               C-36

-------
APPENDIX C:  RAN.GE QF^l
                                            EMISSIONS
C-37

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                               en
                                               C-38

-------
           APPENDIX C:  RANGE OF INFLUENCE OF EMISSIONS
C-39

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
z
o
t75
& 2 N
" ^ \


k 	
I ^ y~ _ — -^ i
^!r*\
(T
I^V,
^\^
*k
Jk
/X
/ V ^^^
                             i
                                   $

                                   B
                             ^ ^3
                             * i'

                             ^ a§

                             e a I
                             5g|
                                0> 2
                             ga
                             9
0»

10
                             I    i
                             §
                             OL
                                           C-40

-------
          APPENDIX C:  RANGE OF INFLUENCE OF EMISSIONS
C-41

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY

-------
                         APPENDIX C: RANGE OF INFLUENCE OF EMISSIONS
o
6
               C-43

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY

-------
APPENDIX O RANGE OF INFLUENCE OF EMISSIONS

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                               z  o

                               g  5 §
                               E  CM

                               00     m

-------
           APPENDIX C:  RANGE OF INFLUENCE OF EMISSIONS
C-47

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                                                O48

-------
,OF EMISSIONS
APPENDIX C: RANGE OF

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                                               C-50

-------
           APPENDIX C:  RANGE OF INFLUENCE OF EMISSIONS
C-51

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                                               C-52

-------
                           APPENDIX C: RANGE OF INFLUENCE OF EMISSIONS
o
£
           /•*
                                                          •' /-N*   ..
                                                            ,'   "^:-:--

                                                    K ^
                C-53

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                                               C-54

-------
               APPENDIX D

SUMMARY OF SCIENCE ADVISORY BOARD REVIEW
   AND PUBLIC COMMENTS AND RESPONSES

-------
                                       APPENDIX D
                   SUMMARY OF SCIENCE ADVISORY BOARD REVIEW
                        AND PUBLIC COMMENTS AND RESPONSES
                                   TABLE OF CONTENTS


PART!:  SCIENCE ADVISORY BOARD REVIEW	D-3

PART 2:  PUBLIC COMMENTS AND RESPONSES	D-6

       Process of Inviting Comments	D-6

       Response to Public Comment Notification	D-6

       General Comments on the Report	D-6

       Definition of Sensitivity  and Risk and Selection of Targeted Populations	D-6

       Geographic Coverage	D-10

       Monitoring and Data  Used in the Report	D-10

       Nitrogen Saturation	D-11

       Watershed Modeling	D-13

       Episodic Acidification	D-15

       Terrestrial Damage	D-16

       Uncertainty versus the Need for Standards	D-17

       Emissions Inventory	D-18

       Deposition Modeling	D-20

       Allowance Trading Program	D-21

       Benefits to Visibility, Human Health, Material, and Cultural Resources	D-22

       Costs	D-23
                                              D-1

-------
                                                   APPENDIX D:  SUMMARY OF COMMENTS AND RESPONSES
                        PART 1:  SCIENCE ADVISORY BOARD REVIEW
The  draft  Acid Deposition  Standard Feasibility
Study Report  to Congress was  reviewed by the
Acid Deposition Effects Subcommittee of the Eco-
logical Processes and Effects Committee (EPEC) of
the Science Advisory  Board (SAB) on  April 12,
1995.  The Subcommittee's review was followed
by  EPEC  review and  finally  by  SAB  Executive
Committee review on  June 29,  1995.  The SAB
concurred with the conclusion of the report that
the current state-of-the-science  with  regard to
acidic deposition effects on aquatic and terrestrial
systems does not support an acid deposition stan-
dard at this time.

The SAB made specific recommendations for im-
provement of the report and two broader recom-
mendations to the  Agency.  This section summa-
rizes the  SAB's  recommendations and  describes
the revisions made to the report in  response to
these recommendations.

  • The Study  should contain  a clear state-
     ment  of the  ecological  resources  and
     resource  endpoints to be protected by an
    acid  deposition standard and the level of
     protection desired.

     Congress  requested information on the fea-
     sibility and level  of  a  standard;  however,
     no guidance was  provided on the level of
     protection desired by Congress or the pub-
     lic.  A goal could be to protect resources at
    a  level measured at a previous point in
    time, to pristine conditions, or to another
     level which balances ecological protection,
    economics,  and  other  societal  values.
     Different  goals can  also be  developed for
     episodic  and  chronic  acidification.   The
     Executive Summary and Introduction have
    been revised to explain these options and
     how  standards  must be  related  to goals.
     Chapter 6 has been expanded to present a
     range of options for a standard or standards
     as a function of potential  goals considering
    the inherent uncertainty in relating  deposi-
    tion  levels  to  resource protection.   Cur-
     rently,  an acid deposition  standard level
    would be driven by aquatic  effects because
     of  the substantial  scientific  information
available relating deposition to aquatic ef-
fects. Much  less is known about the rela-
tionship between deposition and terrestrial
effects (i.e., forests) and current knowledge
suggests these effects are less severe than
aquatic  effects.   Chapter 2 has been ex-
panded  to more fully explain the relation-
ship between ANC and ecological vitality
(fish response) and why ANC was chosen
as the key modeled parameter.

The  Study should better characterize and
quantify the uncertainties in model projec-
tions of acidic deposition effects.

Section  2.5, which describes the selection
and  use of the  MAGIC  model, has been
expanded  to  more clearly  address the ra-
tionale for selection of the model,  uncer-
tainties  associated with its application  in
the Nitrogen Bounding Study (NBS), and
results from the Agency's external peer re-
view of the NBS.    As discussed in that
section, the  uncertainties  related  to that
study are presently best characterized using
a qualitative basis.  It must be recognized
that, as with any projection model, most of
the critical factors contributing to  uncer-
tainty cannot be quantified.   The limited
quantitative   error  estimates   related  to
model  projections  that  might be  made
would suggest an unrealistic confidence of
the model's present capabilities.  The text
emphasizes that the NBS model projections
should be used to provide estimates of the
direction  and magnitudes  of  change  to
different  possible  future  scenarios  of
deposition change.   The bounding nature
of the NBS has  been described  in  more
detail to  address the SAB's  request for
"confidence levels."  Additional qualitative
discussion has also been added to clarify
that  reductions  in  emissions  of  nitrogen
oxides from  motor  vehicle  controls and
implementation  of  the ozone attainment
provisions of Title I, although not  modeled
here because emission inventories were not
available at the time,  may result in long-
term decreases in nitrogen deposition.
                                               D-3

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
  • The Agency should  carefully review  the
     Executive Summary for two types of mis-
     statements: (1) summary conclusions  that
     are inconsistent with the wording of the
     body of the report,  and (2)  conclusions
     that  are  not  adequately  supported by
     available scientific evidence.

     The Executive  Summary has  been rewritten
     to more  closely  capture the conclusions
     and supporting  information  presented in
     the report.  Additional  uncertainty and tar-
     get population language has  been added to
     the Executive Summary.

  • The Study should clearly emphasize the
     distinction between sensitive (at risk) eco-
     logical resources (the  focus of the study)
     and the general populations of ecological
     resources that are more resistant or  fully
     resistant  to  adverse  impact  via acidic
     deposition.

     Section 2.3  provides  a detailed  introduc-
     tion to the concepts of risk and sensitivity.
     It describes  how these concepts apply to
     problems  of assessing  and projecting po-
     tential  receptor  responses in sensitive re-
     sources caused by acidic deposition expo-
     sure. Careful wording  is used to convey to
     the report's  readers  that most of the re-
     sources  having  low  sensitivity (i.e.,  high
     resistance) to potential  acidification effects,
     and deposition  levels  lower than  those
     necessary to produce a response should be
     viewed as being  at low risk.  Only those
     resources  having  high  sensitivities  to
     potential  acidification  effects that are ex-
     posed to  deposition rates sufficient to pro-
     duce the adverse sensitive responses should
     be considered as being at potentially high
     risk.  Section 2.3  also reviews different
     resource scales that can  be  used  to define
     receptor  resource sensitivity and that are
     important for  both  scientific and policy
     considerations. The section also provides
     an overview of the  most important  site-
     specific environmental factors  (e.g.,  soil
     type) that  locally influence the sensitivities
     of potential receptor resources.

  • The   Agency    should   utilize   acidic
     deposition  models   that   include   the
     biological processes controlling nitrogen
dynamics; MAGIC does not include these
processes.

As discussed in Section 2.5,  the Agency
recognizes that  biological  processes con-
trolling nitrogen cycling in watersheds are
important.   When the Nitrogen Bounding
Study began in  mid-1992  to  support this
Report   to   Congress,   no    adequate
combination of dynamic watershed  model
(that   included   controls  by   biological
processes and surface water chemistry) and
statistically based regional  watershed data
existed for the regional modeling of  effects
of  nitrogen  deposition   on   watershed
nitrogen  retention. The MAGIC model was
selected  at that time as the best base from
which to implement a "bounding" analysis
on possible acidification  effects in  target
surface  waters   under  various  possible
scenarios of sulfur and nitrogen  deposition
and over a  selection of possible times to
watershed nitrogen saturation.   The NBS
included  curves of  declining  watershed
nitrogen  retention to mimic, in a simplified
manner,  the possible influence of changing
biological   relationships   to    nitrogen
loadings  over time.  The peer reviewers of
the NBS  concurred that the approach used
was  likely the  best available at the time.
The Agency presently funds two university
efforts  to  develop  appropriate dynamic
watershed models of combined  sulfur and
nitrogen  cycling and effects.  These  model
include   representations   of    specific
biological  control  processes that  affect
nitrogen  cycling.

The Study should more clearly  character-
ize the scientific uncertainty regarding ter-
restrial ecosystem nitrogen saturation.

Both the draft and the revised report have
emphasized the considerable scientific un-
certainty pertaining  to nitrogen  saturation
at watershed and regional scales.  This un-
certainty is  the  reason, for example,  that
the Nitrogen Bounding Study  included a
selection of  modeled times  to potential wa-
tershed saturation ranging from 50 years to
never.    That  later case likely  applies to
most watersheds across North America, in-
cluding  many forested watersheds  poten-
tially sensitive  to acidic deposition.  The
                                                D-4

-------
                                                    APPENDIX D: SUMMARY OF COMMENTS AND RESPONSES
     revised  report includes additional discus-
     sion to emphasize this important point.

     Perhaps more importantly, the revised re-
     port  also  provides additional discussion,
     including citations from a selection of re-
     cently published research  reports, to show
     more  clearly that  nitrogen saturation can
     occur in sensitive watersheds and that spe-
     cific incidences of watershed nitrogen satu-
     ration appear to be occurring now in di-
     verse   watersheds  nationally.     These
     sensitive  watersheds   range   from  the
     Adirondack and Catskill Mountains in New
     York to the Middle Appalachian Mountains
     to the Great Smoky  Mountains  National
     Park  in  the  Southeast to Rocky  Mountain
     National Park  in  Colorado to  the  San
     Bernardino and San Gabriel Mountains  in
     California.    Thus,   watershed   nitrogen
     saturation   cannot   and  should  not  be
     characterized as a low probability, unlikely
     phenomenon  or   as  one  lacking  a
     reasonable basis for concern.

  » The Study should  identify and emphasize
     the importance of  environmental monitor-
     ing of deposition,  ecological indicators,
     and ecological endpoints as a parallel and
     complementary strategy to modeling in or-
     der to assess ecological resource risk from
     acidic deposition.

     A description of existing deposition and ef-
     fects  monitoring programs  and  discussion
     regarding monitoring  to assess  the effec-
     tiveness  of the control program  and en-
     hanced  monitoring needs  to  address  a
     standard or standards  have been added  to
     Chapter  5 of the report.

The SAB also made two recommendations to the
Agency that are broad in scope.  They are:
• The Agency  should  develop a conceptual
  framework  which identifies the  relevant
  science questions for the broader set of
  acidic deposition effects (human  health,
  ecological resource  health, visibility,  ma-
  terials  erosion,  atmospheric  chemistry,
  and other socioeconomic effects).  While
  Congress  explicitly directed the Agency to
  assess the implications of an acid deposi-
  tion standard for ecological resources,  a
  more detailed evaluation of human health
  and other possible benefits would be im-
  portant were the Agency to develop an
  acid deposition standard at some future
  time.

  This report identifies several categories of
  scientific   questions   that   would   be
  necessary to  address in order to determine
  and evaluate an  acid deposition standard
  or standards.  Other efforts are underway in
  the Agency to  identify scientific questions
  and the means to address them in areas
  which cover  effects of acidic deposition as
  part of science policy issues related to, in
  particular, effects of  ground-level ozone
  and fine particulate matter. Efforts are  also
  ongoing under the auspices of the National
  Acid Precipitation Assessment Program to
  gather   current   research   information
  pertaining to the causes  and effects of
  acidic deposition  and to identify remaining
  research gaps.

• Technical conclusions in EPA reports such
  as this one should be based on references
  from the peer-reviewed science-journal lit-
  erature.

  This report has been modified to include
  numerous additional  citations from peer-
  reviewed  journal  literature, particularly on
  the issues of ecological effects and envi-
  ronmental monitoring.
                                               D-5

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                       PART 2:  PUBLIC COMMENTS AND RESPONSES
PROCESS OF INVITING COMMENTS
EPA  announced the availability of the draft Acid
Deposition Standard Feasibility Study Report to
Congress for public comment in the Federal Regis-
ter on February 10, 1995 (FR Vol. 60 No. 28 page
7965).  Interested parties were notified that copies
of the report were available from the Office of Air
and  Radiation Docket Information Center.  Com-
ments were requested to be provided in writing by
April 1, 1995 and a phone number was provided
for those wishing additional information.  In addi-
tion  EPA mailed copies of the report to individuals
and organizations which had previously expressed
an interest in the Report.

A total of 490 copies of the draft report were dis-
tributed.

RESPONSE TO PUBLIC COMMENT NOTIFICATION
In response to its invitation for public comment,
EPA  received  17 written submissions.  A  break-
down, by general category, of respondents is given
below.
      EXHIBIT D-1.  COMMENJERS BY€A TECORY
       Federal Government          4

       State Government            3

       Industry Representatives       4

       Environmental Groups        3

       Academic Researchers        2

       Canadian Government        1
The  comments ranged from general  reactions to
the report and  its conclusions to specific section
by section critiques.  A synthesis of comments on
particular aspects of the report is provided below
along with responses by EPA.  The numbers in pa-
renthesis following comments refer to the source
of the comment as listed at the end of this appen-
dix.
GENERAL COMMENTS ON THE REPORT
Comments
A  number of  comments were  received  on the
overall presentation, content, and goals of the re-
port. One commenter stated that the report is well
written and  the  information presented is factual
and complete (17), another stated that  the report is
a good, comprehensive technical coverage of the
topic (1), and a third stated that  the report does a
good job in describing the factors involved in set-
ting and  implementing acid  deposition standards
(12).   Two  commenters criticized the report for
not adequately addressing or defining "feasibility"
(15,16). Several commenters stated that the report
did not adequately respond to the Congressional
directive in Section 404 of Title IV (Appendix B of
the Act) to describe and specify an  acid deposition
standard (4, 5, 6, 7, 15, 16, 17).

Response
The feasibility  of setting an acid deposition stan-
dard or standards is  dependent  on the scientific
basis for selecting a pollutant to  achieve the goals
of the standard and on the statutory basis and ad-
ministrative complexity of establishing and enforc-
ing a limitation on emissions of that pollutant. The
report  concludes that it  may be  feasible to set
deposition standards,  but that uncertainty, particu-
larly regarding the  scientific understanding of ni-
trogen, remains high  making it difficult to choose
an appropriate level of a standard.  Establishing a
standard or standards would require further guid-
ance from Congress and the public on the degree
of protection desired.

DEFINITION OF SENSITIVITY AND RISK AND
SELECTION OF TARGETED POPULATIONS
Comments
Three commenters (2,6,7) stated that the report did
not adequately characterize many of the most sen-
sitive  aquatic and terrestrial  ecosystems in North
America.  Two commenters (2,3) suggested that
sulfur and nitrogen target deposition loads should
be established for Class I  areas.  One (2) added
that the  report should divide the continent into
separate  resource regions of interest before at-
tempting an analysis  of  target  or critical loads.
                                              D-6

-------
                                                   APPENDIX D: SUMMARY OF COMMENTS AND RESPONSES
Another (6) stated that the report should present to-
tal  numbers  rather than proportions of resources
affected.   Another commenter (12)  agreed that
aquatic systems are the natural resources most at
risk from acidic deposition, but requested more
specific information on the numbers,  areas, and
proportions of the lakes, streams, and watersheds
affected or most at risk.

One commenter (2) stated that the report relies too
much  on the  assumption  that already-acidified
aquatic systems  (e.g., the  Adirondacks)  are the
most  'sensitive'  (2).    However,  this  same
commenter stated that ecosystems already altered
by air pollution should receive the most protection
by standards, as those systems have little available
buffering capacity and  require special protection
from changes due to anthropogenic activities.  This
commenter also suggested that sensitive  systems
that might  be  affected by nitrogen  deposition
should  include tundra ecosystems  of the  moun-
tainous West, the boreal forests of the northern tier
States and Alaska,  and other aquatic and terrestrial
ecosystems in other areas of the West.

Another commenter (4) agreed with the  report's
conclusion that the Adirondacks  should be  in-
cluded among the regions most sensitive to acidic
deposition in the United States. This commenter
also suggested that any acid deposition standards
that may be established should consider the wide
range of sensitivities  of  New York's aquatic re-
sources and varying response times that are neces-
sary to evaluate the impacts of varying deposition
rates.  This commenter stated that  lakes in water-
sheds with thin till and mounded seepage, which
contain biological  communities at risk, are the
most sensitive to acidification in the Adirondacks
and should be used as the  basis for establishing
target sulfur and nitrogen deposition loads.  Two
(6,16)  commenters requested  clarification  of the
description of lake conditions included in the tar-
get population and the  importance of naturally
acidic Adirondack lakes.

One commenter (15)  expressed concern that be-
cause of the large number  of small lakes in the
Adirondacks, the  report's projections  of  surface
water populations of  surface waters are  unrepre-
sentative of the larger general population.   This
commenter was also concerned that  the report
does not use the Congressional terminology criti-
cally sensitive  resources in  the   report.    This
commenter added that the  focus on ANC of 50
ueq/l in the report ignores other cutpoints for ANC
that could be used to  help distinguish  sensitive
from insensitive waters and added that the selected
goals  include  conditions  that "Mother Nature
cannot meet and therefore [are] not a  useful way
to make  environmental policy recommendations."
Finally, this commenter suggested that upper and
lower  bounds for risk  should  be presented and
other sources of acidification should be noted.

Several comments were received on  the use  of
ANC and pH to characterize aquatic resource sen-
sitivity. One commenter (6) urged EPA to consider
appropriate  watershed-based ecological endpoints
or biological indicators of health rather than  ANC.
Another  commenter (10) stated that many studies
show that aquatic biological damage begins  at pH
6.0, and  thus allowing  surface  waters  to become
acidified (ANC<0, pH<5.5) is not sufficiently pro-
tective.   By contrast,  another commenter (16)
stated that "selection of pH 6.0  appears somewhat
arbitrary". This commenter questioned how such a
standard  based on pH 6.0 would apply to a region
that has  many water  bodies with naturally  lower
pH and  indigenous fish populations that live  in
them  and recommended that  more emphasis be
placed on  responses  by  dissolved  aluminum,
which  has been shown often  to be the primary
cause of effects  to fish exposed to acidic condi-
tions.

Response
Appendix A of this report summarizes the exten-
sive, peer reviewed findings from NAPAP studies,
which identify the regions of North America found
to hold natural resources  having potentially high
sensitivities to effects from acidic deposition.  The
appendix also characterizes the nature  of the spe-
cific sensitive  environmental  resources within
these areas.  These findings, and more recent pub-
lished data,  are reviewed in Chapter 2.  The dis-
cussion in this chapter clearly  recognizes that a
continuous  range  exists of  sensitivities by  envi-
ronmental resources  to potential  acidification ef-
fects, ranging from extremely sensitive to insensi-
tive.  Specific environmental characteristics that
tend to increase these  sensitivities are tabulated
and described.  Information provided in Chapter 2
describes which of these  regions  are  at greatest
present and  potential future risk from acidic depo-
sition.  Indeed, ecosystems already altered by air
pollution have the greatest potential for benefiting
from additional reductions.  Chapter 6 projects the
magnitude of future  benefits of emissions reduc-
tions associated with the CAAA and benefits from
additional reductions beyond CAAA requirements.
                                               D-7

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
The discussion of potentially affected resources in
Chapter 2 has also been expanded.

Class I areas  can, by  definition, be described as
sensitive areas for which air quality deterioration is
not acceptable.  Given an adequate level of moni-
toring and assessment  data, Class I  areas could
serve as  potential targets for standards  setting ac-
tivities.  The necessary monitoring and analyses do
not currently exist for surface waters in  most Class
I  areas.   However, pursuant to the CAAA, EPA is
currently initiating development  of regional haze
regulations for Class I areas.  Effects in  Class I  ar-
eas are discussed further under "Geographic Cov-
erage".

The concepts  of sensitivity and risk are  sometimes
poorly distinguished.  As used in the report, sensi-
tivity refers to the potential response relationship
that a receptor, such as a lake or a fish population,
has when affected by a stimulus such as environ-
mental  stress  produced   by acidic  deposition.
Highly sensitive  lake and  stream  systems,  for ex-
ample, have poor abilities to resist adverse chemi-
cal changes produced by acidic deposition.  Alka-
linity or acid neutralizing capacity (ANC) is a
readily   measured   indicator of surface  water
sensitivity,  acidic status, the remaining ability to
buffer   against   future  acidification,  and  the
likelihood that biological  communities  inhabiting
that  aquatic  system will  respond to changes in
acidic addition.  Measures of ANC, thus, provide
useful  information on  many aspects  of  aquatic
systems.   In  turn, risk  is  the probability  that the
receptor  of  concern  (e.g.,  a  lake) will  attain
sufficient exposure to  a stimulus (e.g., a  loading
rate  of acidic deposition)  such  that its sensitivity
(e.g., its buffering capacity)  is  exceeded  and  an
adverse  response  is  produced  (e.g., ANC   is
depleted over the short- or long-term).

Resources can have diverse combinations  of both
sensitivity and risk.  When a resource's sensitivity
is high,  it does not necessarily follow that  the risk
to the resource is high.  A  resource with high sen-
sitivity to acidification, for example, would have a
low risk of adverse effects whenever past, present,
and probable future acidic deposition  is  unlikely to
meet or exceed its sensitivity threshold. Therefore,
when assessing the need for acid deposition stan-
dards,  it is necessary to  consider both resource
sensitivity and risk.

An analysis of sensitivity over scales of nations and
continents requires  a broadly  applicable  and
meaningful measure to help assure that the analy-
sis is both comprehensive and practical.  ANC is
such a measure. As discussed in the report,  it pro-
vides meaningful  regional,  national,  continental
scale  information on  both the chemical and bio-
logical status of aquatic systems.  Other measures,
such as growth and reproductive responses by sen-
sitive resident species, while very important for the
individual systems, provide information of progres-
sively shrinking significance as the scale of  analy-
sis expand across regions, nations, and continents.
Further, potential  sensitivities of most individual
species to changing surface water acidity are un-
known to science. Beyond that, the potential sen-
sitivities and the relative ranking for each species
also shifts between individual waters, over seasons
within the same water, and across the various geo-
graphic regions of concern. These shifts are  due to
a great diversity of site-specific and life-history
considerations related to these species. Cataloging
the extensive matrix of these relationships for indi-
vidual  species across  the acid-sensitive regions of
North America is beyond the scope of this report.

This complexity of potential  resource sensitivity
concerns again points to ANC as the appropriate
indicator measure to  be used in analysis such as
presented in the report.  First, ANC is  the best
overall measure of the ability  of  surface  water,
ground water, and watersheds to buffer against ad-
verse effects from deposited  acidic compounds.
Second, ANC has  a direct relationship to pH and
other acidity related water quality  parameters  in-
cluding aluminum. At lower pH levels,  the bio-
logical availability of many toxic metals, including
aluminum, increase.   As pH decreases,  so does
ANC.  As explained in Chapter 2, for the  sensitive
regions considered, an  ANC of 50 ueq/l  approxi-
mates pH 6.5 and an ANC of 0 ueq/l approximates
pH 5.3.

Close examination of the Congressional history re-
lated to expectations for this Report, as discussed
in Chapter  2,  reveals  that  Congress  followed a
1983 National Academy of Science report in mak-
ing a distinction between 'critically sensitive' and
'sensitive' surface waters as those having alkalinity
< 40 ueq/l  and < 200 ueq/l.  Specifically, the Re-
port of the Committee on Environment and Public
Works, United States Senate (November 20, 1987,
pages 131 and 132) states:

  ... 40  units (microequivalents  of  alkalinity
  per  liter) of buffering capacity is considered
  very low. resources with  buffer capacity be-
  low  40 ueq/l are extremely sensitive to acidic
  deposition. Resources with buffering capacity
                                                D-8

-------
                                                    APPENDIX D: SUMMARY OF COMMENTS AND RESPONSES
  between 40 ueq/l and 200 ueq/l are consider-
  ately  moderately  sensitive  and  subject  to
  damage at existing deposition rates. Alkalinity
  is  chosen  from among  other  possible  indi-
  cators because it is  a  parameter that is now
  routinely  measured  by surveyors; it  can be
  measured with reasonable accuracy by cur-
  rent technologies; alkalinity  is  relatively in-
  sensitive to contamination;  and alkalinity is
  less subject to daily and seasonal variability
  than other measures of water quality such as
  pH. . . . The report is to include an identifica-
  tion of the sensitive (<200 ueq/l alkalinity) and
  critically sensitive  resources (<  40 ueq/l alka-
  linity) ....

As discussed in  Chapter 2, measurement aims for
alkalinity and ANC are the  same,  and many scien-
tists consider the terms to be synonymous.  Recent
assessment of surface water response have focused
on ANC<0  and <50 ueq/l.  Considering  natural
environmental variability  and  projection  times
used  in the  report, an alkalinity value of 40 ueq/l
and  an ANC value of 50 ueq/l are  essentially
equivalent.  This is especially true as related to the
specific intent of Congress to address needs related
to "critically sensitive  resources."   Further, more
recent research  has shown the earlier alkalinity
value of 200  ueq/l used  by MAS to no longer  be
the most  applicable definition of surface water
sensitivity. For example, NAPAP's  1990 Integrated
Assessment reported (page 279):

     In  lakes  with  historically   higher  ANC
     [greater than 50  ueq/l],  ANC reductions
     have not generally been observed except in
     regions  with very  high  levels of  acidic
     deposition  such  as  southern  Sweden and
     near  Sudbury in  Ontario, Canada.    Al-
     though chronic  decreases  in ANC are less
     likely to have occurred in waters with ANC
     greater than 50  ueq/l,  these waters may be
     affected by episodic ANC and pH depres-
     sions (sudden, short-term decreases  in pH
     and ANC related  to rainstorms and snow-
     melt) with  consequent deleterious effects
     on aquatic biota.

Therefore, the analysis in  this  report focused on
how changes  in acidic deposition loadings could
alter acidified surface  waters (ANC<0 ueq/l) and
those most  critically sensitive  to  chronic effects
from continuing  acidic  deposition (ANCS50 ueq/l).
That  is, the  report uses both  ANC<0 ueq/l and
ANC<50  ueq/l  in  considering potential  acidic
deposition effects. The first of these is used primar-
ily to characterize those waters that have or may
become chronically acidic, causing serious long-
term adverse consequences to biological commu-
nities inhabiting these systems.   The second is
used to characterize those waters  most sensitive to
potential long-term acidification and those particu-
larly sensitive to acutely toxic impacts from epi-
sodic acidification, which can cause serious short-
term  adverse consequences  to  resident  aquatic
communities.

The report specified qualitatively  that the numeric
values discussed are based on the specified target
populations of lakes and stream reaches and quan-
titatively identified the representativeness of target
populations to the total number of systems in each
case-study region.  Additional quantification of the
target populations have been  included in  the re-
vised report.  Further, it is important to emphasize
that the legislative language specifically indicated
that the report should focus on sensitive and criti-
cally sensitive resources. In doing this, the report
also tries not to encourage  or misguide the reader
into  using the information presented beyond  the
defined target population.

Following an overview drawn from various NA-
PAP  reports  and  other referenced  information,
Chapter 2 presented  three resulting conclusions.
The third of these is, "to protect aquatic resources
in sensitive watersheds from  effects of  long-term,
chronic acidification, a general  goal is to maintain
the pH of sensitive lakes above pH 6.0-6.5 and in-
organic monomeric aluminum  below 30-50 ug/l.
To protect these resources from potential effects of
episodic, acute acidification, surface water ANC
should be maintained above 50 ueq/l." This was
intended to  provide guidance on how a  general
goal to protect surface waters might appear.  Other
considerations would have to be  given for estab-
lishing possible watershed and water specific goals
related to protecting  sensitive  species  and other
special biological considerations.  Then, a caution
is provided that,   [w]hen  establishing  protection
goals  and objectives  for  sensitive  aquatic  re-
sources, this effort certainly must  include recogni-
tion and allowances that pH levels less than  6.0
and ANC less than  50 ueq/l occur in some natu-
rally acidic surface waters, and that levels of  pH
less than 6.0 can occur naturally in some locations
accompanying periods of episodic stormwater and
snowmelt runoff. That  is,  the specific environ-
mental objectives of any acid deposition standard
should accommodate the natural ranges of chemi-
cal qualities  occurring in waters in the environ-
ment. Furthermore,  they may be  designed to pro-
                                               D-9

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
tect those special biological communities evolved
to inhabit naturally acidic surface waters.

Regarding the comment on the need to bound risk,
the  report strongly  makes the point that time to
watershed nitrogen saturation includes consider-
able uncertainty  for most  individual watersheds
over both regional and nation scales. As such, the
report and the Nitrogen Bounding Study depict the
projections between the various times modeled to
potential nitrogen saturation as providing upper
and lower bounds of the proportion of resources at
risk.  Thus, the name, "Nitrogen Bounding Study."

GEOGRAPHIC COVERAGE
Comments
Two commenters (2,3) suggested that the draft re-
port  inadequately addresses areas having sensitive
Class I designations, especially in the West.  One
(3) suggested that,  while  significant effects from
depositions are suspected in the Pacific  Northwest,
monitoring to document  effects  to sensitive re-
sources are scarce.

Response
The report emphasizes discussion of resource sen-
sitivity in the Adirondacks, Mid Appalachians, and
the  Southern Blue  Ridge Province because  these
three regions  have been  the most extensively
studied.  Also,  as noted  in  the  report, existing
information supports the hypothesis that these are
the three regions  holding the largest proportion of
the natural resources at greatest present risk of ad-
verse effects from acidic deposition.  Support for
this hypothesis is based, in part, on the findings
from NAPAP's SOS/T Report 9 summarized in Ap-
pendix A of the  report.   No new  regional-scale
analysis  has  appeared  since  to   counter  this
hypothesis.  Instead, a 1994 review, supported by
EPA  and summarized in the  report, also supports
this hypothesis.   However, the  lack of observed
acidic deposition effects in other regions, particu-
larly the  West, could be due  to the scarcity of ap-
propriate  monitoring  data  from this area.  One
commenter (3) also recognized this data gap. Bet-
ter and more widespread regionally based moni-
toring data from the West  are needed  before this
hypothesis can be seriously challenged.   To pre-
sent  a  more thorough  picture of the status of po-
tential  effects from  acidic deposition in the West,
additional site-specific information has been  in-
cluded in the report to illustrate apparent examples
of ongoing nitrogen saturation and acidification ef-
fects in select mountain regions of Colorado and
California.

The report provides little direct information or as-
sessment on potential effects to areas designated as
Class I under the CAAA, because relatively sparse
monitoring information is available for these areas.
Data are scarce for these areas primarily because
of their inaccessibility and  because of restrictions
on  monitoring activities stipulated both  by  the
CAAA and by the managers of  these areas to
minimize human  disruption.   Special  status is
given to Class I areas  under the CAAA by Con-
gress. The intent of this report, however, is to pro-
vide  regionally  representative information.   In
some cases, the data and conclusions presented in
the report regarding relationships of acid-sensitive
resources in   non- Class   I  areas have  similar
characteristics to  Class I areas in terms of  acid
impact potential.   Therefore, while specific  data
from most Class I  areas are lacking, data from the
sensitive non-Class I areas discussed  in the report
may implicitly provide surrogate  information on
which to address concerns in certain Class I areas.
The  use  of information from "surrogate"  areas to
address  potential  effects  in  Class I  areas  is a
practice previously established by the  Forest Serv-
ice to assess potential deposition effects. This  is il-
lustrated, for example, by  the  Forest  Service de-
velopment and use of the Glacier Lakes Ecosystem
Experiments Site in Wyoming.

MONITORING AND DATA USED IN THE REPORT
Comments
One commenter (4) noted that the  NBS study and
other parts of the report relied almost entirely on
data  collected during EPA's  National   Surface
Water Survey, which did not include lakes smaller
than 4 hectares in area.  The commenter's concern
was  that these lakes,  which represent about 40
percent of the more than 2,900 Adirondack lakes
and  ponds,  would be among  the  first  aquatic
receptors to be affected by acidic deposition.  This
commenter   also  suggested   that   long-term
monitoring data from  Adirondack lakes  showing
changes  in sulfate, nitrate, and ANC should be
included in the report.  A  second commenter (7)
questioned the extensive use of NAPAP and NSWS
results in the report, which are suggested to  have
limited value and questionable findings, rather
than presenting more recent research findings.

A third commenter (6) suggested  that the report
should include more recent data on  the Adirond-
acks available from the Adirondack  Lake  Survey
                                              D-10

-------
                                                    APPENDIX D: SUMMARY OF COMMENTS AND RESPONSES
Corporation and the New York Department of En-
vironmental  Conservation  (6).  This  commenter
also suggested that EPA "propose  and evaluate a
grid system for deposition  loading  and response
monitoring for the Adirondacks and the other sen-
sitive areas... Such a system is essential to tracking
emissions ...".  Another commenter (2) questioned
why more of the references cited  did not include
those from 1993-1995 to reflect more current  in-
formation, especially regarding nitrogen deposition
effects.

Response
A large body of recent and older literature was  re-
viewed during preparation of the  report. Findings
presented in  the more recent reports did not sig-
nificantly alter major regional scale findings pro-
vided in earlier reviews. The newer literature did
contribute useful details for refining the scientific
understanding regarding the effects of acidic depo-
sition, particularly  regarding the effects from nitro-
gen. The report has been  expanded  to include
discussion of  and  citations for many  newer
research efforts. This has not changed any of the
conclusions presented in the report.

Several   questions  were  raised  regarding target
populations for the modeling analyses.   As noted
under in the  under "Geographic  Coverage", lower
limits on the area  resolution for the maps used  to
establish  the peer-reviewed, statistically based
sampling design used  by the  NSWS caused only
lakes greater that 4 ha in size to be included in the
analysis.  It  is true  that smaller lakes likely are
often more sensitive to the effects of acidic deposi-
tion.   But necessary  watershed  data  to  model
acidic deposition  effects on such lakes over  re-
gional scales do not exist.

It is important to recognize that, for any group  of
streams,  lakes and watersheds sampled and mod-
eled, there will always exist extremes in character-
istics.  Sampling and analysis  could focus on wa-
tersheds  at either end of the extreme.  The disad-
vantage to such focusing is that the target popula-
tion may have marginal utility for policy purposes.
The target  populations of  lakes,  streams  and
watersheds sampled  and analyzed by the NSWS,
NSS and DDRP have received extensive peer and
policy review both before and after these projects.
These populations  have been found to be relevant
to concerns regarding  potential adverse effects  of
acidic deposition.   Focusing modeling efforts  on
systems  of greater vulnerability  would  lead  to
results showing that  higher percentages  of the
target  population  is at risk,  although  the same
number of streams would  be affected.  The water-
shed selection procedures are clearly described in
the description of the NBS and the explanation in
the report has been expanded.

Using  lake  or streams survey data collected in
years following the NSWS would provide  limited
benefit to the  modeling analysis.  The  model is
calibrated for a specific year (in this case, the year
of the  NSWS sampling)—calibrating the  model to
data collected in  a later year or  years,  would
provide results that would not be substantially dif-
ferent  for long-term future projections.   In  fact,
such a process undertaken for only a sub-sample
of lakes having survey data after the NSWS Phase I
survey could actually be counterproductive in that
it would introduce greater procedural uncertainty
into the modeling process.

To  maintain a  degree of scientific consistency
across  the various  regional-scale analyses com-
pleted  by the Agency,  the NSWS lake set contin-
ues to  be used.  Some regional  analyses have fo-
cused on  assessing information contained in  sub-
sets of the original NSWS data that represent the
most sensitive resources.   For example, the DDRP
and the NBS projects assessed potential responses
by the most sensitive lakes and  streams  extracted
from progressively refined subsets  of the  NSWS
data.   That  is, the  environmental  modeling  per-
formed during these  projects, as presented in this
report, were designed to project  the  effects  of
changing potential deposition  loads of sulfur, and
sulfur plus nitrogen, respectively, on the  more sen-
sitive surface waters within each of the modeled
regions. In doing so, the responses studied by both
DDRP and NBS tend to include resources  having
sensitivity characteristics more similar to those se-
nsitive resources (e.g.,  smaller  lakes) that some
commenters considered to be ignored.

EPA agrees that effects and deposition monitoring
are necessary to assess future  impacts and  trends.
A  discussion of these  issues has  been  added to
Chapter 5.

NITROGEN SATURATION
Comments
Twelve commenters  addressed,  either directly or
indirectly, issues related to the watershed process
of nitrogen saturation.  Indirect comments focused
on watershed modeling issues relating to nitrogen
saturation, will be discussed  with  the  responses
addressing comments on that topic.  The following
                                               D-11

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
section  responds  to issues  provided  by  seven
commenters  specifically related to the nitrogen
saturation process.

One commenter (2)  praised the report's emphasis
on the  importance of nitrogen  deposition effects
on ecosystem  integrity.   This commenter  noted
that deposition of nitrogen in the West is princi-
pally  "buffered" ammonium nitrate, which  is not
"acidic  deposition."   This commenter suggested
that the report should discuss the interaction of ni-
trogen depositions with insects and disease in pro-
ducing  forest effects  and altering nitrogen  reten-
tion.  Further, this commenter requested that more
consideration be given to potential nitrogen depo-
sition effects  in  the tundra  ecosystems  of  the
mountainous West, boreal forests of the northern
tier states, and recent data from California.  This
commenter further suggested  that the role  of in-
sects and  disease in exacerbating the effects of at-
mospherically deposited nitrogen on terrestrial  and
aquatic  systems, e.g., gypsy moth in Shenandoah
National Park and adelgid in Great Smoky Moun-
tains  National  Park  appear to relate to the com-
bined effects of stress from excess nitrogen deposi-
tion and insect infestation.  Insect stress in forests
reduces apparent times to nitrogen saturation by
impairing the ability of affected trees and forests to
incorporate available nitrogen.

Two commenters  (13,16) suggested that nitrogen
saturation is a hypothesis without definitive  proof,
especially in  North America.  They suggested that
since  nitrogen  saturation  is  unlikely  to  occur
where forests are disturbed, the potential for nitro-
gen saturation  are low outside major parks  and
wilderness areas.  One (16) suggested that the re-
port  incorrectly characterizes nitrogen  saturation
as "when" not  "if" events.   This commenter also
suggested that  the report asserts that the  knowl-
edge  of nitrogen  cycling  is  poorly understood,
when considerable  knowledge exists about  the
process  because of its use as a plant fertili2er. Both
commenters provided various suggestions on  im-
proving the  discussion regarding the nitrogen cy-
cle, especially a clear acknowledgment that deni-
trification can be an important mechanism for ni-
trogen loss from terrestrial systems.

In contrast,  two commenters  (4,6) suggested  that
recent research shows that estimated time to wa-
tershed  nitrogen saturation in the Adirondacks  is
likely less than  50 years.  These commenters noted
that some recent research results  suggest that ni-
trate concentrations  are  increasing in Adirondack
waters,  while sulfate concentrations are decreasing
and ANC levels appear largely unchanged.   One
(4) acknowledged that while the extent of nitrogen
saturation in Adirondack watersheds  or the  time
scale for this phenomenon is not fully understood,
that considerable progress has  been made in un-
derstanding the nitrogen saturation process.

Another  commenter (8) suggested that additional
emphasis should be placed  on  the  statement that
the time needed for nitrogen saturation  is ". . . very
speculative.   We  do not  really know the  time
needed for  nitrogen saturation to  occur".   This
commenter  went on to note  that  historical evi-
dence  of nitrogen  saturation,  as  indicated  by
drainage loss of nitrogen from Hubbard Brook Wa-
tershed, for example, shows that watershed loss of
nitrate can display unanticipated  shifts in both di-
rections between high and  low  loss  rates.   This
commenter  also cautioned that it should not  be
implied that all watersheds will  eventually become
nitrogen  saturated.   It  is likely that many water-
sheds will not because land  use changes can  have
considerable influence on the potential for  nitro-
gen saturation.

Another commenter (12) suggested that concluding
anything about the benefits from changes in ni-
trogen  deposition is premature.  This  commenter
suggested that  retention and loss of  nitrogen in
ecosystems is regulated almost exclusively by bio-
logical processes.  The commenter also expressed
the view that natural and man-caused  changes on
the land can significantly affect  these processes.
Noting that  the report uses results from European
studies to help establish that  nitrogen saturation
can occur, this commenter claimed that deposition
of nitrogen  in  Europe  can be many times that
found in the United States.

One commenter (17) suggested that  the  time to
saturation scenarios  used in the report should  be
applicable to a significant  component of  the Ca-
nadian aquatic resources, and  the  results for the
Adirondacks  probably  reflect  the Canadian  situ-
ation quite well, particularly in southern  Ontario
and Quebec.  Further, this commenter noted that
the uncertainties raised with respect to the effect of
nitrogen  deposition are similar to those raised by
the recently  completed  Nitrogen Assessment  by
the Canadian Environmental Conservation Service.

Response
As conveyed  in  the diversity of  comments re-
ceived, much uncertainty and little consensus exist
regarding either the process of  nitrogen saturation
or the times required for watersheds  to reach ni-
                                               D-12

-------
                                                    APPENDIX D: SUMMARY OF COMMENTS AND RESPONSES
trogen saturation.  This diversity underscores these
two sources of existing uncertainty, as stated in the
report, that prevent the  Agency from recommend-
ing at this time either specific national or specific
regional deposition standards.

Despite this  uncertainty and debate, the Agency
concludes that the bulk of the scientific  evidence
suggests that the  process of nitrogen saturation is
not a "hypothesis lacking proof," as suggested in
some public comments.   Instead, watershed nitro-
gen saturation is a process that can occur at some
times in  some watersheds.  The  report has been
revised to include additional evidence to support
this  conclusion.  Determining  whether  nitrogen
saturation had occurred or is occurring in any in-
dividual watershed requires site-specific  monitor-
ing data.

Symptoms of nitrogen saturation and resulting sur-
face water acidification effects do  not and likely
will  not occur  in  most  watersheds.  Many forest
management actions,  natural  disturbances,  and
low  sensitivity  to potential acidification  impacts
will prevent these effects in most forest systems, as
suggested by several commenters.  Yet, various
acid-sensitive watersheds,  when subjected to high
loadings of nitrogen from the atmosphere, or from
other sources, can become nitrogen saturated and
acidify surface waters into which they drain.  Ad-
ditional discussion of these important relationships
has  been added to the  report  to  clarify these
points.

Further, some comments  reflect  different defini-
tions used for nitrogen saturation within the scien-
tific community.  As now  further  explained in  the
Nitrogen Bounding Study (NBS) section of the re-
port, some investigators use the term nitrogen satu-
ration to indicate the first point where the supply
of nitrogen compounds  from the atmosphere  ex-
ceeds the demands for these compounds by water-
shed plants and soil microbes. In comparison,  the
definition for watershed  saturation used in the dis-
cussion of NBS follows  that used by an  early re-
searcher of nitrogen saturation and that used dur-
ing the Direct Delayed  Response Project (DDRP)
regarding sulfur saturation.  That  is, the NBS uses
nitrogen saturation to mean the  point when  the
watershed retains less than 5 percent of all depos-
ited  nitrogen on a net annual basis.  Under this
definition, any year's deposition may be stored for
release during a later year, as long as at least 95
percent of the mass  of  nitrogen deposited within
any given year is released. The important differ-
ence in these two definitions is that the first refers
generally to a time occurring relatively earlier in
the nitrogen saturation process, when relatively lit-
tle  nitrogen is being lost from the watershed.  In
contrast, the definition used for the NBS refers to a
later time in the saturation process, when nearly
all of the annually deposited nitrogen is discharged
from the watershed.

Other specific issues regarding nitrogen saturation
raised by the public commenters are addressed in
the revised report.

WATERSHED MODELING
Comments
Twelve comments were received relating to water-
shed modeling issues  including comments on the
dynamics of nitrogen  in watersheds and potential
times to watershed saturation.  Comments on the
latter two points are summarized and discussed  in
the previous  response.  This section  summarizes
and responds to comments relating directly to wa-
tershed modeling. Seven commenters (5, 8, 9, 12,
13, 15, 16)  questioned  the  applicability  of the
MAGIC model for assessment because of  its per-
ceived limitations.

One commenter (8)  stated that  most versions  of
MAGIC do not include biotic  processing and that
this limitation should be identified,  especially  in
watersheds where nitrogen is important.   This
commenter  also noted that  a  new version  of
MAGIC, which includes some nitrogen transforma-
tions, is under development.  Another commenter
(12) claimed  that the  NBS is severely limited be-
cause the basic mechanisms of nitrogen  cycling
and processes affecting the rate of nitrogen uptake
and release are not included in the model and that
the only use of MAGIC  is to bound possible out-
comes.  Two  commenters (15,16) stated that the
NBS needs to be subjected  to  additional  peer
review, beyond the EPA peer-review process.

Another commenter  (13) suggested that,  if the
process of nitrogen saturation is real, terrestrial sys-
tems will differ in the capacity to both assimilate
and lose nitrogen and that some portions of a sin-
gle watershed could  reach saturation  at one time
and other portions at another.  This commenter
and another  (16) questioned  how differences  in
time to nitrogen saturation over scales of individ-
ual watersheds would affect the  results from  NBS
and what implications this would have on  poten-
tial standards development.

One of these commenters (16) requested that spe-
cific technical information be  supplied to  help  in
                                              D-13

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
evaluating the Nitrogen Bounding Study (NBS) for
use in evaluating the feasibility of an acid deposi-
tion standard:

  *  In  relation  to the  total  number of water-
     sheds  in  the  modeled regions, how may
     watersheds were modeled in each region
     by NBS,  how were the modeled water-
     sheds chosen, and what are the distribution
     of watershed characteristics?

  *  In acknowledging that NBS included an at-
     tempt to evaluate uncertainty in time to ni-
     trogen  saturation,  better clarification was
     requested on  quantifying other uncertain-
     ties inherent in the MAGIC model  to allow
     judging the  significance of  differences
     among results presented within the various
     NBS plots exhibited.

  *  Providing information regarding the impli-
     cations of  compounding  the  uncertainty
     from the RADM projections when  these re-
     sults were used in the NBS model.

This  commenter and another (15) also requested
more information on the model selection process.
Both claimed that the  ILWAS model  is the best
model  available, because, unlike  MAGIC, it con-
tains a lake  component  that can simulate episodic
acidification. One of these commenters (15) sug-
gested  that  the report  should have discussed  the
results  of tests  comparing  MAGIC's  predictions
with real data  and with  predictions using  other
models.   This  commenter also  questioned  the
projections presented in the report from the Direct
Delayed  Response   Project  (DDRP).     The
commenter  suggested  that  using  projections of
surface  waters in the Northeast studies by DDRP
to represent effects on broad classes  of  surface
waters produces incorrect conclusions, and that
the report lacks discussion of sensitivities for the
surface waters modeled.

Another commenter (9) claimed that both the NBS
and the DDRP investigations  inappropriately com-
bined responses of watersheds in the Mid-Appala-
chians that  have low sulfur-retention characteris-
tics with those  that have high retention.   This
commenter  further  claimed  that  the  generaliza-
tions from this  aggregating were  misleading, and
that effects occurring in the most  sensitive Class I
areas were effectively ignored in the model output.
This commenter suggested further that a  more
focused assessment would have lead to different
conclusions regarding this region and that DDRP's
implementation of MAGIC could not be calibrated
successfully to streams of the region's Class I areas.
The  commenter referred to recent  model results
(not  yet peer reviewed or published) that suggest
that  reductions in sulfur deposition of  70 to 80
percent are needed to prevent additional stream
acidification in this region.

Response
Additional discussion has been added to  Chapter 2
that  explains that NBS  was designed to  address
one of the major goals of this study, to project the
dynamic regional-scale  effects on surface water
chemistry from sulfur and nitrogen deposition into
areas containing potentially vulnerable  lakes and
streams.  The Chapter 2  discussion also  notes that
when this effort began in 1992, limited options ex-
isted for  model choice and regional sets available
for modeling.  Because of these limitations, it was
possible only to investigate the potential  bounds of
nitrogen deposition effects on surface water chem-
istry in combination with process modeling of ef-
fects of sulfur deposition.

In comparing the ILWAS and the MAGIC models
for regional-scale modeling, EPA determined that
although both models  provide  comparable  re-
gional results,  MAGIC could be successfully cali-
brated for significantly more watersheds than IL-
WAS.  Thus, use of the MAGIC model in the NBS
was the chosen approach.   Chapter 2 explains the
selection process in further detail.

The DDRP modeling and data, completed in sup-
port of the  National Acid Precipitation Assessment
Program  (NAPAP), formed the  basis of subsequent
NAPAP modeling using many alternative scenarios
of atmospheric sulfur deposition.  Documentation
of the DDRP and subsequent NAPAP modeling ac-
tivities based upon DDRP can  be found  in several
publications cited in Chapter 2 of the report.

As,  noted in the report,  the NBS did not simulate
how nitrogen deposition might alter watershed re-
tention of nitrogen; no combination of model and
regional data available in early-mid 1992 could do
that  at the regional scales required for the Acid
Deposition  Standard Feasibility Study.  Rather, the
NBS  illustrated what  would  be the  result  on
surface water chemistry at regional scales //"certain
scenarios  of   changes  in  watershed   nitrogen
retention were to occur.   In  so  doing, the NBS
effectively  bounded all  reasonable possibilities of
such effects.   This is the first time this has been
accomplished.
                                              D-14

-------
                                                   APPENDIX D: SUMMARY OF COMMENTS AND RESPONSES
As discussed under "Monitoring and Data Used in
the Report," the target populations of surface wa-
ters included in the NSWS, NSS, DDRP, and now
NBS have  received extensive peer and policy re-
view  both  prior to and following those  projects.
They  have been  found relevant for concerns re-
garding potential adverse effects of acidic deposi-
tion.  In the Mid-Appalachians, for example,  the
model was calibrated successfully to low ANC
streams in  the  region to represent adequately  the
target population defined  by the sampling frame.
If the analysis were to focus on lakes  and streams
of potentially greater (or lesser) vulnerability,  the
regional utility of the modeling would be lost.

The  report  now  includes additional  information
regarding uncertainty in the NBS modeling exer-
cise.  The  expanded discussion notes that uncer-
tainty in any computation or modeling analysis is
the potential difference  between  the calculated
value (under a set of conditions)  and the  "true"
value.  Quantifying levels of actual error in model
projections, such as those made within the NBS (or
for any watershed acidification model used in a
predictive sense),  is not possible  because future
effects  are  not  yet  observable.      Certain
components  of  model uncertainty (e.g.,  sample
uncertainty, input uncertainty)  may be estimated
quantitatively,   but  other   components   (e.g.,
aggregation  uncertainty,  structural   uncertainty)
cannot.  These  latter sources of uncertainty  likely
overwhelm  the former.    The computation   of
"confidence  limits"  around  model  projections
based only on estimations of those few sources of
uncertainty  that  can  be  quantified  can  be
misleading,  because they  underestimate the total
uncertainty involved.

For example, one public  comment requested that
the effect of uncertainty in RADM projections be
propagated through the watershed modeling.   In
fact,  such an estimate of propagated uncertainty
would be very  minor.  RADM simulations were
used until the year 2010;  after that, prescribed  ex-
plicit  scenarios  of  deposition  were used.  Any
computation  of effects of  uncertainties  of  the
RADM projections or the  MAGIC  projections  for
the period up to year 2010 would not be meaning-
ful within the context of the study.

The purpose of the model runs performed for  the
Acid Deposition Standard Feasibility Study was to
test the sensitivity of potential watershed responses
to varying scenarios of (1) nitrogen deposition, and
(2) watershed transition to nitrogen saturation  in
relation to  projected effects of sulfur deposition.
In this light, and as described in the report,  the
utility of the watershed simulations is to examine
direction  and  magnitude  of  projected  relative
changes, rather than to focus on explicit numerical
estimates of percentages of target populations.  In
addition, the Agency is funding the development
of further  nitrogen modeling capabilities  in  the
MAGIC  model.    The Agency  is  also  funding
additional    watershed   nitrogen    modeling
development as well as a watershed manipulation
project  to allow  comparing short term  model
projections with effects of elevated deposition over
watershed scales.

The  NBS methodologies were peer reviewed fol-
lowing standard EPA peer review procedures. The
number of peer reviews requested and received on
the report exceeded requirements.  The revised re-
port  describes the peer review process and its re-
sults.

EPISODIC ACIDIFICATION
Comments
Five  commenters (2,4,6,13,16) addressed episodic
acidification.  One (2) agreed  that  the  report
places  correct  emphasis  on the  importance  of
acidic episodes in low ANC aquatic systems. Two
commenters (4,6) reported that recent research has
shown that acidic episodes have great potential  to
affect the survival of aquatic  organisms.  One (6)
emphasized that nitrogen plays a very significant
role  in acidification  episodes and suggested that
the report should increase its  discussion of this re-
lationship. Both (4,6) recommended that a stronger
emphasis be placed  on reductions  in NOX emis-
sions because most snowmelt episodes are acidic
due  to high nitrate levels.   One (4) emphasized
that any acid deposition standard must account for
the episodic events resulting primarily from spring
snowmelt that affects the most  sensitive Adirond-
ack water classes.  This commenter claimed that,
based on information contained in the report (3.5
times the number of chronically acidic Adirondack
lakes could be affected by episodic acidification
and potentially 43 percent of  the lakes in the Adi-
rondacks may be  acidic in 2040 after CAAA im-
plementation if 50 years to nitrogen saturation is
assumed), then 100 percent of  the Adirondack
lakes may be subject to episodic acidification  in
2040.

A  fourth (13) commenter  suggested  that most  of
the work documenting effects related to episodic
acidification has  occurred in small  watersheds,
which may or may  not reflect impacts at larger
                                              D-15

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
landscape scales.   This commenter additionally
suggested that there is no direct evidence linking
atmospheric deposition and episodic acidification.

A fifth commenter (16) questioned the source and
the basis for quantitative estimates oi  episodic
acidification.  This commenter also claimed that
reducing  rates of  "deposition  can  lead  to an
increased number  of  systems that  experience
episodic acidification  if the  number of systems
switching from chronic  to  episodic  acidification
exceeds the number no longer being acidified epi-
sodically."  The commenter suggested lhat more
detailed  discussion is needed on  other biogeo-
chemical  sources,  snowmelt dynamics, and in-
stream use of nitrogen that can potentially affect
receiving  water effects produced during episodic
acidification events.  The commenter further stated
that  documented cases  of  episodic  acidification
are limited to first order ephemeral streams, i.e.,
small streams with  no tributaries that have flows
only during portions of most years.

Response
Several of these comments, particularly  those re-
lated to the importance  of  nitrogen to episodic
acidification  events, are discussed  extensively in
the report.  In response to these comments,  this
discussion has been expanded to more clearly ex-
plain this phenomenon.  The focus of much past
research, as noted in the report, regarding poten-
tial  effects  from  episodic  impacts occurring in
smaller,  lower order streams  is appropriate  be-
cause these are the aquatic systems  that tend to be
most sensitive to potential  acidification effects.
Such streams often  provide critical  spawning  and
rearing habitat for fish populations, including acid-
sensitive fish populations.

EPA  agrees  that  in  certain  areas, as  effects
accompanying  chronic acidification due to sulfur
deposition are reduced, overall  effects due to
episodic acidification  would  likely continue to
impair the water quality, but the  extent of these
effects would likely be reduced because ieducing
the chronic sulfur effects also decreases potential
episodic effects as well.  Such a hypothetical re-
sponse would need to be evaluated through  sur-
face water monitoring.  The potential for  such  a
response  should  be examined  in  detail through
appropriate  modeling to project whether the need
for future deposition reductions should be acceler-
ated.

The  comment that documentation of episodic
acidification is limited to observations in first order
ephemeral streams is not correct. All streams, for
example,  included in the EPA  study  "Episodic
Acidification   of  Streams  in  the  Northeastern
United States," as discussed  in the Report, were
perennial, i.e., had continuous flows.  Included in
this study were first and second order streams and
a pond outlet in the Adirondack Mountains; first,
second, and  third order streams  in  the  Catskill
Mountains; six second order streams in north cen-
tral  Pennsylvania.    All of  the  surface  waters
showed characteristics  of  episodic  acidification.
Similarly,  in  an   article  published  in   Water
Resources Research during  1993 by O'Brien et al.
and   cited   in  the   report,   found   episodic
acidification  effects  associated  with  increased
water concentrations of sulfur, nitrogen, or both in
first-, second-, and third-order perennial streams.

TERRESTRIAL DAMAGE
Comments
Three commenters (13,15,16) felt that the  Execu-
tive  Summary oversimplified the relationship  be-
tween  acidic deposition  and  terrestrial  effects.
One commenter (13)  suggested that the report in-
adequately explains that very  limited data exist
that  support  the occurrence of degradation of soil
properties or stress to forest systems due to acidifi-
cation.  This  commenter also claimed that  acidic
deposition appears to be a relatively minor factor
affecting the current health  and  productivity  of
forests in the United States.   Two  commenters
(15,16) stated that the report did not document the
sensitivity of  high elevation  red spruce  in  the
Northern  Appalachian  acidic  deposition.   Both
stated that the report should acknowledge  that  a
serious and unresolved  debate  continues whether
acidic deposition causes any forest effects.

One commenter  (16) claimed that  the potential
productivity  and  economic  benefits  of   acidic
deposition in producing fertilization effects  in  "90
percent" of the managed and non-sensitive forests
in the United States would  be lost if sulfur and ni-
trogen deposition were reduced.  This commenter
also claimed that the apparent trend of reducing
base cation deposition may be contributing to the
overall  acidification  response.  This commenter
also suggested that discussion of acidification ef-
fects on soils should be revised.

A fourth  commenter (17) was concerned that sus-
tainability of forest landscapes and effects on bio-
diversity were inadequately  discussed  in  the re-
port. In noting that forests most at risk from acidic
deposition in eastern Canada are within southeast-
                                               D-16

-------
                                                    APPENDIX D: SUMMARY OF COMMENTS AND RESPONSES
ern Ontario, southern Quebec, and the Maritimes,
this commenter stated that this was  unfortunately
not recognized in the study, "especially since the
US NAPAP Review [SOS/T  Report  18] identifies
acidic deposition as a possible cofactor in maple
decline in these forests.  In  addition, acidic fog is
increasingly being  recognized as  a  factor in the
decline of coastal  red spruce and white birches
along the portions of the Gulf of Maine and Bay of
Fundy coasts."

Response
The Executive Summary has been revised to clarify
the current state of knowledge regarding the  rela-
tionship between acidic deposition and terrestrial
effects.  Chapter 2  noted that, besides a few new
references  cited therein, most technical  informa-
tion  on potential direct and indirect effects  from
acidic deposition to soils, trees, and forests were
based  on analyses completed and  reported by the
NAPAP. In addition,  Appendix A  summarizes the
relevant NAPAP State of the Science  and Technol-
ogy (SOS/T) reports.

The report does not claim widespread soil damage
or forest decline due to acidic deposition. Chapter
2  includes the statement  "(w)hile  control studies
quantitatively  link changes  in soil chemistries to
tree and other plant responses, similar studies  link-
ing acidic deposition effects  in  nature  remain
inconclusive."    Further,   Chapter  2 cites   the
conclusion in NAPAP's SOS/T 16 report, "(t)he vast
majority of forests in the United States and Canada
have  not declined."   The  report  does note that
certain high elevation forests (i.e., high elevation
red spruce, primarily in northern Appalachians and
the Northeast) are sensitive to and may be affected
by acidic  deposition, from  particularly  acidic
cloud water,  interacting with other air pollutants.
This  follows  from  the NAPAP 1990 Integrated
Assessment  (page  45) finding,  ".  .  .  there is
evidence  -  from  controlled   experiments  for
alteration of plant nutrition,  cold hardening, and a
wide  variety  of physiological processes  -  of red
spruce  being  affected by  high  levels  of  acidic
deposition  in  cloud water."  That  report states
further, "Localized areas of forest decline (i.e.,  high
elevation red spruce)  do occur, as a result of the
combined action of multiple stress factors,  and in
those areas where high deposition  amounts occur
in combination with  other stress  factors,  acidic
deposition can  increase the total stress level on the
forest  system."  The  report does  not make  any
claims of further potential damage  beyond what
was  presented and   discussed  in  the  NAPAP
reports.  Potential  effects  to  Ontario maple are
among those noted in the body and Appendix A in
both the Draft Report and its successor.

UNCERTAINTY VERSUS THE NEED FOR STANDARDS
Comments
Eleven comments were received in this area rang-
ing from urging EPA to  set standards immediately
to statements  that there is no demonstrated need
for  additional  emissions  reductions.    Several
commenters (4,5,6,7) urged  EPA  to recommend
deposition standards for the Adirondacks in the re-
port. These commenters and one other (16) argued
that the report does  not fully respond to the Con-
gressional directive to "describe the nature and nu-
merical value of a  standard  or  standards that
would be sufficient  to protect sensitive and criti-
cally aquatic and terrestrial resources".  They add
that the Adirondacks is the region most  strongly
impacted by acidic  deposition and that further re-
ductions in sulfur  and  nitrogen  deposition  are
needed to  protect sensitive aquatic resources in
this region.  One commenter (7) suggests that "EPA
implement deposition standards  in the range of 4
kilograms of sulfur per hectare and 7.5 kg of nitro-
gen per hectare on an annual basis" for the Adi-
rondacks.   Another commenter  (6) argued that a
deposition  standard  should account for  episodic
acidification, define sensitive lake resources by in-
cluding physical and biological communities, and
include specifics on  the  interaction  on  nitrogen
and sulfur to protect sensitive resources in the Adi-
rondacks.  This commenter and another (4) suggest
that  a sulfur  deposition  standard  of 3.5 -  3.7
kg/ha/yr would be required.

Two other commenters (10,11) supported develop-
ing acid deposition  standards to protect resources
such  as  those  in  eastern   Canada.     Both
commenters felt that setting such standards is fea-
sible and that  at a minimum critical loads should
be established.  Another commenter (11) recom-
mended that EPA establish regional critical loading
goals for wet sulfur  (sulfates) now and that critical
loading goals  for nitrogen deposition be set once
ongoing nitrogen saturation studies are completed.
This commenter recommended that these regional
critical loads be  set as goals not standards  in order
to assist decision makers in determining whether
adequate environmental protection was occurring
and to foster a dialogue among the public, elected
officials,  and  scientists over the adequacy of the
reductions   mandated by the  Clean  Air  Act
Amendments.  Further, these goals could serve as
                                               D-17

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
an interim step  in developing effects-based stan-
dards for sensitive regions. Another comrnenter (6)
stated that sufficient data are available to establish
critical loads for the Adirondacks.

Other  commenters  (12,13,15,16)  felt  that the
Agency does not have enough information or un-
certainties are too large at this time to recommend
deposition standards.  One commenter (12) stated
that the report "should clearly delineate that a ni-
trogen  deposition standard is not feasible at this
point" and that there is little risk in  waiting to as-
sess the full impact of the Clean Air Act Amend-
ments of 1990 because sulfur and nitrogen deposi-
tion will be decreasing in response to mandates of
the Act.  Another  commenter  (13)  believes that
EPA  does not have the authority to establish and
implement  deposition standards under  sections
108, 109, and 110 of the Act.

Response
In the report EPA concluded that an acid deposi-
tion standard or standards should provide adequate
protection for sensitive  resources  at  the greatest
risk and that the most sensitive resources at the
greatest risk (i.e., receiving the highest levels  of
deposition)  are  lakes and streams in the Eastern
United  States.  The subset of these resources  at
greatest  risk are Adirondack lakes.   The report
concluded  further  that   effects associated with
acidic deposition are  minimized as  pH and acid
neutralization capacity (ANC) are kept high.  Gen-
eral goals set forth are to keep pH above 6.0 and
ANC above  50 ueq/l.  The report does not set criti-
cal or target loads,  but does provide the scientific
basis upon which critical  loads can  be developed.
Scientific uncertainties regarding  times to water-
shed nitrogen saturation  make choosing the ap-
propriate  level of a standard or standards difficult
at this time. After reviewing this report, EPA's Sci-
ence Advisory Board unanimously concurred with
this conclusion.  EPA  has revised Chapter 6 of the
report to  provide deposition  levels  for  stlfur and
nitrogen which  may  be  necessary to  achieve a
range of  environmental  goals under different as-
sumed times to nitrogen  saturation.  The summary
states that under some assumed times  to  nitrogen
saturation, deposition  reductions  expected from
implementation  of the CAAA would achieve envi-
ronmental goals; under other assumed times to ni-
trogen saturation additional reductions in both sul-
fur and nitrogen deposition would be  necessary.
Further, scientific understanding of time to nitro-
gen saturation would provide needed insight.
Although some have suggested that geographically
targeted  acid  deposition  standards  could  be
established under existing statutory authority, no
definitive  determination  has  been  made at this
time.  The  purpose of this  report  is to provide
Congress with the scientific information  necessary
to determine the feasibility  (i.e., scientific basis,
implementation and costs issues) and desirability
of setting a standard or standards.  EPA would then
look   to   Congress   for   direction   regarding
establishing a standards or standards, including the
environmental goals that such  standards should
achieve.

EMISSIONS INVENTORIES
Comments
Several comments  were  received  regarding the
completeness of the baseline emissions inventories
for NOX and SO2 and the assumptions made in de-
veloping  the  projection  inventories.     Four
commenters  (4,12,15,16)  noted that  compliance
with the ozone provisions of Title I of the CAAA
will   result  in  significant   decreases   in  NOX
emissions from stationary and  mobile sources  in
ozone nonattainment areas and in  the  Northeast
Ozone   Transport  Region.     Two   of   these
commenters also noted that Title II of the Act will
result in decreases in  NOX emissions from mobile
sources nationwide.  These  commenters believe
that these reductions should have been factored in
to  the   NOX  emissions   reduction   scenario
constructed for the report.   Another commenter
(15)  was  concerned that the baseline  emissions
inventory does not  adequately characterize VOC,
anthropogenic SO2, and  natural SO2 and  NOX.
This   commenter  stated  that  VOC  inventories
currently  being  prepared  by  States for  ozone
nonattainment areas should have been included in
the baseline  inventory.  Another commenter (16)
stated  that the 1 to 2  percent of  total nitrogen
emissions estimated to be from natural  sources
seems low in light of  significant  NO  emissions
from soils.

One commenter (14) took issue with the statement
made  in  the report  that  utility SO2  emissions
should remain near the 8.95 million ton cap for
the years beyond 2010.  This commenter estimates
that utility SO2 emissions will decline to less than
50 percent of 2010 levels by  2040 and that utility
NOX emissions will decline to around 50 percent
of 2010  levels by 2040.  The  SO2 estimates are
based on the assumption that  all "SIP" units will be
over 60 years old by 2040 and will be retired or
                                               D-18

-------
                                                   APPENDIX D:  SUMMARY OF COMMENTS AND RESPONSES
will  have undergone reconstruction  or  modifica-
tion. Such units or their replacements will be sub-
ject to New Source Review (NSR) which requires
the application of Best Available Control Technol-
ogy  (BACT).   BACT currently requires  emissions
reductions of 95 percent which translates to  emis-
sions rates of 0.1-0.3 lb/106 BTU. Thus SO2  emis-
sions from these plants will have dropped from the
cap rate of 1.2 lb/106 to less than a quarter of that
rate. Since electricity demand is expected to dou-
ble with new units  having to achieve BACT the
combination  of increased  capacity  and reduced
emissions rate will result in more than halving na-
tionwide utility SO2 emissions in 2040.  For analo-
gous, but less definitive reasons, the commenter
projects that  NOX emissions from utilities will fall
by 50 percent by the year 2040.

Regarding NOX emissions, the commenter believes
that  given the long  time  horizon (2040), current
average electric utility NOX emission rates of 0.5
to 0.6 Ibs NOX per million Btu, and the availability
of NOX control technology capable  of meeting a
rate of 0.1 to 0.2 Ibs  NOX per million Btu, that it is
probable that a 50 percent NOX  reduction relative
to 2010 will be achieved by 2040.

Response
Several ongoing sulfur and nitrogen related efforts
within EPA will affect acidic deposition loadings.
These  include periodic review  and analyses of
NO2 and SO2 national ambient  air  quality  stan-
dards,  and  the efforts by states and  Northeast
Ozone Transport Commission  under Title I to  re-
duce NOX emissions to facilitate compliance with
the ozone standard.  Other ongoing efforts include
potential regulatory action towards a fine panicu-
late standard (PM25), which would likely result in
significant sulfate and nitrate reductions, and a re-
gional  haze rulemaking effort under Title I.   Com-
pliance with  the provisions of Title I  of the CAAA
will  result in  significant decreases in NOX  emis-
sions from stationary sources in  ozone nonattain-
ment areas and in the Northeast Ozone Transport
Region and compliance with  the Title  II mobile
source provisions will  result in a decrease in na-
tionwide NOX emissions in the short term.

Deposition modeling requires that emissions be re-
solved spatially and temporally.  Thus, source-spe-
cific emissions for point sources  and area source
emissions by category and location must be devel-
oped.  For SO2, the baseline inventory for 2010 is
based on a comprehensive utility model based on
full implementation  of Title IV  that incorporates
regional growth, energy availability, plant retire-
ments, control device  requirements, and  cost-ef-
fective compliance choices.   EPA  considers this
inventory to be a realistic projection of emissions
in 2010.  On the  other hand, implementation of
Title I by the states is ongoing and region-specific
inventories for NOX (as well as for VOCs) are not
yet available.  Therefore, the 1990  interim inven-
tory was used as the baseline to estimate the im-
pact of reducing NOX emissions in future years.

As noted in the report, estimates  of natural  emis-
sions of SO2 and NOX are not as well character-
ized as anthropogenic emissions.  These emissions
are, however, have been found to be less impor-
tant than man-made sources.  Natural sulfur emis-
sions are estimated to  be  6 percent of  anthropo-
genic emissions and natural emissions of nitrogen
compounds are estimated  to be 1 to 2 percent of
total nitrogen emissions. Further research  is  being
conducted  by EPA and others to better character-
ize natural  emissions.

A projection model to forecast electric utility SO2
emissions beyond 2010 at  a sufficient level of dis-
aggregation necessary to support deposition  mod-
eling does  not exist.  Even if such  a model were
available,   the uncertainty and  subjectivity that
would be involved with specifying the geographic
location of unplanned new capacity needed  to re-
place generation from  retiring units and to meet
new electricity demand (i.e., between 2010 and
2040), would limit the usefulness of such a  long-
term projection.  In fact, any emissions projections
made that far into the future  (2040) are based on
so many economic assumptions, that the confi-
dence  in  such  projections would  be very  low.
With regard to the comment that  all  "SIP" units
will have retired or undergone significant modifi-
cation by 2040 and thus would have dramatically
lower SO2  emissions, the Inventory of Power
Plants  in  the United  States  1993  published by
DOE's Energy Information Administration  states
"Electric utilities have  found it to be  more  eco-
nomically feasible  to modify existing electric gen-
erating units than  to construct new ones. ...unit
modifications are not subject to the new  sources
laws,  which  would  be  applicable under the
...(CAAA) of 1990". For purposes of the report, it
was assumed  conservatively that SO2  emissions
would  remain constant beyond 2010 —  the last
year sufficiently  desegregated  projections  were
available.   To the extent  that SO2  emissions de-
cline beyond 2010, deposition levels will  be cor-
respondingly reduced.
                                              D-19

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
As noted in the report, electric utilities contribute
about 33 percent of NOX emissions, highway vehi-
cles about  33 percent, and other sources together
about 33  percent.  EPA projects that total NOX
emissions will decrease between  1990 and 2000
and then gradually  increase thereafter. The near-
term decrease in NOX emissions will be produced
by implementation  of Titles  I, II,  and IV  of the
CAAA. After 2000 the growth in highway mileage
driven will begin to overtake emissions reductions
and will control the rate of increase in nationwide
NOX emissions. Thus, any postulated decrease on
electric utility NOX emissions after 2010 may serve
to moderate the rate of increase of total NOX emis-
sions.  Because reliably predicting  regional NOX
emissions after 2010 would be subject to large un-
certainties, the modeling in the report is based on
the assumption that NOX  emissions will  remain
constant from 1990  through 2040.

DEPOSITION MODELING
Comments
Three organizations provided  detailed comments
on the use of RADM  for deposition and visibility
modeling.  Two commenters questioned the ability
of RADM to accurately model  nitrogen deposition
(13,15).   They stated that the  lack of plume treat-
ment  in the model  leads to instantaneous mixing
and too rapid oxidation of NO.  One commenter
(16)  stated that the ability of  RADM to  predict
deposition  in future years  is  limited because the
model results have  only been calibrated to meas-
ured deposition for  the years  1985 through 1990.
This  same  commenter disagreed  with the state-
ment  made in the report that the uncertainties in
RADM lead to a  spread of about 10 percent in
predicting changes in deposition due to changes in
emissions.  This commenter stated that the variabi-
lity for wet deposition  of sulfate alone is a factor of
2 and that  the inability to measure dry deposition
of sulfate accurately leads  to even greater  uncer-
tainties.   This commenter noted that significant
uncertainty is  associated  with the  location  of
"unplanned" sources in 2010 and stated that these
sources represent up to 50 percent of year 2010
emissions.

Response
Accurate Modeling of Nitrogen
It is  correct that the instantaneous  mixing in the
large  RADM grids will cause  NO  to be converted
more quickly in the model than in the real  world.
The instantaneous  mixing likely means that more
nitric acid (HNO3) is formed in the model close to
strong  NOX  emissions  sources  than should  be
formed.  Comparison  of a  20-km  RADM version
with the 80-km  version shows this effect, where
the high HNO3  levels spread more downwind in
the 20-km model.  The result is a displacement of
maximum HNO3 formation by  about an 80-km
grid cell, causing the maximum to be closer to  the
emissions source. The difference between the two
models disappears by the time half of the source's
nitrogen  is deposited, roughly 300-500 km.  Thus,
RADM predictions are expected to show structure
in the deposition pattern associated with variations
in emissions density that is  more  enhanced than it
should be.    However, such  an  enhancement of
deposition near strong NOX sources  is not appar-
ent in  comparisons against wet deposition  meas-
urements. This is, in part, because nitric acid is a
secondary pollutant and transport distances  are
large,  500-800  km, smoothing  out  the  overall
deposition field for nitrogen.  In  other words,  the
effect of the too  rapid  conversion is hard to see in
comparisons  with  large-scale  regional   fields.
Nonetheless,  RADM is expected  to somewhat  ac-
centuate spatial  differences due  to spatial  differ-
ences in emissions.  The model can be used to  ex-
plore differences in the spatial location of deposi-
tion associated with different emissions  sectors if
the spatial separations, are large, as  they are for  the
eastern  United States.   Any quantitative conclu-
sions drawn from such an analysis  would need to
be carefully caveated.  However, in this report  the
RADM results are used  to draw attention to  the
main qualitative  result that  the influence of utility
and mobile emissions on  nitrogen deposition is
quite spatially distinct, and, hence, one will need
to be alert to this when one  considers possible sen-
sitive areas that need l.o benefit from future reduc-
tions in NOX emissions.   Based on  the grid-size
sensitivity analysis, the errors that  result from  the
large grid-size of the  RADM model do not pre-
clude the model  from providing valuable insight at
the qualitative level.  The RADM results show that
the dominance of certain emissions sectors is suffi-
ciently separated geographically that this  knowl-
edge must be factored into any future analysis of
NOX emissions reductions.  This is  fully consistent
with the use of the RADM results  in this report.

Ability to Predict Future  Deposition
The criticism that the ability of RADM to predict
future years is limited because it is  calibrated to
1990 misses  the central point underlying the  de-
velopment of RADM in the NAPAP research effort
                                              D-20

-------
                                                   APPENDIX D: SUMMARY OF COMMENTS AND RESPONSES
and incorrectly associates RADM with a regression
model, which it most definitely is not. RADM was
developed to be a state-of-the-science model that
incorporates  the  basic  physical  and  chemical
processes necessary  to  describe  the  transport,
transformation  and  fate  of acidifying  species.
RADM was  developed as a  result of the concern
that  simple,  parameterized  (fitted)  models,  no
matter how  advanced,  might  not  be  able  to
reliably  predict future year  deposition  resulting
from  emissions  changes  (either  increases  or
decreases) due to the  effects of oxidant limitation.
These effects cause nonlinearity in the system and
can cause the "effective" parameterizations of the
linear models  to change  as  one extrapolates to
conditions quite different from those upon  which
the  parameterizations  are   based.    Therefore,
advanced,  comprehensive  models,  like RADM,
that  can internally  account  for   change and
dynamically  compute the resultant effect on the
processes are the models  with  which  one can
more reliably extrapolate  to  conditions  quite
different  from those of 1990. This can not be done
with the  simple, parameterized (fitted) models.

Estimated Uncertainty of Prediction of
Deposition Change
The  estimate  for the  uncertainty in the relative
change predictions from  RADM  are  based  on  an
extensive bounding study that is documented as a
chapter  in NAPAP State of Science and Technol-
ogy (SOS/T)  Report No. 5 and is referenced in the
deposition modeling chapter  of this  report.  The
bounding analysis showed  that errors that could
contribute   to  scatter   in   comparisons   with
measurements do not result  in the same level of
uncertainty  in the relative  change predictions.
Several model errors were explored that, based  on
our current understanding, would be expected to
contribute most to model biases and/or to chang-
ing the sensitivity to emissions changes.  The  re-
sults were that the relative change predictions dif-
fered on  the  order of 10 percent, that is, a baseline
predicted change in sulfur deposition of 40 percent
had an uncertainty the order of ±4 percent.  These
results  and  conclusions were independently  re-
viewed as part of the  NAPAP SOS/T  review proc-
ess.  The analysis was deemed valid  and the best
estimate  at this time.  The reviewer is incorrectly
associating scatter with accuracy of change predic-
tions. Much  of the scatter relates to site represen-
tativeness and issues of comparing a volume aver-
age to a point.  Such scatter could be inherent and,
as such,  is not a reflection on the  ability of the
model to  represent a change in deposition.  The
assumption here is that the scatter represents a sys-
tematic difference  in  pattern  that  is preserved
when moving to a different level of deposition.
Also, because scatter may be inherent, bias is the
more fundamental  measure related  to accuracy.
One must be careful, because scatter and bias can
be quite unrelated when dealing with an advanced
air  quality model.   RADM  is  not  a regression
model and arguments and  experience about pre-
dictability for regression models do not carry over
to advanced  science models.   The discussion of
uncertainty in Chapter 3 has been expanded to try
to reduce  misinterpretation.  Finally, it should be
noted that the uncertainty associated with the lo-
cation of "unplanned" sources in 2010 is minimal
since unplanned capacity is projected to represent
less than 2 percent of total emissions in that year.

ALLOWANCE TRADING PROGRAM
Comments
One commenter (12) concurred  with  the analysis
that  emissions trading is cost-effective and  main-
tains  environmental benefits.   Two commenters
(5,7) questioned the report's conclusion that the al-
lowance trading program would have less than a
10 percent impact on deposition of  sulfur  in the
Adirondacks in 2010.  One of these commenters
(7) stated  that EPA  should evaluate the effects of
the   allowance trading  program by correlating
emissions  monitoring data from Continuous Emis-
sions Monitoring  Systems  (CEMS)  with  actual
deposition data rather than  relying on mathemati-
cal projections of trading and resulting deposition.
The other  commenter (5) claimed that the allow-
ance trading completed  in March 1995 will work
to the detriment of the Adirondacks because  al-
lowance prices decreased from the previous two
years and  because the allowances purchased will
be used in areas contributing to deposition  in the
Adirondacks.  Another commenter (16) was pleased
to see a  comparison  of  acidic deposition  re-
ductions from emissions trading and a realistic al-
ternative  approach.   This  commenter suggested
that the conclusion that the current Title IV allow-
ance  trading system is  administratively  efficient
should be included among the summary conclu-
sions.

Response
EPA  agrees that emissions  measured  with  CEMS
provide the most accurate  and precise emissions
values.  However, CEMS data are only now avail-
able for the first year of compliance for sources
                                              D-21

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
subject to Phase I of the Acid Rain Program.  The
analysis in this report is designed to evaluate the
impact of the allowance  trading program in the
year 2010 when all affected electric utility sources
are in compliance.

EPA also agrees that deposition monitoring is a key
component in assessing the effectiveness of emis-
sions reductions programs on deposition.  Deposi-
tion monitoring programs currently in place pro-
vide data on trends and spatial distributions of
acidic  deposition species.   The  National Atmos-
pheric Deposition  Program/National Trends  Net-
work which  operates  at  about 200  sites in the
United States provides data on deposition of wet
sulfate and nitrates.  EPA also operates a 50-station
dry deposition  network.   As available,  data  from
these and other monitoring networks will continue
to be used to evaluate the  effectiveness oif the Acid
Rain Program.  Nevertheless, the  projection of the
effectiveness  of  the Acid  Deposition  Program,
when it is fully implemented in 2010, can only be
estimated by modeling future emissions  end depo-
sition scenarios. The models used in the report for
such projections  have been  extensively reviewed
and selected as  the most  realistically available
models.

EPA disagrees with the comments that the allow-
ance trading program  and  decreasing  allowance
prices work to  the  detriment of deposition in the
Adirondacks.  The  fact that (1) allowance prices
were lower in the March 1995 auction than in the
previous year,  and (2) prices have  been  much
lower than most analysts predicted prior to the first
EPA auction only means that the  cost of reducing
SC>2 is not as high  as previously expected.  Many
analysts  attribute the current low cost of allow-
ances to the availability of relatively inexpensive
fuel switching alternatives. Lower allowance costs
also result in the banking of allowances with con-
comitant early  emissions reduction  benefits.  Fur-
ther, Duke Power and Virginia Electric Power pur-
chased 109,530 of the 176,400 (about 62 percent)
of the allowances sold at the March 1995 auction.
Emissions from  plants owned by these utilities con-
tribute significantly less than 1 percent to acidic
deposition  in  the  Adirondacks.    Even  if  full
economic trading is assumed (i.e., electric; utilities
pursue trading to  the fullest extent  in order to
minimize the cost of compliance),  EPA has dem-
onstrated in the study that allowance trading is ex-
pected to have less than  a 2 percent  impact on
acidic deposition in the Adirondacks.
BENEFITS TO VISIBILITY, HUMAN HEALTH,
MATERIAL, AND CULTURAL RESOURCES
Comments
One commenter (16) felt that the discussion of im-
provement in visibility due to reductions in ambi-
ent sulfate concentrations  is oversimplified.  As
sulfate concentrations decrease, the relative  con-
tributions  to visibility degradation of other light-
scattering  species such as dust, water, nitrate, and
elemental  carbon increase.  Thus changes in  con-
centrations and extinction  coefficients  of  these
species  will become increasingly important.  A
second commenter (13)  stated that  RADM  should
not be used in visibility assessment due to short-
comings relative to oxidant estimation and the in-
ability of the model to allow nitrate  particles to be
formed.

Another commenter (15) stated that since the legis-
lative language does not specifically mention im-
pacts other than  aquatic  and terrestrial benefits,
the report should  not include them.  Furthermore,
the  CAAA  has  specific  sections  dealing  with
visibility which provide EPA regulatory authority
to impose restrictions on emissions that are precur-
sors to visibility impairment. This commenter also
stated that claims of additional health benefits
from reductions in precursors of acidic deposition
are inappropriate and are more properly addressed
by  Title I  of the CAAA.  Finally, the commenter
argued that the materials damage section is biased
and misleading because the real issue is whether
cost savings that  may  result from  reductions  in
acidic deposition  outweigh the costs  to achieve
those savings.

Response
An acid deposition  standard or standards  would
most likely be derived using sensitive  aquatic re-
sources as the environmental endpoint in selecting
the level of a standard.  This report recognizes that
environmental and  human  health  issues do not
function in isolation of one another.  For example,
what may improve surface water conditions in the
Adirondacks may also benefit visibility in the mid-
Appalachian  area  (e.g.,   Shenandoah  National
Park) and provide a further safeguard  against the
health effects associated with fine paniculate mat-
ter (e.g., sulfates).

Anthropogenic visibility impairment in the eastern
United  States  is primarily caused by sulfates (65
percent),  organics (14 percent), elemental  carbon
(11 percent), nitrates (5 percent), nitrogen oxide (3
percent),  and  suspended  dust  (2  percent).  The
                                               D-22

-------
                                                    APPENDIX D:  SUMMARY OF COMMENTS AND RESPONSES
contribution of sulfates to visibility impairment in
the Southwest and Northwest are 39 percent and
33  percent, respectively.  The largest constituent
(sulfates) as well as others  (nitrates and  nitrogen
dioxide) share  the same  precursors as those re-
sponsible for acidic deposition.  The visibility sec-
tion of  this report is intended to identify visibility
improvement as another important benefit of SO2
and NOX emissions reductions, and to estimate the
visibility benefits that could be associated with an
acid deposition  standard based on aquatic effects
in sensitive regions.

The shortcomings  of current  regional  photo-
chemical models like  RADM, as identified in the
peer-review process, has very little to do with the
prediction of sulfate air concentrations important
to visibility degradation.  Earlier concerns about
RADM's predictions of hydrogen peroxide turned
out to be problems with the measurements.  Sub-
sequent comparisons with measurements  suggest
RADM is projecting somewhat too much hydrogen
peroxide, not too little  as was earlier thought.  This
could lead to too much conversion to sulfate in
cloud water  in  nonprecipitating clouds that are
later evaporated.  Nevertheless,  comparisons  of
annual  average  concentrations  of  sulfates  with
RADM  annual average predictions are good, with
essentially no bias for higher levels of atmospheric
sulfate.  This indicates  that RADM can be  used for
visibility assessments.  It is  true that RADM  does
not allow nitrate particles to be formed. They ac-
count for 5 percent of the  light  extinction.  EPA
agrees that as sulfate particles decrease in  number
there ought to be an increase in the importance of
nitrate  particles, assuming that ammonia concen-
trations  do  not  change.     However,  organics
(associated  with  volatile  organic  compounds,
VOCs) account for 11  percent of the light extinc-
tion.  They would be  expected to decrease by at
least  25-30  percent  due  to  oxidant  controls
brought about by compliance with Title  I of the
CAAA.  Thus, there are offsetting trends that will
affect non-sulfate light extinction, causing  it to
change  much less that might be  expected.  Thus,
the  simple  approach used with  RADM, that only
uses change in sulfate  and ignores changes in ni-
trate- and  organic-associated light extinction, is
expected to be  a  reasonable  zeroeth  approxima-
tion of a plausible change in  overall  light extinc-
tion.

The report's discussion of potential health  benefits
associated with further  reductions of acidic deposi-
tion precursors provides a brief summary of current
scientific information as well as regulatory-related
issues  associated with Title I.  This discussion  is
qualitative in nature  because there are  several
regulatory and implementation issues under Title I
which  are currently  being  explored,  reviewed,
and/or developed by EPA  and others.  Based on
the  current  stage of  review  or  development,
modeling inventories are simply not yet available.
The Agency  is exploring whether  reduced emis-
sions, particularly of sulfur dioxide, nitrogen oxide,
and the  subsequent reduction in  fine paniculate
matter (i.e., sulfates and nitrates), is  likely to have
a  beneficial  impact  on  human health.    Other
Agency efforts will quantify the relationship and
benefits.

The costs and cost savings of air pollution,  specifi-
cally acidic deposition, on both materials of func-
tional as well as cultural or historical importance
is  a growing  field of research.  Both wet and dry
acidic  deposition are believed to contribute to the
overall decline in functionality and  aesthetic well-
being of materials damaged by air pollution. The
monetary benefits associated with control of those
pollutants (i.e., sulfates) may be significant and are
currently being investigated by EPA, NAPAP, and
others.

COSTS
Comments
One commenter (15) noted that, although the total
costs of control  are equal  for the geographically
targeted and the regional [national]  emissions red-
uctions approaches, the costs of the geographically
targeted  approach will  be  absorbed  by  fewer
sources.  Another commenter (6) questioned why
existing detailed economic  models which consider
emissions variations and timing of achieving redu-
ctions were not used.  This commenter also stated
that detailed  costs for NOX  control should  have
been calculated.  A third commenter (14) provided
"zeroeth  order"  cost-effectiveness  comparison of
two alternative control strategies (based on appli-
cation  of Best Available Control Technology after
a specified boiler life) with Title IV.

A  fourth  commenter (16)  raised several points or
questions:  although utility generating  units are
already required  to  have continuous  emissions
monitors (CEMs), the costs of monitoring emissions
from individual  sources using CEMs could be an
impediment to implementing a control scenario
based on trading;  did  the costs calculated  for the
additional  national  SO2   reductions  scenarios
assume the use  of allowance trading, and, if so,
how was trading accounted for in cost estimates;
                                               D-23

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
average costs of control  are inappropriate as the
basis  for cost  comparisons  among  reductions
scenarios   (marginal   costs   would   be   more
appropriate);   economic   impacts   to   local
economies  should  be considered  for  the  geo-
graphically  targeted  reductions scenarios; further
reduction requirements would  provide greater im-
petus for  technological innovation  only under a
trading system as opposed to a targeted approach
that gives little flexibility to regulated sources; and
the 2010 CAAA Scenario (with trading)  is not the
appropriate  cost baseline because it incorporates
the allowance bank  projected to be built up be-
tween 1995 and 2009 and thus, understates rela-
tive to post-2010  costs after  the banked allow-
ances are used up.

Response
EPA agrees with the comment that even though the
total cost of  an additional 50 percent SO2 emission
reduction from electric utilities is about the  same
in the geographically targeted and non-targeted
scenarios, the costs are born by fewer sources in
the targeted reduction scenario than in the non-
targeted  scenario.   Cost-effectiveness ($/ton SO2
removed)  is slightly  higher in the geographically
targeted approach.

A detailed,  comprehensive economic model was
used by EPA to determine costs of  the Acid Rain
Program in 2010. As noted in the report, detailed
economic evaluation of the SO2 reduction scenar-
ios would require  the implementation of at least
two economic sector models—one for ihe electric
utility sector and  one for the  industrial sector.
Using such models would be very costly and time
intensive, and was beyond the scope of this study.
Given  the   large  uncertainty  associated  with
predicting economic  behavior far into the  future
(i.e., beyond 2010), EPA believes that little mean-
ingful additional information would be gained by
implementing such detailed models  over the  scop-
ing economic analysis conducted for this report.
In the report,  EPA recommended that if Congress
decides to give further consideration to an acid
deposition   standard,  more  detailed   economic
analysis should be conducted at that time.

The NOX control scenario analyzed  for this report
was  developed to represent realistic  future NOX
reductions.   Because implementation of the sec-
tions of Title I and Title IV of the Art requiring
NOX reductions is still ongoing, no detailed inven-
tory of either baseline (full implementation  of the
CAAA) nor additional control scenario exist.  EPA
is currently developing regulations that will specify
the degree of control necessary for NOX emissions
and types of control  technologies that will be re-
quired under Titles I and IV of the CAAA.  Until fi-
nal regulation for Title  I and IV have been devel-
oped, EPA does not believe that it can produce a
reasonable estimate of  (1) NOX  emissions far into
the future, and (2) the cost of reducing NOX emis-
sions as a result of Titles I and IV. Therefore, EPA
did not attempt to estimate NOX emissions or con-
trol costs post-implementation of the 1990 CAAA.

EPA  agrees with  the comment that the cost of
monitoring emissions from individual sources us-
ing CEMS should be considered  in determining the
costs  of control  scenarios   involving   industrial
sources  of SO2 which  are not required  to install
CEMS under Title IV. For a control program to be
integrated with the current allowance trading pro-
gram, industrial sources emitting SO2 must be able
to  demonstrate  that  they can achieve equivalent
reliability in their reported emissions as utilities al-
ready in the program. Before any program such as
this would be implemented both the technical fea-
sibility and  costs of achieving  this requirement
would have to be analyzed.

The costs calculated  for the additional SO2 reduc-
tion scenarios did not explicitly include allowance
trading,  because  EPA does  not know  how SO2
allowance allocations would be modified in order
to  reduce the SO2 allowance cap by 50 percent.
Under EPA's scoping approach for estimating  costs
under the additional SO2  reduction scenarios, EPA
assumed that units with the potential for significant
additional reductions in SO2 (i.e., those with emis-
sion rates greater than 0.8 Ibs SO2 per million Btu
in  2010) would be considered for additional cont-
rol. Using EPA's generic  retrofit scrubbing model,
EPA ranked these units  and their SO2 emission re-
duction  potential  in  order of average cost-effec-
tiveness  (i.e., measured as cost per ton SO2  re-
moved).  EPA then assumed that units would apply
retrofit scrubber in order of cost-effectiveness until
the 50   percent  additional  SO2 reduction   was
achieved. Note that this is consistent with an al-
lowance trading approach (i.e.,  the  units with the
most cost-effective SO2 reductions make the  addi-
tional reductions).   It may be appropriate to
assume that some level of cost  savings associated
with an  unrestricted  national trading program (as
assessed  for  implementation of the Acid  Rain
Program under  Title  IV) could also  result  in
reduced  costs of compliance with broad emission
reductions  beyond  the  current  program.   This
could widen  the  cost difference between a geo-
                                              D-24

-------
                                                   APPENDIX D: SUMMARY OF COMMENTS AND RESPONSES
graphically targeted and national emissions reduc-
tion strategies.

Next, EPA evaluated the potential for allowance
trading.  For this evaluation,  EPA grouped  units
into three categories:

  *  UNITS CONTROLLED BY 2010:   Units with
     emission rates  in 2010 of less than 0.8  Ibs
     SO2 per million Btu before consideration of
     an additional 50 percent reduction fall into
     this category.   EPA assumed that because:
     (1) these units would  already  be  tightly
     controlled,  and  (2)  the  allowance cap
     would be reduced by 50 percent and addi-
     tional SO2  reduction requirements  would
     only drive up allowance prices, that emis-
     sions from  these units would not  change
     (i.e., they would not pursue any additional
     allowance trading).

  *  UNITS APPLYING ADDITIONAL CONTROL: The
     units identified as applying retrofit  scrub-
     bers in order to achieve the additional  50
     percent  SO2 reduction fall into this cate-
     gory.  Even though  EPA  did not consider
     how allowance allocations would be modi-
     fied under a 50 percent reduction in the  al-
     lowance cap,  it is  reasonable to assumed
     that these units may have excess  allow-
     ances available.

  •  UNITS NOT  AS TIGHTLY CONTROLLED:  After
     assuming an  additional   50 percent SO2
     emission reduction, only  a small  set  of
     units  remain   that  have  SO2  emissions
     greater than 0.8 Ibs. SO2 per million Btu.
     Because EPA assumed that the  allowance
     cap would  be  reduced by 50 percent and
     SO2 allowance prices could only increase,
     EPA  assumed  these units would not  in-
     crease their emissions as a result of trading.
     Therefore,   EPA assumed  no  change  in
     emissions from these units.

The  commenter  is  correct to  point  out  that the
2010 scenario (with trading)  forecasts that about
0.5 million tons  of  previously  banked allowances
are used in 2010.   As a result, total utility SO2
emissions in  2010 are  forecast to be about  9.41
million tons instead  of 8.95 million tons—the long
term  SO2 allowance cap.  This use of banked  al-
lowances  in 2010 was forecast  by the  sector
model that EPA used to evaluate the impacts of al-
lowance trading.  This sector model is not de-
signed to  forecast  impacts beyond 2010.  There-
fore, EPA had little choice but to  rely on  this 2010
forecast.  It should be noted that because  of the
model's assumption that banked allowances will
be used in  2010,  utility SO2 emissions are only
about 5 percent greater than the 8.95 million ton
cap—well within the  range of uncertainty  of the
scenarios evaluated in the study.

EPA  disagrees with the commenter regarding the
appropriateness of  using average  cost per ton SO2
removed as a  measure of cost-effectiveness. The
commenter believes that marginal cost per ton SO2
removed would be a better measure. Both average
cost   and  marginal   cost  can  be  appropriate
measures of cost-effectiveness. Which measure is
more appropriate depends on what the cost-effec-
tiveness measure is to  be used for. If EPA were at-
tempting to assess  the impact of alternative addi-
tional reduction scenarios on the allowance mar-
ket price,  marginal cost-effectiveness would  be
appropriate because it would indicate the underly-
ing market  value  of  allowances.   This was  not
EPA's goal in this report, however.   In the case of
the study,  EPA wanted to provide  a  meaningful
metric that would facilitate a broad comparison of
the overall  cost-effectiveness of  each alternative
additional reduction scenario.  For this purpose,
EPA  believes that  average cost-effectiveness will
be more meaningful to members  of Congress than
marginal cost-effectiveness.

Any additional reduction scenario will have differ-
ing impacts on local  economies.  Until detailed
regulatory proposals are evaluated,  it is not possi-
ble to determine the specific  geographic distribu-
tion of emissions reductions.

EPA  agrees  with the commenter that approaches
provide flexibility in compliance by using allow-
ance trading will provide greater  impetus for tech-
nological  innovation.   Note, however,  that EPA
believes that a regionally targeted approach could
be developed  that  would continue to rely on  al-
lowance -trading.  Under a regionally targeted ap-
proach, regional allowance  markets could  de-
velop.
                                              D-25

-------
ACID DEPOSITION STANDARD FEASIBILITY STUDY
                                 EXHIBIT D-2. LISJ OF COMMFNTFRS
   1.  John M. Barnes, Department of Agriculture, Washington, DC. Letter of March 8, 1995.
   2.  Kathy A. Tonnessen, National Biological Service, Denver, Colorado. Letter of April 1, 1995.
   3.  Robert J. Devlin, Forest Service, Portland, Oregon. Letter of March 24, 1995.
   4.  Gary Spielman, New York State Department of Environmental Conservation, Albany, New York.
      Letters of March 13 and May 3, 1995.
   5.  Bernard Melewski, The Adirondack Council, Elizabethtown, New York.  Letter of March 31, 1995.
   6.  Robert Glennon, Adirondack Park Agency, Ray Brook, New York.  Letter of March 31, 1995.
   7.  David Gibson and Daniel R. Plurnley, Association for the Protection of The Adirondacks, Schenec-
      tady, New York.  Letter of March 30, 1995.
   8.  Myron J. Mitchell, State University of New York, Syracuse New York. Letter of March 30, 1995.
   9.  Rick Webb, University of Virginia, Charlottesville, Virginia. Letter of March 31, 1995.
  10.  Lisa J. Thorvig, Minnesota Pollution Control Agency, St.  Paul, Minnesota.  Letter of March 31, 1995.
  11.  Donald Theiler, Wisconsin Department of Natural Resources, Madison, Wisconsin.  Letter of
      March 31, 1995.
  12.  Samuel A. Leonard, General Motors Corporation, Detroit, Michigan. Letter of April 1, 1995.
  13.  Jon M. Loney, Tennessee Valley Authority, Knoxville, Tennessee. Letter of March 31, 1995.
  14.  Roger Morris, Department of Energy, Washington, DC.   Letter of March 31, 1995.
  15.  Quinlan j. Shea and Michael L. Teague, Hunton & Williams, Washington, DC.  Letter of March 31,
      1995.  Representing the Utility Air Regulatory Group and the National Mining Association.
  16.  Mary Ann Allan, Electric Power Research Institute, Palo  Alto, California.  Letter of March 30, 1995.
  17.  H.A. Clarke, Environment Canada, Ottawa, Ontario. Letter of April 3, 1995.
                                              D-26
                                                           • U.S. GOVERNMENT PRINTING OFFICE: 1995-715-635/82458

-------