&EPA
United State
Environmental PwtocSon
Agency
Policy Assessment for the Review of
the Secondary National Ambient Air
Quality Standards for Oxides of
Nitrogen and Oxides of Sulfur
Appendices
-------
-------
EPA-452/R-ll-005b
February 2011
Policy Assessment for the Review of the Secondary
National Ambient Air Quality Standards for Oxides of
Nitrogen and Oxides of Sulfur
Appendices
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Health and Environmental Impacts Division
Research Triangle Park, North Carolina
-------
DISCLAIMER
This document has been reviewed by the Office of Air Quality Planning and Standards,
U.S. Environmental Protection Agency, and approved for publication. This final document has
been prepared by staff from the Office of Air Quality Planning and Standards, U.S.
Environmental Protection Agency. Any opinions, findings, conclusions, or recommendations are
those of the authors and do not necessarily reflect the views of the U.S. Environmental
Protection Agency. Mention of trade names or commercial products is not intended to constitute
endorsement or recommendation for use. Any questions or comments concerning this document
should be addressed to Richard Scheffe, U.S. Environmental Protection Agency, Office of Air
Quality Planning and Standards, C304-02, Research Triangle Park, North Carolina 27711 (email:
scheffe.richard@epa.gov).
-------
APPENDICES
Appendix A: Analysis of Critical Loads, Comparing Aquatic and Terrestrial Acidification
Appendix B: Methodologies and Assumptions Used In Steady State Ecosystem Modeling
Appendix C: Ecoregions, Level III: Description And Summary of Environmental Conditions
Appendix D: Maps and Calculation Procedures for Alternative Standards
Appendix E: Derivation to Use Measured Total Nitrate as a Surrogate for NOy
Appendix F: Evaluation of Variability, Sensitivity and Uncertainty in the Acidification Index
Appendix G: Cumulative Uncertainty Analysis
-------
-------
Appendix A
Analysis of Critical Loads, Comparing Aquatic and Terrestrial Acidification
Background
Critical load is defined as, "a quantitative estimate of ecosystem exposure to one or more
pollutants below which significant harmful effects on specified sensitive elements of the
environment do not occur, according to present knowledge" (McNulty et al, 2007), and critical
loads can be estimated for aquatic and terrestrial ecosystems. Within the Risk and Exposure
Assessment for Review of the Secondary National Ambient Air Quality Standards for Oxides of
Nitrogen and Sulfur (hereafter referred to as REA Report) (US EPA, 2009), critical loads of
acidification for aquatic systems were determined by relating specific amounts of acidifying
nitrogen and sulfur deposition to selected Acid Neutralizing Capacities (ANC) within freshwater
lakes or streams. The presence and abundance offish species served as the biological indicator
of the impacts of the exceedance of critical acid loads by nitrogen and sulfur deposition.
Estimation of critical acid loads for terrestrial systems within the REA Report (US EPA, 2009)
related acidifying nitrogen and sulfur deposition to the base cation to aluminum (Bc/Al) ratio in
the soil solution, and the health of sugar maple and red spruce in forest ecosystems served as the
biological indicator of the impacts of critical acid load exceedance. A main distinction between
these two critical loads is that aquatic critical loads are largely an integrated function of the
chemistry of run-off waters that feed the lake or stream within a watershed, while terrestrial
critical acid loads are determined by the rooting zone section of the soil profile in a forest
ecosystem. Therefore, it is possible to have different critical load values for aquatic and
terrestrial ecosystems within the same watershed.
The goal of this Task was to determine the relative degree of protection offered by
aquatic versus terrestrial critical acid loads within a landscape. Critical acid loads for lakes and
streams within watersheds of the Adirondacks and Shenandoah Valley were compared against
terrestrial critical loads calculated for same watersheds to determine which estimate had the
lowest, most protective critical load for acidifying nitrogen and sulfur deposition.
A-l
-------
Methods
For the REA Report (US EPA, 2009), critical acid loads were determined for 169 lakes
and 60 streams in the Adirondacks and Shenandoah Case Study Areas, respectively. These
critical loads were calculated using four different ANCs, 0, 20, 50 and 100 |ieq/L, that ranged in
the level of protection offered to fish species abundance and diversity, and the resulting critical
acid loads were classified into four "current condition of acidity and sensitivity to acidification"
categories. "Highly Sensitive" water bodies had critical loads less than or equal to 50
meq/m2/yr, "Moderately Sensitive" systems had critical loads ranging from 51 to 100 meq/m2/yr,
"Low Sensitivity" lakes and streams had critical loads that ranged from 101 to 200 meq/m2/yr,
and "Not Sensitive" systems had critical acid loads greater than 201 meq/m2/yr.
For the purposes of this Task, aquatic critical acid loads corresponding to an ANC of 50
meq/m2/yr were selected, and the locations of the lakes and streams in the Adirondacks and
Shenandoah Case Study Areas were mapped by HUC12 watersheds. Availability of data for
terrestrial acidification estimates was determined for each HUC, and only HUCs that had
sufficient data were mapped. Data from the U. S. Department of Agriculture- Natural Resources
Conservation Service (USDA-NRCS) SSURGO soils database (USDA-NRCS, 2008) had the
poorest coverage. This data restriction limited the number of water bodies that could be included
in the analysis to 62 and 35 for the Adirondacks and Shenandoah Case Study Areas, respectively.
To examine a representative selection of water bodies in each Case Study Area, four
watersheds containing lakes or streams from each of the four "current condition of acidity and
sensitivity to acidification" categories were randomly selected. Therefore, a total of 16
watersheds were chosen for each Case Study Area. All four "current condition of acidity and
sensitivity to acidification" categories were evenly represented for the Adirondacks Case Study
Area (four watersheds for each of the four categories). However, due to the limited number of
watersheds in the Shenandoah Area and a lower proportion of lakes with low sensitivities to
acidifying nitrogen and sulfur deposition ("Low Sensitivity" and "Not Sensitive"), it was not
possible to have equal representation of all "current condition of acidity and sensitivity to
acidification" categories. Therefore, there was a larger representation of streams that were more
sensitive to acidification ("Highly Sensitive" and "Moderately Sensitive"). All water bodies that
were located in each of the selected HUCs were included in the analyses. In many cases, these
A-2
-------
water bodies ranged in sensitivity to acidification. In total, 29 lakes and 20 streams were
analyzed in the Adirondacks and Shenandoah Case Study Areas, respectively (Table A-l and A-
2).
A-3
-------
Table A-l. Watersheds (HUC 12) and fresh water lakes in the Adirondacks Case Study Area that were used in the
comparison of aquatic and terrestrial critical acid loads. Lake IDs and associated aquatic critical acid loads (CL) in meq/m2/yr,
based on an ANC of 50 |ieq/L, are indicated in each cell and are from the REA REPORT (US EPA, 2009).
HUC
020100010103
020100040203
020100080304
020100081602
020200020101
020200020704
020200040805
041501011001
041503020801
041503040102
041503040204
041503050103
041503050104
041503050302
CURRENT CONDITION OF ACIDITY AND
SENSITIVITY TO ACIDIFICATION CATEGORY
Highly Sensitive (CL <
50 meq/m2/yr)
1A2-078O (CL = 33)
NY029L (CL = 39)
NY284L (CL = 23)
NY285L (CL = 42)
1A1-089O (CL = 43)
NY290L (CL = 30)
NY289L (CL = 50)
Moderately Sensitive (CL =
51-100 meq/m2/yr)
NY013L(CL = 64)
NY536L (CL = 69)
NY278L (CL = 57)
050215AO(CL = 74)
NY793L (CL = 97)
Low Sensitivity
(CL = 101-200 meq/m2/yr)
1A2-028O (CL = 106)
NY310L(CL=147)
NY292L(CL=117)
NY008L (CL = 146)
NY007L (CL = 165)
Not Sensitive
(CL > 201 meq/m2/yr)
NY534L (CL = 1043)
NY308L (CL = 485)
NY312L (CL = 588) NY313L
(CL = 598)
NY500L(CL = 610)
NY783L (CL = 455)
A-4
-------
HUC
CURRENT CONDITION OF ACIDITY AND
SENSITIVITY TO ACIDIFICATION CATEGORY
Highly Sensitive (CL <
50 meq/m2/yr)
Moderately Sensitive (CL =
51-100 meq/m2/yr)
Low Sensitivity
(CL = 101-200 meq/m2/yr)
Not Sensitive
(CL > 201 meq/m2/yr)
041503050407
NY767L (CL = 51) NY529L
(CL = 73) NY528L (CL = 82)
NY769L (CL = 99)
NY768L(CL=114)
041503050601
NY004L (CL = 168)
A-5
-------
Table A-2. Watersheds (HUC 12) and streams in the Shenandoah Case Study Area that were used in the comparison of
aquatic and terrestrial critical acid loads. Stream IDs and associated aquatic critical acid loads (CL) in meq/m2/yr, based on an
ANC of 50 neq/L, are indicated in each cell and are from the REA REPORT (US EPA, 2009).
HUC
020700050401
020700050502
020700050703
020700050705
020700050801
020700050803
020700060101
020801030301
020801030402
020802010702
020802010703
020802010801
020802010803
020802020102
020802020401
020802030601
CURRENT CONDITION OF ACIDITY AND SENSITIVITY TO ACIDIFICATION CATEGORY
Highly Sensitive (CL
< 50 meq/m2/yr)
VT37 (CL = 26)
VT57 (CL = 39)
VT40(CL= 13)
VT35(CL = 37)
VT36 (CL = 24)
DR01 (CL = 33)
WOR1 (CL = 43)
VT53 (CL = 40)
VT10(CL= 15)
VT11 (CL= 14)
VT14(CL= 14)
VT15(CL= 13)
VT16(CL = 20)
VT41 (CL= 15)
Moderately Sensitive (CL
= 51-100 meq/m2/yr)
VT54 (CL = 69)
VT62 (CL = 68)
VT12(C = 75)
VT38(CL = 66)
VT46 (CL = 52)
Low Sensitivity (CL
= 101-200 meq/m2/yr)
VT60(CL= 198)
Not Sensitive
(CL > 201 meq/m2/yr)
VT61 (CL = 231)
A-6
-------
Terrestrial critical acid loads were calculated for each of the 16 watersheds using the
simple mass balance method (UNECE, 2004) and data sources outlined in the REA Report (US
EPA, 2009), and Bc/Al soil solution indicator values of 1.2 and 10.0. Briefly, average values for
base cation deposition (calcium, potassium, magnesium and sodium), chloride deposition, and
annual runoff (m3/ha/yr) were determined for each watershed (Table A-3). The Kgibb constant
(m6eq2) was determined by the average percent organic matter in the soil (Table A-4), and N
immobilization in the soil was set to the constant value of 42.86 eq/ha/yr (McNulty et al, 2007).
It was assumed that active harvesting did not occur in each of the watersheds. Therefore base
cation (calcium, magnesium and potassium) and nitrogen uptake were 0 eq/ha/yr (UNECE,
2004). Similarly, it was assumed that the majority of each watershed consisted of upland sites.
Therefore, denitrification losses were assumed to be 0 eq/ha/yr (McNulty et al., 2007). Base
cation weathering was estimated using the clay substrate model (equations 1-3) (McNulty et al.,
2007).
Acid Substrate: BCe = (56.7 x %clay)- (o.32 x (%clay)2) (1)
Intermediate Substrate: BCe = 500 + (53.6 x %clay)- (o. 18 x (%clay)2) (2)
Basic Substrate: BCe = 500 + (59.2 x %clay) (3)
where
BCe = empirical soil base cation (Ca2+ + K+ + Mg2+ + Na+) weathering rate
(eq/ha/yr)
% clay = the percentage of clay within the top 50cm of the soil.
The U.S. Department of Agriculture- Natural Resources Conservation Service (USDA-NRCS)
SSURGO soils database (USDA-NRCS, 2008)) and state-level geology (U.S. Geological Survey
(USGS) state-level integrated map database for the United States (USGS, 2009)) were used to
determine parent material acidity classification. Parent material acidity was determined for each
SSURGO polygon within each watershed using the criteria outlined in the REA Report (US
EPA, 2009), and the contributions of base cations from the weathering of acid, intermediate and
basic substrates (eq/ha/yr) were determined by a weighted average based on the proportion of
area occupied by each parent material acidity class. Rooting depth was assumed to be 50 cm and
masses of calcium, magnesium, potassium, sodium and nitrogen were converted to eq/ha/yr units
based on molar charge equivalents. Unless indicated otherwise, the units used in the calculation
A-7
-------
of critical acid loads were eq/ha/yr. The estimated terrestrial critical loads for the 16 watersheds
in the Adirondacks and Shenandoah Case Study Areas are presented in Table A-5.
Table A-3. Name, type and source of data used in the simple mass balance estimates of
terrestrial critical acid loads for the watersheds in the Adirondacks and Shenandoah Case
Study Areas.
DATA
Base cation (Ca2+,
Mg2+, Na+, K+)
deposition — wet
Chloride (Cl")
deposition — wet
Runoff
Soil horizon depth
Percentage of clay
by soil horizon
Percentage of
organic matter by
soil horizon
Soil parent
material
State-level
bedrock geology
NAME
CMAQ/
NADP
NADP
Annual run-off
(1: 7,500,000
scale)
SSURGO
SSURGO
SSURGO
SSURGO
State
Geological
Map
Compilation
TYPE
GIS
datalayers
GIS
datalayer
GIS
datalayer
GIS
datalayer
GIS
datalayer
GIS
datalayer
GIS
datalayer
GIS
datalayer
SOURCE
Provided by U.S. Environmental
Protection Agency (EPA)/NADP,
2003a,c, d, e
NADP, 2003b
Gebertetal, 1987
USDA-NRCS, 2008
USDA-NRCS, 2008
USDA-NRCS, 2008
USDA-NRCS, 2008
USGS, 2009
Note: CMAQ = Community Multiscale Air Quality Model; NADP = National Atmospheric
Deposition Program; GIS = Geographic Information System; SSURGO = Soil Survey
Geographic Database
A-8
-------
Table A-4. Gibbsite equilibrium (Kga,/,) constant determined by percentage of soil organic
matter (modified from McNulty et al. 2007).
Soil Type Layer
Mineral soils: C layer
Soils with low organic matter: B/C layers
Soils with some organic material: A/E
layers
Peaty and organic soils: organic layers
Organic
Matter %
<5
5tol5
15 to 70
>70
Kgibb (m6/eq2)
950
300
100
9.5
A-9
-------
Table A-5. Terrestrial critical acid loads (in eq/ha/yr) for the watersheds in the
Adirondacks and Shenandoah Case Study Areas.
Case Study
Area
Adirondacks
Adirondacks
Adirondacks
Adirondacks
Adirondacks
Adirondacks
Adirondacks
Adirondacks
Adirondacks
Adirondacks
Adirondacks
Adirondacks
Adirondacks
Adirondacks
Adirondacks
Adirondacks
Shenandoah
Shenandoah
Shenandoah
Shenandoah
Shenandoah
Shenandoah
Shenandoah
Shenandoah
Shenandoah
Shenandoah
Shenandoah
Shenandoah
Shenandoah
Shenandoah
Shenandoah
Shenandoah
HUC12
020100010103
020100040203
020100080304
020100081602
020200020101
020200020704
020200040805
041501011001
041503020801
041503040102
041503040204
041503050103
041503050104
041503050302
041503050407
041503050601
020700050401
020700050502
020700050703
020700050705
020700050801
020700050803
020700060101
020801030301
020801030402
020802010702
020802010703
020802010801
020802010803
020802020102
020802020401
020802030601
Terrestrial Critical Acid Load
(eq/ha/yr)
Bc/Al = 1.2
2045
1316
1329
1670
1484
1707
1770
1770
1664
1627
1436
1774
1794
1754
1447
1203
1440
1560
1762
1852
1799
1975
1638
1511
1393
1603
1642
1635
1573
1519
1264
1660
Bc/Al = 10.0
1134
712
731
922
819
935
951
955
912
880
786
957
968
947
789
656
802
871
979
1032
1003
1102
914
843
776
890
912
909
876
845
703
918
A-10
-------
Maps were generated to compare the aquatic and terrestrial critical acid loads in each
watershed to determine which estimate provided the greatest protection against acidifying
nitrogen and sulfur deposition. In each watershed, the terrestrial critical load estimate was
compared against each aquatic critical load, and the load with the lowest value was set to
represent the most sensitive component in the watershed. All critical load estimates were
converted to eq/ha/yr for the comparisons.
Results
Maps indicating and comparing the sensitivities of the terrestrial and aquatic critical loads
to nitrogen and sulfur deposition in each watershed of the Adirondacks and Shenandoah Case
Study Areas are presented in Figures A-l to A-4 and Tables A-6 to A-9.
In the Adirondacks Case Study Area, 7 of the 16 watersheds had terrestrial critical acid
loads (based on a Bc/Al of 10.0) that were lower and therefore more sensitive to acidification
than all the lakes in the watershed. However, when the terrestrial critical loads were calculated
with a Bc/Al soil solution ratio of 1.2, only 5 of the 16 watersheds were protected by a terrestrial
critical load that was lower than the aquatic critical loads of the lakes. Three watersheds in the
Adriondacks Case Study Area had terrestrial critical loads (based on a Bc/Al of 10.0) that were
lower and higher than the critical loads for the lakes in the watershed, and one watershed had a
similar mixture of aquatic versus terrestrial acid load protections for terrestrial critical loads
estimated with a Bc/Al of 1.2. In general, a main trend in the Adirondacks Case Study Area was
that watersheds with "Highly Sensitive" and "Moderately Sensitive" lakes were more protected
by aquatic than terrestrial critical acid loads, while the watersheds with "Low Sensitivity" and
"Not Sensitive" lakes were more protected by terrestrial critical acid loads.
Similar trends were found in the Shenandoah Case Study Area. However, there was little
distinction between terrestrial acid loads that were calculated with a Bc/Al of 10.0 versus 1.2.
Terrestrial critical acid loads offered a higher level of protection than did the stream aquatic
critical loads in only one watershed. The two streams in this watershed had "Low Sensitivity" or
were "Not Sensitive" to acidifying nitrogen and sulfur deposition. The 15 watersheds that had
streams with aquatic critical loads lower and more protective than the terrestrial critical loads
were all "Highly Sensitive" or "Moderately Sensitive" to acidifying nitrogen and sulfur
deposition.
A-ll
-------
In summary, a comparison of the terrestrial and aquatic critical acid loads for watersheds
in the Adirondacks and Shenandoah Case Study Areas indicated that, in general, the aquatic
critical acid loads offered greater protection to the watersheds than did the terrestrial critical
loads. In situations where the terrestrial loads were more protective, the lakes or streams in the
watershed were rated as having "Low Sensitivity" or "Not Sensitive" to acidifying nitrogen and
sulfur deposition. Conversely, when the water bodies were more sensitive to deposition
("Highly Sensitive" or "Moderately Sensitive"), the aquatic critical acid loads consistently
provided a greater level of protection against acidifying nitrogen and sulfur deposition in the
watershed.
A-12
-------
Figure A-l. Comparison of aquatic and terrestrial critical loads of acidification (in
eq/ha/yr) in the 16 watersheds of the Adirondacks Case Study Area, based on an ANC of 50
eq/L for the aquatic loads and a Bc/Al of 10.0 for the terrestrial loads. Colored circles
indicate the locations of the waters bodies within each watershed. Green circles indicate lakes
with critical load values less than the terrestrial critical load for the same watershed. Red circles
indicate a condition where the terrestrial critical load is lower than the lake critical load.
Bc/AI = 10.0
020100080304
04150305040/ •
041503050302
041501011001
«04
Legend
Aquatic more sensitive
Terrestrial more sensitive
Adirondacks State Park
HUC 12 Watersheds
A-13
-------
Figure A-2. Comparison of aquatic and terrestrial critical loads of acidification (in
eq/ha/yr) in the 16 watersheds of the Adirondacks Case Study Area, based on an ANC of 50
eq/L for the aquatic loads and a Bc/Al of 1.2 for the terrestrial loads. Colored circles
indicate the locations of the waters bodies within each watershed. Green circles indicate lakes
with critical load values less than the terrestrial critical load for the same watershed. Red circles
indicate a condition where the terrestrial critical load is lower than the lake critical load.
041503040204
041603050407 Q
041501011001
% 041503050104
• Aquatic more sensitive
• Terrestrial more sensitive
Adirondacks State Park
• HUC 12 Watersheds
A-14
-------
Figure A-3. Comparison of aquatic and terrestrial critical loads of acidification (in
eq/ha/yr) in the 16 watersheds of the Shenandoah Case Study Area, based on an ANC of 50
eq/L for the aquatic loads and a Bc/Al of 10.0 for the terrestrial loads. Colored circles
indicate the locations of the waters bodies within each watershed. Green circles indicate streams
with critical load values less than the terrestrial critical load for the same watershed. Red circles
indicate a condition where the terrestrial critical load is lower than the stream critical load.
Bc/AI = 10.0
WEST VIRGINIA
020802020102 VIRGINIA
X* /
020802010702
020802010703
020700050703
> /
020802020401
Legend
• Aquatic more sensitive
• Terrestrial more sensitive
HUC 12 Watersheds
National Forests
A-15
-------
Figure A-4. Comparison of aquatic and terrestrial critical loads of acidification (in
eq/ha/yr) in the 16 watersheds of the Shenandoah Case Study Area, based on an ANC of 50
eq/L for the aquatic loads and a Bc/Al of 1.2 for the terrestrial loads. Colored circles
indicate the locations of the waters bodies within each watershed. Green circles indicate streams
with critical load values less than the terrestrial critical load for the same watershed. Red circles
indicate a condition where the terrestrial critical load is lower than the stream critical load.
ippahannock /
020801030301
•'-.» —J
020801030402
-'••
WEST VIRGINIA
020700050803
£
'v-*
020700050801
O
020802020102 VIRGINIA
•
020802010702
020802010803
w
• i
020802030601
Legend
• Aquatic more sensitive
• Terrestrial more sensitive
HUC 12 Watersheds
National Forests
A-16
-------
Table A-6. Relative sensitivities of aquatic versus terrestrial critical loads in the 29 lakes and 16 watersheds of the Adirondack
Case Study Area (based on an ANC of 50 jieq/L for the aquatic loads and a Bc/Al of 10.0 for the terrestrial critical loads and
common unit of eq/ha/yr) to acidifying nitrogen and sulfur deposition. Lake IDs are indicated in each cell and are from the REA
REPORT (US EPA, 2009). Green text indicates lakes where the aquatic critical load was less than the terrestrial critical load value for
the watershed. Red text indicates a condition where the terrestrial critical load for the watershed was lower than the aquatic critical
load for the lake within the same watershed.
HUC
020100010103
020100040203
020100080304
020100081602
020200020101
020200020704
020200040805
041501011001
041503020801
041503040102
041503040204
041503050103
041503050104
CURRENT CONDITION OF ACIDITY AND
SENSITIVITY TO ACIDIFICATION CATEGORY
Highly Sensitive
(CL < 50
meq/m2/yr)
Moderately Sensitive
(CL = 51-100
meq/m2/yr)
NY793L
Low Sensitivity
(CL = 101-200
meq/m2/yr)
1A2-028O
NY310L
NY292L
Not Sensitive
(CL > 201
meq/m2/yr)
NY534L
NY308L
NY312L
NY313L
NY500L
NY783L
A-17
-------
HUC
041503050302
041503050407
041503050601
CURRENT CONDITION OF ACIDITY AND
SENSITIVITY TO ACIDIFICATION CATEGORY
Highly Sensitive
(CL < 50
meq/m2/yr)
Moderately Sensitive
(CL = 51-100
meq/m2/yr)
NY528L
NY769L
Low Sensitivity
(CL = 101-200
meq/m2/yr)
NY008L
NY007L
NY768L
NY004L
Not Sensitive
(CL > 201
meq/m2/yr)
A-18
-------
Table A-7. Relative sensitivities of aquatic versus terrestrial critical loads in the 29 lakes and 16 watersheds of the Adirondack
Case Study Area (based on an ANC of 50 jieq/L for the aquatic loads and a Bc/Al of 1.2 for the terrestrial critical loads and
common unit of eq/ha/yr) to acidifying nitrogen and sulfur deposition. Lake IDs are indicated in each cell and are from the REA
REPORT (US EPA, 2009). Green text indicates lakes where the aquatic critical load was less than the terrestrial critical load value for
the watershed. Red text indicates a condition where the terrestrial critical load for the watershed was lower than the aquatic critical
load for the lake within the same watershed.
HUC
020100010103
020100040203
020100080304
020100081602
020200020101
020200020704
020200040805
041501011001
041503020801
041503040102
041503040204
041503050103
CURRENT CONDITION OF ACIDITY AND
SENSITIVITY TO ACIDIFICATION CATEGORY
Highly Sensitive
(CL < 50
meq/m2/yr)
Moderately Sensitive
(CL = 51-100
meq/m2/yr)
Low Sensitivity
(CL = 101-200
meq/m2/yr)
NY310L
Not Sensitive
(CL > 201
meq/m2/yr)
NY534L
NY308L
NY312L
NY313L
NY500L
NY783L
A-19
-------
HUC
041503050104
041503050302
041503050407
041503050601
CURRENT CONDITION OF ACIDITY AND
SENSITIVITY TO ACIDIFICATION CATEGORY
Highly Sensitive
(CL < 50
meq/m2/yr)
M t
Moderately Sensitive
(CL = 51-100
meq/m2/yr)
Low Sensitivity
(CL = 101-200
meq/m2/yr)
i M 1 '
NY004L
Not Sensitive
(CL > 201
meq/m2/yr)
A-20
-------
Table A-8. Relative sensitivities of aquatic versus terrestrial critical loads in the 20 streams and 16 watersheds of the
Shenandoah Case Study Area (based on an ANC of 50 jieq/L for the aquatic loads and a Bc/Al of 10.0 for the terrestrial
critical loads and common unit of eq/ha/yr) to acidifying nitrogen and sulfur deposition. Stream IDs are indicated in each cell
and are from the RE A REPORT (US EPA, 2009). Green text indicates streams where the aquatic critical load was less than the
terrestrial critical load value for the watershed. Red text indicates a condition where the terrestrial critical load for the watershed was
lower than the aquatic critical load for the stream within the same watershed.
HUC
020700050401
020700050502
020700050703
020700050705
020700050801
020700050803
020700060101
020801030301
020801030402
020802010702
020802010703
020802010801
CURRENT CONDITION OF ACIDITY AND
SENSITIVITY TO ACIDIFICATION CATEGORY
Highly Sensitive
(CL < 50
meq/m2/yr)
'.-,-. i
i . i
Moderately Sensitive
(CL = 51-100
meq/m2/yr)
Low Sensitivity
(CL = 101-200
meq/m2/yr)
VT60
Not Sensitive
(CL > 201
meq/m2/yr)
VT61
A-21
-------
HUC
020802010803
020802020102
020802020401
020802030601
CURRENT CONDITION OF ACIDITY AND
SENSITIVITY TO ACIDIFICATION CATEGORY
Highly Sensitive
(CL < 50
meq/m2/yr)
Moderately Sensitive
(CL = 51-100
meq/m2/yr)
I i '
Low Sensitivity
(CL = 101-200
meq/m2/yr)
Not Sensitive
(CL > 201
meq/m2/yr)
A-22
-------
Table A-9. Relative sensitivities of aquatic versus terrestrial critical loads in the 20 streams and 16 watersheds of the
Shenandoah Case Study Area (based on an ANC of 50 jieq/L for the aquatic loads and a Bc/Al of 1.2 for the terrestrial critical
loads and common unit of eq/ha/yr) to acidifying nitrogen and sulfur deposition. Stream IDs are indicated in each cell and are
from the REA REPORT (US EPA, 2009). Green text indicates streams where aquatic critical loads were less than the terrestrial
critical load value for the watershed. Red text indicates a condition where the terrestrial critical load for the watershed was lower than
the aquatic critical load for the stream within the same watershed.
HUC
020700050401
020700050502
020700050703
020700050705
020700050801
020700050803
020700060101
020801030301
020801030402
020802010702
020802010703
020802010801
CURRENT CONDITION OF ACIDITY AND
SENSITIVITY TO ACIDIFICATION CATEGORY
Highly Sensitive
(CL < 50
meq/m2/yr)
'.-,-. i
Moderately Sensitive
(CL = 51-100
meq/m2/yr)
Low Sensitivity
(CL = 101-200
meq/m2/yr)
VT60
Not Sensitive
(CL > 201
meq/m2/yr)
VT61
A-23
-------
HUC
020802010803
020802020102
020802020401
020802030601
CURRENT CONDITION OF ACIDITY AND
SENSITIVITY TO ACIDIFICATION CATEGORY
Highly Sensitive
(CL < 50
meq/m2/yr)
Moderately Sensitive
(CL = 51-100
meq/m2/yr)
I i '
Low Sensitivity
(CL = 101-200
meq/m2/yr)
Not Sensitive
(CL > 201
meq/m2/yr)
A-24
-------
REFERENCES
Gebert, W.A., DJ. Graczyk, and W.R. Krug, 1987. Average Annual Runoff in the United States,
1951-80: U.S. Geological Survey Hydrologic Investigations Atlas HA-710, Scale
1:7,500,000. GIS datalayer. U.S. Department of Interior, U.S. Geological Survey,
Madison, WI. Available at: http://water.usgs.gov/GIS/dsdl/runoff.eOO.gz (accessed
September 9, 2009).
McNulty, S.G., E.G. Cohen, H. Li, and J.A. Moore-Myers. 2007. Estimates of critical acid loads
and exceedences for forest soils across the conterminous United States. Environmental
Pollution J49:281-292.
NADP (National Atmospheric Deposition Program). 2003a. Annual Calcium Wet Deposition,
2002. GIS datalayer. National Atmospheric Deposition Program, Illinois State Water
Survey, Champaign, IL. Available at http://nadp.sws.uiuc.edu/maps/2002.
NADP (National Atmospheric Deposition Program). 2003b. Annual Chloride Wet Deposition,
2002. GIS datalayer. National Atmospheric Deposition Program, Illinois State Water
Survey, Champaign, IL. Available at http://nadp.sws.uiuc.edu/maps/2002.
NADP (National Atmospheric Deposition Program). 2003c. Annual Magnesium Wet Deposition,
2002. GIS datalayer. National Atmospheric Deposition Program, Illinois State Water
Survey, Champaign, IL. Available at http://nadp.sws.uiuc.edu/maps/2002.
NADP (National Atmospheric Deposition Program). 2003d. Annual Potassium Wet Deposition,
2002. GIS datalayer. National Atmospheric Deposition Program, Illinois State Water
Survey, Champaign, IL. Available at http://nadp.sws.uiuc.edu/maps/2002.
NADP (National Atmospheric Deposition Program). 2003e. Annual Sodium Wet Deposition,
2002. GIS datalayer. National Atmospheric Deposition Program, Illinois State Water
UNECE (United Nations Economic Commission for Europe). 2004. Manual on Methodologies
and Criteria for Modeling and Mapping Critical Loads and Levels and Air Pollution
Effects, Risks, and Trends. Convention on Long-Range Transboundary Air Pollution,
Geneva Switzerland. Available at http://www.icpmapping.org (accessed August 16,
2006).
USDA-NRCS (United States Department of Agriculture-Natural Resources Conservation
Service). 2008. Soil Survey Geographic (SSURGO) Database. GIS datalayer. U.S.
Department of Agriculture, Natural Resources Conservation Service, Washington, DC.
Available at http://datagateway.nrcs.usda.gov.
US EPA (United States Environmental Protection Agency). 2009. Risk and Exposure Assessment
for Review of the Secondary National Ambient Air Quality Standards for Oxides of
Nitrogen and Sulfur. Final. U.S. Environmental Protection Agency, Office of Research
and Development, National Center for Environmental Assessment, Research Triangle
Park, NC. September.
USGS (U.S. Geological Survey). 2009. State Geological Map Compilation. U.S. Department of
the Interior, U.S. Geological Survey, Reston, VA. Available at:
http://tin.er.usgs.gov/geology/state (accessed January 28, 2009).
A-25
-------
A-26
-------
Appendix B
Methodologies and assumptions used in steady state ecosystem modeling
Technical summary of critical loads modeling in the REA
The critical load of acidity for lakes or streams was derived from present-day water
chemistry using a combination of steady-state models. Both the Steady-State Water Chemistry
(SSWC) model and First-order Acidity Balance model (FAB) are based on the principle that
excess base-cation production within a catchment area should be equal to or greater than the acid
anion input, thereby maintaining the ANC above a preselected level (Reynolds and Norris, 2001;
Posch et al. 1997). These models assume steady-state conditions and assume that all SC>42 in
runoff originates from sea salt spray and anthropogenic deposition. Given a critical ANC
protection level, the critical load of acidity is simply the input flux of acid anions from
atmospheric deposition (i.e., natural and anthropogenic) subtracted from the natural (i.e.,
preindustrial) inputs of base cations in the surface water (REA 2009 Appendix 4).
Atmospheric deposition of NOx and SOx contributes to acidification in aquatic
ecosystems through the input of acid anions, such as NO3- and SC>42 . The acid balance of
headwater lakes and streams is controlled by the level of this acidifying deposition of NCV and
SO42 and a series of biogeochemical processes that produce and consume acidity in watersheds.
The biotic integrity of freshwater ecosystems is then a function of the acid-base balance, and the
resulting acidity-related stress on the biota that occupy the water. The calculated ANC of the
surface waters is a measure of the acid-base balance:
ANC = [BC]* - [AN]* (1)
where [BC]* and [AN]* are the sum of base cations and acid anions (NCV and SC>42 ),
respectively. Equation (1) forms the basis of the linkage between deposition and surface water
acidic condition and the modeling approach used. Given some "target" ANC concentration
[ANClimit] that protects biological integrity, the amount of deposition of acid anions [AN] or
depositional load of acidity CL(A) is simply the input flux of acid anions from atmospheric
deposition that result in a surface water ANC concentration equal to the [ANClimit] when
balanced by the sustainable flux of base cations input and the sinks of nitrogen and sulfur in the
lake and watershed catchment.
B-l
-------
Critical loads for nitrogen and sulfur (CL(N) + CL(S) ) or critical load of acidity CL(A) were
calculated for each waterbody from the principle that the acid load should not exceed the
nonmarine, nonanthropogenic base cation input and sources and sinks in the catchment minus a
target ANC (ANClim) to protect selected biota from being damaged:
CL(N) + CL(S) or CL(A) = BC*dep + BCw - Ecu - AN - ANClimit (2)
Where,
BC*dep = (BC*=Ca*+Mg*+K*+Na*), nonanthropogenic deposition flux of base cations,
BCw = the average weathering flux producing base cations,
Ecu (Bc=Ca*+Mg*+K*) = the net long-term average uptake flux of base cations in the biomass
(i.e., the annual average removal of base cations due to harvesting),
AN = the net long-term average uptake, denitrification, and immobilization of nitrogen anions
(e.g. NO3") and uptake of SO42 , and
ANClimit = the lowest ANC-flux that protects the biological communities.
Since the average flux of base cations weathered in a catchment and reaching the lake or
streams is difficult to measure or compute from available information, the average flux of base
cations and the resulting critical load estimation were derived from water quality data (Henriksen
and Posch, 2001; Henriksen et al., 1992; Sverdrup et al., 1990). Weighted annual mean water
chemistry values were used to estimate average base cation fluxes, which were calculated from
water chemistry data collected from the Temporally Integrated Monitoring of Ecosystems
(TIME)/Long-Term Monitoring (LTM) monitoring networks, that include Adirondack Longterm
Monitoring (ALTM), Virginia Trout Stream Sensitivity Study (VTSSS), and the Shenandoah
Watershed Study (SWAS), and Environmental Monitoring and Assessment Program (EMAP)
(see REA Section 4.1.2.1 of Chapter 4).
The preacidification nonmarine flux of base cations for each lake or stream, BC*0, is
BC*0 = BC*dep + BCw - Ecu (3)
B-2
-------
Thus, critical load for acidity can be rewritten as
CL(N) + CL(S) = BC*0 - AN - ANClimit = Q([BC*]0 - [AN] - [ANC]limit), (4)
where the second identity expresses the critical load for acidity in terms of catchment runoff (Q)
m/yr and concentration ([x] = X/Q). The sink of nitrogen in the watershed is equal to the uptake
(Nupt), immobilization (Nimm), and denitrification (Nden) of nitrogen in the catchment. Thus,
critical load for acidity can be rewritten as
CL(N) + CL(S) = (fNupt + (1 - r)(Nimm + Nden)} + ( [BC]0* - [ANClimit])Q (5)
where f and r are dimensionless parameters that define the fraction of forest cover in the
catchment and the lake/catchment ratio, respectively. The in-lake retention of nitrogen and sulfur
was assumed to be negligible.
Equation (5) described the FAB model that was applied in the REA when sufficient data
was available to estimate the uptake, immobilization, and denitrification of nitrogen and the
neutralization of acid anions (e.g. NO3-) in the catchment. In the case were data were not
available, the contribution of nitrogen anions to acidification was assumed to be equal to the
nitrogen leaching rate (Nleach) into the surface water. The flux of acid anions in the surface
water is assumed to represent the amount of nitrogen that is not retained by the catchment, which
is determined from the sum of measured concentration of NO3- and ammonia in the stream
chemistry. This case describes the SSWC model and the critical load for acidity is
CL(A) = Q([BC*]0 - [ANC]limit) (6)
where the contribution of acid anions is considered as part of the exceedances calculation (see
REA App 4 Section 1.2.5). For the assessment of current condition in both case study areas in
the REA, the critical load calculation described in Equation (6) was used for most lakes and
streams. The lack of sufficient data for quantifying nitrogen denitrification and immobilization
prohibited the wide use of the FAB model. In addition, given the uncertainty in quantifying
B-3
-------
nitrogen denitrification and immobilization, the flux of nitrogen anions in the surface water was
assumed to more accurately reflect the contribution of NO3- to acidification. Several major
assumptions are made: (1) steady-state conditions exist, (2) the effect of nutrient cycling between
plants and soil is negligible, (3) there are no significant nitrogen inputs from sources other than
atmospheric deposition, (4) ammonium leaching is negligible because any inputs are either taken
up by biota or adsorbed onto soils or nitrate compounds, and (5) longterm sinks of sulfate in the
ratrVimpnt emle arp npaliaihlp
soils are negligible.
Pre-industrial Base Cation Concentration
The pre-industrial concentration of base cations [BC]o effectively set the long term
capacity of the catchment to neutralize acidic deposition, because it represents the only source of
base cation input that is sustainable over the long-term. Input of cations from weathering is
assumed to be a relatively constant process driven largely by the reaction of CC>2 with primary
minerals in the soils and bedrock. Base cations are removed by leaching from the soil solution
through surface water runoff. At a steady-state, the leaching rate of base cation occurs at lesser
or greater rates than the weathering supply. However, base cation leaching is not at steady-state
today because anthropogenic acid deposition actually increases the leaching of base cations
through ion-exchange within catchment soils. Soils contain a store of adsorbed base cations, as
measured as base saturation, which are derived from weathering, but have accumulated in the
soil over millennia, until eventually a steady-state is achieved, whereby the supply of base
cations from weathering was in approximate equilibrium with the removal of base cations by
rainwater, itself in equilibrium with the atmosphere. For this reason, [BC]o* cannot be derived
from measured data in runoff, but derived from a empirical relationships (i.e., pre-industrial base
cation concentration).
The pre-industrial base cation concentration is the sum of weathering ([BC w]) supply
plus base cation deposition ([BC dep]), if it is assumed that base cation deposition has not
significantly changed since pre-industrial times, minus long-term average uptake of base cations
in the biomass (i.e., the annual average removal of base cations due to harvesting):
[BC]0* = [BCW] + [BC*dep] - [BCU] (7)
B-4
-------
F-factor
An F-factor was used to correct the concentrations and estimate preindustrial base concentrations
for lakes in the Adirondack Case Study Area (REA 2009). In the case of streams in the
Shenandoah Case Study Area, the preindustrial base concentrations were derived from the
MAGIC model as the base cation supply in 1860 (hindcast) because the F-factor approach is
untested in this region. An F-factor is a ratio of the change in nonmarine base cation
concentration due to changes in strong acid anion concentrations (Henriksen, 1984; Brakke et al.,
1990):
F =([BC*]t - [BC*]0)/([SO4*]t - [SO4*]0 + [NO3*]t - [NO3*]0), (8)
where the subscripts t and 0 refer to present and preacidification conditions, respectively. If F=l,
all incoming protons are neutralized in the catchment (only soil acidification); at F=0, none of
the incoming protons are neutralized in the catchment (only water acidification). The F-factor
was estimated empirically to be in the range 0.2 to 0.4, based on the analysis of historical data
from Norway, Sweden, the United States, and Canada (Henriksen, 1984). Brakke et al. (1990)
later suggested that the F-factor should be a function of the base cation concentration:
F = sin (Ti/2 Q[BC*]t/[S]) (9)
Where
Q = the annual runoff (m/yr),
[S] = the base cation concentration at which F=l; and
for[BC*]t>[S]Fissettol.
For Norway [S] has been set to 400 milliequivalents per cubic meter (meq/m3)(circa.8
mg Ca/L) (Brakke et al., 1990). We assumed the steady-state concentration of nitrate ([AA]0)
was zero ([AA]0*= 0). The preacidification SO42- concentration in lakes, [SO4*]0, is assumed
to consist of a constant atmospheric contribution and a geologic contribution proportional to the
concentration of base cations (Brakke et al., 1989; Harriman and Christie, 1995). The
preacidification SO42- concentration in lakes, [SO4*]0 was estimated from the relationship
between [SO42-]o* and [BC]t* based on work completed by Henriksen et al., 2002 as described
by the following equation:
B-5
-------
[SO42-]o* = 15 + 0.16* [BC]t* (10)
This F value is then used to calculate the pre-industrial base cation concentration according to
the following equation:
[BCV = [BC*]t - F([AA]t* - [AA]0*) (11)
Tables B-l and B-2 list the factors used in the SSWC and FAB approaches.
B-6
-------
Table B-l Illustrates SSWC Approach - Environmental Variables
CL(A) = BC*dep + BCW - Bcu - ANClimit
CL(A) = Q ([BC*]0 - [ANC]Hmit)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Variable
Code
BC dep
BCW
Bcu
ANClimit
Ca*
Mg*
Na*
K*
SO/
CL
SO/
NO3*
Q
[BC*]0
[S0/]o
[N03*]o
F
Description
Sum (Ca*+Mg*+K*+Na*), nonanthropogenic
deposition flux of base cations
Average weathering flux of base cations
Sum (Ca+Mg+K), the net long-term average
uptake flux of base cations in the biomass
Lowest ANC-flux that protects the biological
communities
Sea Salt corrected Surface water concentration
(ueq/L) growing season average. (Ca - (CL x
0.0213))
Sea Salt corrected Surface water concentration
(ueq/L) growing season average. (Mg - (CL
x 0.0669))
Sea Salt corrected Surface water concentration
(ueq/L) growing season average. (Na - (CL x
0.557))
Sea Salt corrected Surface water concentration
(ueq/L) growing season average. (K - (CL x
0.0.0206))
Sea Salt corrected Surface water concentration
(ueq/L) growing season average. (SO4 - (CL x
0.14))
Surface water concentration (ueq/L) growing
season average.
Surface water concentration (ueq/L) growing
season average.
Surface water concentration (ueq/L) growing
season average.
The annual runoff (m/yr)
Preindustrial flux of base cations in surface
water, corrected for sea salts
Preindustrial flux of sulfate in surface water,
corrected for sea salts
Preindustrial flux of nitrate, corrected for sea
salts
Calculated factor
Source
Wet NADP and Dry
CASTNET
Calculated (5-17)
USFS-FIA data
Set
Water quality data
Water quality data
Water quality data
Water quality data
Water quality data
Water quality data
Water quality data
Water quality data
USGS
Calculated from water
quality data
Estimated
Equal to 0
Fix values
B-7
-------
Table B-2 FAB Approach - Environmental Variables
DL(N) + DL(S) = |fNupt + (1 - r)(Nimm + Nden) + (Nret + Sret)| + ( [BC]0* - [ANClimit])Q
1
2
O
4
5
6
7
8
9
10
11
12
13
14
14
15
16
17
18
19
20
Variable
Code
Ndepo
ANClimit
[BC*]0
Ca*
Mg*
Na*
K*
SO/
CL
SO/
NO3*
Q
f
r
Nret
Sret
Nupt
J^imm
Nden
Lake Size
WSH
Description
Total N deposition
Lowest ANC-flux that protects the biological
communities
Preindustrial flux of base cations in surface
water, corrected for sea salt
Sea Salt corrected Surface water concentration
(ueq/L) growing season average. (Ca - (CL x
0.0213))
Sea Salt corrected Surface water concentration
(ueq/L) growing season average. (Mg - (CL x
0.0669))
Sea Salt corrected Surface water concentration
(ueq/L) growing season average. (Na - (CL x
0.557))
Sea Salt corrected Surface water concentration
(ueq/L) growing season average. (K - (CL x
0.0.0206))
Sea Salt corrected Surface water concentration
(ueq/L) growing season average. (SO4 - (CL x
0.14))
Surface water concentration (ueq/L) growing
season average.
Surface water concentration (ueq/L) growing
season average.
Surface water concentration (ueq/L) growing
season average.
The annual runoff (m/yr)
f is a dimensionless parameter that define the
fraction of forest cover in the catchment
r is a dimensionless parameter that define the
lake/catchment ratio
The in-lake retention of nitrogen
The in-lake retention of sulfur
The net long-term average uptake flux of N in
the biomass
Immobilization of N in the soils
Denitrification
Lake size (ha)
Watershed area (ha)
Source
NADP/CMAQ
Set
Calculated from
water quality data
Water quality data
Water quality data
Water quality data
Water quality data
Water quality data
Water quality data
Water quality data
Water quality data
USGS
Estimated
Estimated
USFS-FIA data
Estimated fix value
Estimated fix value
DLMs
Calculated
-------
Data requirements for MAGIC
The MAGIC model (Cosby et al., 1985a; 1985b; 1985c) is a mathematical model (a
lumped-parameter model) of soil and surface water acidification in response to atmospheric
deposition based on process-level information about acidification. A process model, such as
MAGIC, characterizes acidification into (1) a section in which the concentrations of major ions
are assumed to be governed by simultaneous reactions involving SC>42" adsorption, cation
exchange, dissolution-precipitation- speciation of aluminum, and dissolution-speciation of
inorganic carbon; and (2) a mass balance section in which the flux of major ions to and from the
soil is assumed to be controlled by atmospheric inputs, chemical weathering, net uptake and loss
in biomass and losses to runoff. At the heart of MAGIC is the size of the pool of exchangeable
base cations in the soil. As the fluxes to and from this pool change over time owing to changes in
atmospheric deposition, the chemical equilibria between soil and soil solution shift to give
changes in surface water chemistry. The degree and rate of change of surface water acidity thus
depend both on flux factors and the inherent characteristics of the affected soils.
There are numerous input data required to run MAGIC making it rather data intensive.
Atmospheric deposition fluxes for the base cations and strong acid anions are required as inputs
to the model. These inputs are generally assumed to be uniform over the catchment. The volume
discharge for the catchment must also be provided to the model. In general, the model is
implemented using average hydrologic conditions and meteorological conditions in annual
simulations, i.e., mean annual deposition, precipitation and lake discharge are used to drive the
model. Values for soil and surface water temperature, partial pressure of carbon dioxide and
organic acid concentrations must also be provided at the appropriate temporal resolution.
The aggregated nature of the model requires that it be calibrated to observed data from a
system before it can be used to examine potential system response. Calibrations are based on
volume weighted mean annual or seasonal fluxes for a given period of observation. The length of
the period of observation used for calibration is not arbitrary. Model output will be more reliable
if the annual flux estimates used in calibration are based on a number of years rather than just
one year. There is a lot of year-to-year variability in atmospheric deposition and catchment
runoff. Averaging over a number of years reduces the likelihood that an "outlier" year (very dry,
etc.) is used to specify the primary data on which model forecasts are based. On the other hand,
B-9
-------
averaging over too long a period may remove important trends in the data that need to be
simulated by the model.
The calibration procedure requires that stream water quality, soil chemical and physical
characteristics, and atmospheric deposition data be available for each catchment. The water
quality data needed for calibration are the concentrations of the individual base cations (Ca, Mg,
Na, and K) and acid anions (Cl, SC>42", and N(V) and the pH. The soil data used in the model
include soil depth and bulk density, soil pH, soil cation-exchange capacity, and exchangeable
bases in the soil (Ca, Mg, Na, and K). The atmospheric deposition inputs to the model must be
estimates of total deposition, not just wet deposition. In some instances, direct measurements of
either atmospheric deposition or soil properties may not be available for a given site with stream
water data. In these cases, the required data can often be estimated by: (a) assigning soil
properties based on some landscape classification of the catchment; and (b) assigning deposition
using model extrapolations from some national or regional atmospheric deposition monitoring
network. Soil data for model calibration are usually derived as aerially averaged values of soil
parameters within a catchment. If soils data for a given location are vertically stratified, the soils
data for the individual soil horizons at that sampling site can be aggregated based on horizon,
depth, and bulk density to obtain single vertically aggregated values for the site, or the stratified
data can be used directly in the model.
Example of the two ways to calculate NECO
The steady-state critical load model suggested for use in the NAAQS by the PA could be
constrained by a quantity of N which would be taken up, immobilized or denitrified by
ecosystems and used to adjust the quantity of deposition required to meet a specified critical
load. This term is abbreviated by Neco, and could be derived multiple ways. The first is by taking
the mean value calculated to represent the long-term amount of N an ecosystem can immobilize
and denitrify before leaching (i.e., N saturation) that is derived from the FAB model. This
approach requires the input of multiple ecosystem parameters. Its components are expressed by
equation (13).
Neco=JNiipt + Nret+(\-r\Nimm +Nden) (13)
Where,
Nupt= nitirogen uptake by the catchment,
B-10
-------
Nimm= nitrogen immobilization by the catchment soil,
Nden=denitrification of nitrogen in the catchment,
Nret = in-lake retention of nitrogen,
f =forest cover in the catchment (dimensionless parameter),
r = fraction lake/catchment ratio (dimensionless parameter),
The second approach for estimating Neco is to take the difference between N deposition
and measured N leaching in a catchment as expressed by equation (14).
Neco=DL(N)-Nleach (14)
The site specific values of critical loads can be used to derive such a deposition loading,
here called the deposition metric, which represents a group or percentage of water bodies that
reach a specified ANC (or higher) in a given spatial area. For example, if it is desired that all
water bodies reach a specified ANC, the allowable amount of deposition for all water bodies is
equal to the lowest critical load of the population of water bodies. Because the deposition metric
represents a percentage of individual catchments from a population of water bodies, and not an
individual catchment, the deposition metric is noted by the follow abbreviation DLo/oEC0.
Two methods to calculate pre-industrial base cation weathering: F factor and MAGIC
The preindustrial concentration of base cations ([BC]o*) is calculated to represent
conditions prior to industrialization (-about 1860). It incorporates the main source of base
cations to an ecosystem including preindustrial weathering from soil and pre-industrial base
cation deposition. It is therefore considered one of the governing factors of critical loads. [BC]o
is commonly calculated using one of two approaches: dynamic modeling (i.e. MAGIC) and
calculation by the F-factor approach (Henriksen and Posch 2001).
In this section, critical load estimates obtained from two steady-state approaches were
compared. The exercise is not intended to provide an assessment of the accuracy of the two
models, but rather to provide a means for evaluating the relative performance of the two different
models. The EPA conducted analysis compared steady-state CL values based on Henriksen and
Posch (2001) F-factor approach and output from the MAGIC model. The primary purpose of the
F-factor is to obtain estimates of preindustrial surface water base cation concentrations ([BC]o )
for equation (3). The MAGIC (Cosby et al., 1985) model can also be used to derive preindustrial
B-ll
-------
surface water base cation concentrations. MAGIC is a process-based model designed to mimic
the geochemical effects of mineral weathering, soil cation exchange, and other watershed
processes. Once the model has been calibrated for a watershed, it can be run to simulate how
surface water chemistry changes with time and to predict preindustrial cation concentrations
([BC]o ) to be used in the steady-state SSWC CL model in equation (6).
The comparison of CLs between the F-factor and MAGIC approached was done for two
regions, streams in Southern Appalachia and lakes in the Adirondack Mountains. For 67 streams
and 99 lakes, [BC]o* were determined for both approaches and CLs were calculated using the
same value of Q. The results are show in Figure B-l. For this analysis, the steady-state MAGIC
model yielded critical load values that show the same trend for both regions, but were on average
16 meq S /m2-yr for the Adirondack Lakes and 5 meq S/m2"yr for Southern Appalachia streams
higher than those from the SSWC F-factor approach. The two models converge at low critical
loads, but diverge as the buffering potential for watersheds increase. This is particularly the case
for CL values above 80 meq/m2-yr for lakes in the Adirondack Mountains. These results are
consistent with similar comparison of critical loads done by Holdren et al. 1992. Holdren et al
1992 found that the MAGIC model yielded CL values that were on average 29 meq S /m2-yr
higher than those from the SSWC model. Holdren et al 1992 also found that as the buffering
potential for watersheds increased, as indicated by increasing CLs, the results from the two
models gradually diverge.
B-12
-------
A
o o
S l
g £
w •
w "-
400
300
200
o-
o
E
y = 0.9976X - 4.591
R2 = 0.8453
-100
-100
100
200
300
400
SSWC CL
MAGIC - BCo
meq/m2/yr
B
400
300
o ™
° S
V) n
(0 11.
^i
Ji
o
E
y = 0.4996x +19.541
R2 = 0.7787
-100
-100
100
200
300
400
MAGIC CL
MAGIC BCo
meq/m2/yr
Figure B-l Relationship between CLs using pre-industrial base cation weathering
([BC]0* =BC0) calculated by MAGIC versus the F-factor methods: A) 67
streams in the Southern Appalachian and B) 99 lakes in the Adirondacks
Mountains.
B-13
-------
References
Brakke, D.F., A. Henriksen, and S.A. Norton. 1989. Estimated background concentrations of
sulfate in dilute lakes. Water Resources Bulletin 25(2):247-253.
Brakke, D.F., A. Henriksen, and S.A. Norton. 1990. A variable F-factor to explain changes in
base cation concentrations as a function of strong acid deposition. International
Association of Theoretical and Applied Limnology, Proceedings 24:146-149.
Cosby, B.J., R.F. Wright, G.M. Hornberger, and J.N. Galloway. 1985a. Modelling the effects of
acid deposition: Assessment of a lumped parameter model of soil water and streamwater
chemistry. Water Resources Research 21:51-63.
Cosby, B.J., R.F. Wright, G.M. Hornberger, and J.N. Galloway. 1985b. Modelling the effects of
acid deposition: Estimation of long-term water quality responses in a small forested
catchment. Water Resources Research 21:1591-1601.
Cosby B.J., G.M. Hornberger, J.N. Galloway, and R.F. Wright. 1985c. Time scales of catchment
acidification: a quantitative model for estimating freshwater acidification. Environ Sci
Technol, 19, 1144-1149.
Harriman, R. and Christie, A. E. G.: 1995, 'Estimating Critical Loads for Biota: the Steady-state
Water Chemistry (Henriksen) Model', in Critical Loads of Acid Deposition for United
Kingdom Freshwaters, Critical Loads Advisory Group, Sub-group on Freshwaters,
Institute of Terrestrial Ecology: Edinburgh, pp. 7-8.
Henriksen, A. 1984. Changes in base cation concentrations due to freshwater acidification.
International Association of Theoretical and Applied Limnology, Proceedings 22:692-
698.
Henriksen A., P.J. Dillon, and J. Aherne. 2002. Critical loads of acidity to surface waters in
south-central Ontario, Canada: Regional application of the steady-state water chemistry
model. Can J Fish Aquat Sci, 59, 1287-1295.
Henriksen, A., and M. Posch. 2001. Steady-state models for calculating critical loads of acidity
for surface waters. Water, Air, and Soil Pollution: Focus 1:375-398.
Henriksen, A., J. Kamari, M. Posch, and A. Wilander. 1992. Critical loads of acidity: Nordic
surface waters. Ambio 21:356-363.
Holdren, G., T. Strickland, P. Shaffer, P. Ryan, P. Ringold, and R. Turner. 1992. Sensitivity of
Critical Load Estimates for Surface Waters to Model Selection and Regionalization
Schemes. Journal of Environmental Quality, 22: 279-289.
Posch, M., J. Kamari, M. Forsius, A. Henriksen, and A. Wilander. 1997. Exceedance of critical
loads for lakes in Finland, Norway and Sweden: Reduction requirements for acidifying
nitrogen and sulfur deposition. Environmental Management 21: 291-304.
Reynolds, B., and D.A. Norris. 2001. Freshwater critical loads in Wales. Water, Air, and Soil
Pollution: Focus 1:495-505.
Sverdrup, H., W. de Vries, and A. Henriksen. 1990. Mapping Critical Loads. Miljorapport 14.
Nordic Council of Ministers, Copenhagen, Denmark.
US EPA. 2009. Risk and Exposure Assessment for Review of the Secondary National Ambient
Air Quality Standards for Oxides of Nitrogen and Oxides of Sulfur-Main Content - Final
Report. U.S. Environmental Protection Agency, Washington, D.C., EPA-452/R-09-008a.
B-14
-------
Appendix C
Ecoregions, Level III: Description and Summary of Environmental Conditions
Introduction
This appendix provides descriptions of each of the level III ecoregions as described by Omernik
(U.S. EPA 2010) and used in this Policy Assessment. The level III ecoregions are presented
grouped by the level II ecoregion in which they are located. Figure C-l illustrates the location of
the level II ecoregions across the U.S. The subsequent figures illustrate the level III ecoregions
within each level II ecoregion and present a summary of the raw water quality data. Figure C-2
is a summary of the water quality data at the national level for comparison with the level III
ecoregion data. These maps also indicate the extent of ANC and critical load values available
for each ecoregion. The general regional descriptions are taken from Omernik (U.S. EPA 2010).
Omernik Ecoregion II Index Map
Eco_LevelJILUS
NA_L2NAME
| | CBJTRAL LISA PLAINS
| | CCLD DESERTS
I | E./EFSXADEE
| | MWINEWESTCOAi" =CREST
I | MISSISSIPPI ALLJ\/!.AL.1NDSOL,THEAST JSA COASTAL PLAINS | | TEXAS4_OLI3iAKA COASTAL PLAIM
| | Mixm WOOD PLAINS | | UPPER GILA MOUfcTAI MS
| | WKro WOCO SHIELD | | WARM DESERTS
| | CCARKOUACHITA-APFXyjiCHlANFOBESTS | | WEST-CglTRALSEMI-ARIDPRAIRIES
| | 2O JTH CENTRAL SE\'!-W 3 F=A.IR£S | | WESTERN CCRDLLmA
| | SOUTHEASTERN USA =LAINS I I WESTERN SIEf!F!AM.«RE RH3MCNT
I i TAMIJLPAS-TEXASSEMIASIDPLAW
| | MEDITERRArtW CALFORNIA | | TEMPERATE PRAIRIES
Figure C-l. Omernik Ecoregions, Levels II and III
C-l
-------
I §J
8-
8-
I I I I I
0 500 1000 1500 2000
ANC (ueq/L)
All Regions
§ n
- ^
8_
500 1000 1500 2000
BCo (ueq/L)
§-.
8_
8
in
8
o
I I I I I I
0 10 20 30 40 50
DOC (mg/L)
8 J
0 100 200 300 400 500
SO4 (ueq/L)
5" 0
0 CD
D"
2. o
U. 0-
o
°
1—
•
I IrhTn-i-ijri 1 1 n |
1 1 1
0123
Q (rrfyr)
s-
I
o
8H
0 50 100 150 200
N03 (ueq/L)
Figure C-2. National Water Quality Data Summary
C-2
-------
Region 5.2 Mixed Wood Shield
Reg_ll_5.2
US_L3NAME
]J Northern Lakes and Forests
^ Northern Minnesota Wetlands
o Reg_ll_5.2_.ANC
* Reg_ll_5.2_CL
Figure C-3. Region 5.2
C-3
-------
Region 5.2.1 Northern Lakes and Forests
The Northern Lakes and Forests is a region of relatively nutrient-poor glacial soils, coniferous
and northern hardwood forests, undulating till plains, morainal hills, broad lacustrine basins, and
extensive sandy outwash plains. Soils in this ecoregion are thicker than in those to the north and
generally lack the arability of soils in adjacent ecoregions to the south. The numerous lakes that
dot the landscape are clearer and less productive than those in ecoregions to the south.
Region
5.2.1
o
8-
8-
8-
8-
8-
8-
i i
0 1000
3000
ANC(ueq/L)
5000
I I I I I I
0 1000 3000 5000
BCo (ueq/L)
0 5 10 15 20 25 30
DOC (mg/L)
S-\
-------
Region 5.2.2 Northern Minnesota Wetlands
Much of the Northern Minnesota Wetlands is a vast and nearly level marsh that is sparsely
inhabited by humans and covered by swamp and boreal forest vegetation. Formerly occupied by
broad glacial lakes, most of the flat terrain in this ecoregion is still covered by standing water.
Region
5.2.2
CO
o
CD
O
CN
O
q
o
I I
0 1000
I I I I
3000 5000
cq
o
CD
O
•^
o
CN
O
q
o
ANC(ueq/L)
0 1000 3000 5000
BCo (ueq/L)
CO
o
CD
O
•^
o
CN
O
q
o
I I I I I
8 9 10 11 12
DOC (mg/L)
CO
o
>• CD
C O
q
o
50 100 150 200 250
SO4 (ueq/L)
CO
o
CD
O
q
o
q
CN
in
o
q
o
0.14
0.16 0.18
Q (rrfyr)
0.20
-1.0 -0.8 -0.6 -0.4 -0.2 0.0
N03 (ueq/L)
Figure C-5. Region 5.2.2 Water Quality Data Summary
Note: This region had only 1 data point
C-5
-------
Region 5.3 Atlantic Highlands
Reg_ll_5.3
USJ.3NAME
^\ North Central Appalachians
^| Northeastern Highlands
• Reg_ll_5.3_ANC
• Reg_ll_5.3_CL
Figure C-6. Region 5.3
C-6
-------
Region 5.3.1 Northeastern Highlands
The Northeastern Highlands cover most of the northern and mountainous parts of New England
as well as the Adirondacks and higher Catskills in New York. It is a relatively sparsely populated
region characterized by hills and mountains, a mostly forested land cover, nutrient-poor soils,
and numerous high-gradient streams and glacial lakes. Forest vegetation is somewhat transitional
between the boreal regions to the north in Canada and the broadleaf deciduous forests to the
south. Typical forest types include northern hardwoods (maple-beech-birch), northern
hardwoods/spruce, and northeastern spruce-fir forests. Recreation, tourism, and forestry are
primary land uses. Farm-to-forest conversion began in the 19th century and continues today. In
spite of this trend, alluvial valleys, glacial lake basins, and areas of limestone-derived soils are
still farmed for dairy products, forage crops, apples, and potatoes. Many of the lakes and streams
in this region have been acidified by sulfur depositions originating in industrialized areas upwind
from the ecoregion to the west.
Region
5.3.1
L
8-
8-
S-
8-,
0 1000 3000
ANC (ueq/L)
5000
i i
0 1000
3000
BCo (ueq/L)
5000
I I I I
10 15 20 25
DOC (mg/L)
55-
100 200 300
SO4 (ueq/L)
400
0.5 0.6 0.7 0.8 0.9 1.0
Q(trVyr)
0 20 40 60 80 120
NO3 (ueq/L)
Figure C-7. Region 5.3.1 Water Quality Data Summary
C-7
-------
Region 5.3.3 North Central Appalachians
More forest-covered than most adjacent ecoregions, the North Central Appalachians ecoregion is
part of a vast, elevated plateau composed of horizontally bedded sandstone, shale, siltstone,
conglomerate, and coal. It is made up of plateau surfaces, high hills, and low mountains, which,
unlike the ecoregions to the north and west, were largely unaffected by continental glaciation.
Only a portion of the Poconos section in the east has been glaciated. Land use activities are
generally tied to forestry and recreation, but some coal and natural gas extraction occurs in the
west.
Region
5.3.3
0
-------
Region 6.2 Western Cordillera
Reg_ll_6.2
US_L3NAME
Blue Mountains
^] Canadian Rockies
Cascades
Eastern Cascades Slopes and Foothills
Idaho Batholith
Klamath Mountains
Middle Rockies
| | North Cascades
Northern Rockies
Sierra Nevada
Southern Rockies
\Afesatch and Uinta Mountains
Reg_ll_6.2_ANC
Reg_ll_6.2_CL
Figure C-9. Region 6.2
C-9
-------
Region 6.2.3 Northern Rockies
The Northern Rockies ecoregion is mountainous and rugged. Despite its inland position, climate
and vegetation are, typically, marine-influenced. Douglas-fir, subalpine fir, Englemann spruce,
and ponderosa pine and Pacific indicators such as western red cedar, western hemlock, and grand
fir are found in the ecoregion. The vegetation mosaic is different from that of the Idaho Batholith
(6.2.15) and Middle Rockies (6.2.10) which are not dominated by maritime species. The
Northern Rockies ecoregion is not as high nor as snow- and ice-covered as the Canadian Rockies
(6.2.4) although alpine characteristics occur at highest elevations and include numerous glacial
lakes. Granitics and associated management problems are less extensive than in the Idaho
Batholith.
Region
6.2.3
0 1000 2000 3000
ANC (ueq/L)
I I I I
1000 2000 3000 4000
BCo (ueq/L)
DOC (mg/L)
in nmm
8-
100 200 300 400 500
0.2 0.4 0.6 0.8 1.0
SO4 (ueq/L) Q (rrfyr)
Figure C-10. Region 6.2.3 Water Quality Data Summary
5 10 15 20 25
N03 (ueq/L)
C-10
-------
Region 6.2.4 Canadian Rockies
As its name indicates, most of this region is located in Canada. It straddles the border between
Alberta and British Columbia in Canada and extends southeastward into northwestern Montana.
The region is generally higher and more ice-covered than the Northern Rockies, and portions are
strongly influenced by moist maritime air masses. Vegetation is mostly Douglas-fir, Engelmann
spruce, subalpine fir, and lodgepole pine in the forested elevations, with treeless alpine
conditions at higher elevations. A large part of the region is in national parks where tourism is
the major land use. Forestry and mining occur on the non-park lands.
Region
6.2.4
1000 2000 3000 4000
ANC (ueq/L)
1000 2000 3000 4000
BCo (ueq/L)
I \
2 3
DOC (mg/L)
CN
in _
s-
c
§ q _
0
in
o
q _
in -
--
0* CO -
c
s
D"
® CN -
O -
I
n
II
I MM
0 50 100 150 0.2 0.4 0.6 0.8 1.0 1.2 1.4
SO4 (ueq/L) Q (rrfyr)
Figure C-ll. Region 6.2.4 Water Quality Data Summary
02 4 6 8 10 12 14
N03 (ueq/L)
C-ll
-------
Region 6.2.5 North Cascades
The terrain of the North Cascades is composed of high, rugged mountains. It contains the
greatest concentration of active alpine glaciers in the conterminous United States and has a
variety of climatic zones. A dry continental climate occurs in the east and mild, maritime,
rainforest conditions are found in the west. It is underlain by sedimentary and metamorphic rock
in contrast to the adjoining Cascades (6.2.7) which are composed of volcanics.
Region
6.2.5
8-1
i irfln n n n n n
nil nn
nnn mn
I I I I I I I
0 1000 3000 5000
\ I I I I I
2000 4000 6000
ANC(ueq/L)
BCo (ueq/L)
I I
4 6
DOC (mg/L)
I
10
o !Ji -
0 100 200 300 0123
SO4 (ueq/L) Q (rrfyr)
Figure C-12. Region 6.2.5 Water Quality Data Summary
8-
i-
8-
o _
0 2 4 6 8 10 12
N03 (ueq/L)
C-12
-------
Region 6.2.7 Cascades
This mountainous ecoregion is underlain by Cenozoic volcanics and much of the region has been
affected by alpine glaciation. It is characterized by steep ridges and river valleys in the west, a
high plateau in the east, and both active and dormant volcanoes. Elevations range upwards to
14,411 feet. Its moist, temperate climate supports an extensive and highly productive coniferous
forest that is intensively managed for logging. Subalpine meadows and rocky alpine zones occur
at high elevations.
Region
6.2.7
-------
Region 6.2.8 Eastern Cascades Slopes and Foothills
The Eastern Cascade Slopes and Foothills ecoregion is in the rainshadow of the Cascade Range.
It experiences greater temperature extremes and receives less precipitation than ecoregions to the
west. Open forests of ponderosa pine and some lodgepole pine distinguish this region from the
higher ecoregions to the west where fir and hemlock forests are common, and the lower dryer
ecoregions to the east where shrubs and grasslands are predominant. The vegetation is adapted to
the prevailing dry continental climate and is highly susceptible to wildfire. Historically, creeping
ground fires consumed accumulated fuel, and devastating crown fires were less common in dry
forests. Volcanic cones and buttes are common in much of the region.
Region
6.2.8
q
CO
in
CN
q
CN
in
o
q
o
500
\ I
1500
I I I
2500 3500
q
CO
in
CN
q
CN
in
o
q
o
500
\ I
1500
ANC(ueq/L)
2500
BCo (ueq/L)
3500
\ I I I \
12345
DOC (mg/L)
I I I I I I I
0 50 100 200 300
j?
I
D"
0
j?
I
D"
0
I I I I I I
0.2 0.6 1.0
JL
_n
804 (ueq/L) Q (rrfyr)
Figure C-14. Region 6.2.8 Water Quality Data Summary
0 5 10 15 20 25 30 35
N03 (ueq/L)
C-14
-------
Region 6.2.9 Blue Mountains
The Blue Mountains ecoregion is a complex of mountain ranges that are generally lower and
more open than the neighboring Cascades (6.2.7), Northern Rockies (6.2.3), and the Idaho
Batholith (6.2.15) ecoregions. Like the Cascades, but unlike the Northern Rockies, the region is
mostly volcanic in origin. Only the few higher ranges, particularly the Wallowa and Elkhorn
Mountains, consist of granitic intrusive and metamorphic rocks that rise above the dissected lava
surface of the region. Unlike the bulk of the Cascades, Idaho Batholith, and Northern Rockies,
much of this ecoregion is grazed by cattle.
Region
6.2.9
I I I I I I
0 1000 3000 5000
o
I I I I I I
0 1000 3000 5000
ANC (ueq/L)
BCo (ueq/L)
I I
2 3
DOC (mg/L)
8-
-
j?
0 m
8-
o _
8-
rflOn
Irmi mlm
ill „
0 500 1000 1500 2000 0.0 0.2 0.4 0.6 0.8 1.0
SO4 (ueq/L) Q (rrfyr)
Figure C-15. Region 6.2.9 Water Quality Data Summary
0 5 10 15 20 25 30
N03 (ueq/L)
C-15
-------
Region 6.2.10 Middle Rockies
The climate of the Middle Rockies lacks the strong maritime influence of the Northern Rockies
(6.2.3). Mountains have Douglas-fir, subalpine fir, and Engelmann spruce forests, as well as
some large alpine areas. Pacific tree species are never dominant and forests can have open
canopies. Foothills are partly wooded or shrub- and grass-covered. Intermontane valleys are
grass- and/or shrub-covered and contain a mosaic of terrestrial and aquatic fauna that is distinct
from the nearby mountains. Many mountain-fed, perennial streams occur and differentiate the
intermontane valleys from the Northwestern Great Plains (9.3.3). Granitics and associated
management problems are less extensive than in the Idaho Batholith (6.2.15). Recreation,
logging, mining, and summer livestock grazing are common land uses.
Region
6.2.10
§-,
•Hflnm.
HrfLnnlK. -
I I I I
0 2000 6000
ANC (ueq/L)
0 2000 4000 6000 8000
BCo (ueq/L)
8-,
8-
8-
20
I
40
DOC (mg/L)
I
60
8-,
8-
8-
i i i i i i i i
0 10000 20000 30000
0.0
0.2 0.4 0.6
SO4 (ueq/L) Q (rrfyr)
Figure C-16. Region 6.2.10 Water Quality Data Summary
0.8
8-
8-
nn..
0 20 40 60 80 100
N03 (ueq/L)
C-16
-------
Region 6.2.11 Klamath Mountains
This physically and biologically diverse ecoregion covers the highly dissected ridges, foothills,
and valleys of the Klamath and Siskiyou mountains. It also extends south in California to include
the mixed conifer and montane hardwood forests that occur in the North Coast Range mountains.
The region's mix of granitic, sedimentary, metamorphic, and extrusive rocks contrasts with the
predominantly volcanic rocks of the Cascades (6.2.7) to the east. It was unglaciated during the
Pleistocene epoch, when it served as a refuge for northern plant species. The regions diverse
flora, a mosaic of both northern Californian and Pacific Northwestern conifers and hardwoods, is
rich in endemic and relic species. The mild, subhumid climate of the Klamath Mountains is
characterized by a lengthy summer drought.
Region
6.2.11
i i i i i i
0 1000 3000 5000
o _
I I I I I I I
0 1000 3000 5000
ANC(ueq/L)
BCo (ueq/L)
DOC (mg/L)
0 100 300 500 0.5 1.0 1.5 2.0
SO4 (ueq/L) Q (rrfyr)
Figure C-17. Region 6.2.11 Water Quality Data Summary
rnl I nn
20 40 60 80
N03 (ueq/L)
C-17
-------
Region 6.2.12 Sierra Nevada
The Sierra Nevada is a deeply dissected fault-block mountain range that rises sharply from the
arid basin and range ecoregions on the east and slopes gently toward the Central California
Valley to the west. The eastern portion has been strongly glaciated and generally contains higher
mountains than are found in the Klamath Mountains (6.2.11) to the northwest. Much of the
central and southern parts of the region is underlain by granite as compared to the mostly
sedimentary and metamorphic formations of the Klamath Mountains and the volcanic rocks of
the Cascades (6.2.7). The higher elevations of this region are largely federally owned and include
several national parks. The vegetation grades from mostly ponderosa pine and Douglas-fir at the
lower elevations on the west side, pines and Sierra juniper on the east side, to fir and other
conifers at the higher elevations. Alpine conditions exist at the highest elevations.
Region
6.2.12
o _
8-
I I I I I
0 500 1000 1500 2000
I I
500 1000
2000
ANC (ueq/L)
BCo (ueq/L)
o _
imn
i i
4 6 I
DOC (rrg/L)
\
10
s-
8-
o _
o _
o -
50 100
SO4 (ueq/L)
150
0.2 0.4 0.6 0.8 1.0 1.2
Q (rrfyr)
§-,
5 10 15
NO3 (ueq/L)
Figure C-18. Region 6.2.12 Water Quality Data Summary
C-18
-------
Region 6.2.13 Wasatch and Uinta Mountains
This ecoregion is composed of a core area of high, precipitous mountains with narrow crests and
valleys flanked in some areas by dissected plateaus and open high mountains. The elevational
banding pattern of vegetation is similar to that of the Southern Rockies (6.2.14) except that areas
of aspen, interior chaparral, and juniper-piny on and scrub oak are more common at middle
elevations. This characteristic, along with a far lesser extent of lodgepole pine and greater use of
the region for grazing livestock in the summer months, distinguish the Wasatch and Uinta
Mountains ecoregion from the more northerly Middle Rockies (6.2.10).
Region
6.2.13
» «-\
0
D
a-
0 o
I I I I I I I
0 1000 3000 5000
ANC(ueq/L)
1 mn HUmJI JMlli i
L
I I I I I I
0 1000 3000 5000
i
10
15
BCo (ueq/L)
DOC (mg/L)
o
I 8H
o _
L
„ ™n
(D
in
I
D"
0 CO
8-
i tlml!
0 500 1000 2000 0.0 0.1 0.2 0.3 0.4 0.5
SO4 (ueq/L) Q (rrfyr)
Figure C-19. Region 6.2.13 Water Quality Data Summary
10 20 30 40
N03 (ueq/L)
C-19
-------
Region 6.2.14 Southern Rockies
The Southern Rockies are composed of steep, rugged mountains with high elevations. Although
coniferous forests cover much of the region, as in most of the mountainous regions in the
western United States, vegetation, as well as soil and land use, follows a pattern of elevational
banding. The lowest elevations are generally grass or shrub covered and heavily grazed. Low to
middle elevations are also grazed and covered by a variety of vegetation types including
Douglas-fir, ponderosa pine, aspen, and juniper-oak woodlands. Middle to high elevations are
largely covered by coniferous forests and have little grazing activity. The highest elevations have
alpine characteristics.
Region
6.2.14
o _
g
0
D
D-
0 2000 6000
ANC(ueq/L)
1000 sooo
BCo (ueq/L)
5000
I
i 10
DOC (mg/L)
I
15
8.
s-
I §'
£
8-
§
a- o
0 •>-
j?
I
0 5000 10000 15000 0.0 0.2 0.4 0.6
SO4 (ueq/L) Q (rrfyr)
Figure C-20. Region 6.2.14 Water Quality Data Summary
0.8
20 40 60
N03 (ueq/L)
C-20
-------
Region 6.2.15 Idaho Batholith
This ecoregion is a dissected, partially glaciated, mountainous plateau. Many perennial streams
originate here and water quality can be high if basins are undisturbed. Deeply weathered, acidic,
intrusive igneous rock is common and is far more extensive than in the Northern Rockies (6.2.3)
or the Middle Rockies (6.2.10). Soils are sensitive to disturbance especially when stabilizing
vegetation is removed. Land uses include logging, grazing, and recreation. Mining and related
damage to aquatic habitat was widespread. Grand fir, Douglas-fir, and, at higher elevations,
Engelmann spruce and subalpine fir occur. Ponderosa pine, shrubs, and grasses grow in very
deep canyons. Maritime influence lessens toward the south and is never as strong as in the
Northern Rockies.
Region
6.2.15
i i i i i i
0 500 1500 2500
ANC(ueq/L)
I I I I I I
0 500 1500 2500
BCo (ueq/L)
8-i
>,
0
1 ° -
D"
0
it
in -
o -
1
|
I
1 1 Inn n n n n
1 1 1 1 1 1
0 2 4 6 8 10
DOC (mg/L)
0 100 200 300 0.2 0.4 0.6 0.8 1.0 1.2 1.4
SO4 (ueq/L) Q (rrfyr)
Figure C-21. Region 6.2.15 Water Quality Data Summary
o _
in -
o -
0 2 4 6 8 10
N03 (ueq/L)
C-21
-------
Region 7.1 Marine West Coast Forest
Reg_ll_7.1
USJ.3NAME
^] Coast Range
^] Puget Lowland
| | Wllamette Valley
* Reg_ll_7.1_ANC
* Reg_ll_7.1_CL
Figure C-22. Region 7.1
C-22
-------
Region 7.1.7 Puget Lowland
This broad rolling lowland is characterized by a mild maritime climate. It occupies a continental
glacial trough and is composed of many islands, peninsulas, and bays in the Puget Sound area.
Coniferous forests originally grew on the ecoregion's ground moraines, outwash plains,
floodplains, and terraces. The distribution of forest species is affected by the rainshadow from
the Olympic Mountains.
Region
7.1.7
q
CO
q
-------
Region 7.1.8 Coast Range
The low mountains of the Coast Range are covered by highly productive, rain-drenched
coniferous forests. Sitka spruce forests originally dominated the fog-shrouded coast, while a
mosaic of western redcedar, western hemlock, and serai Douglas-fir blanketed inland areas.
Today, Douglas-fir plantations are prevalent on the intensively logged and managed landscape.
In California, redwood forests are a dominant component in much of the region.
Region
7.1.8
o _
JLD
0 1000 2000 3000 4000
ANC (ueq/L)
0 1000 3000
BCo (ueq/L)
2 4 6
DOC (mg/L)
mmn n
0 200 400 600 800
SO4 (ueq/L)
0.5 1.0 1.5 2.0 2.5 3.0
Q (rrfyr)
Figure C-24. Region 7.1.8 Water Quality Data Summary
j
0 10 20 30 40 50
N03 (ueq/L)
C-24
-------
Region 7.1.9 Willamette Valley
This ecoregion contains terraces and floodplains of the Willamette River system, along with
scattered hills, buttes, and adjacent foothills. Originally, it was covered by prairies, oak savannas,
coniferous forests, extensive wetlands, and deciduous riparian forests. Elevation and relief are
lower and the vegetation mosaic differs from the coniferous forests of the surrounding Coast
Range (7.1.8), Cascades (6.2.7), and Klamath Mountains (6.2.11). Mean annual rainfall is 37 to
60 inches and summers are generally dry; overall, precipitation is lower than in the surrounding
mountains. Today, the Willamette Valley contains the bulk of Oregon's population, industry,
commerce, and cropland. Productive soils and a temperate climate make it one of the most
important agricultural areas in Oregon.
Region
7.1.9
CO
o
(Q
O
q
o
I I I I
400 600 800 1000
in
o
q
o
CO
o
(Q
O
ANC(ueq/L)
400 600 800 1000 1200
BCo (ueq/L)
q
o
1.0 1.5 2.0 2.5
DOC (mg/L)
CO
o
(Q
O
-------
Region 8.1 Mixed Wood Plains
Reg_ll_8.1
USJ.3NAME
^\ Acadian Plains and Hills
| | DriftlessArea
^] Eastern Great Lakes Lowlands
| | Erie Drift Plain
| | North Central Hardwood Forests
^\ Northeastern Coastal Zone
^] Northern Allegheny Plateau
| Southern Michigan/Northern Indiana Drift Plains
• Reg_ll_8.1_ANC
* Reg_ll_8.1_CL
Figure C-26. Region 8.1
C-26
-------
Region 8.1.1 Eastern Great Lakes Lowlands
This glaciated region of irregular plains bordered by hills generally contains less surface
irregularity and more agricultural activity and population density than the adjacent Northeastern
Highlands (5.3.1) and Northern Allegheny Plateau (8.1.3). Although orchards, vineyards, and
vegetable farming are important locally, a large percentage of the agriculture is associated with
dairy operations. The portion of this ecoregion that is in close proximity to the Great Lakes
experiences an increased growing season, more winter cloudiness, and greater snowfall.
Region
8.1.1
CN
O _
CO -
I I I I I
0 2000 4000 6000 8000
ANC(ueq/L)
CD
in
o _
I
0 2000 4000 6000 8000
BCo (ueq/L)
I
10
I
15
I
20
DOC (mg/L)
CN
O _
CO -
CD -
^- -
CN -
O —
[
n
m nn n n
I I I I
0 500 1000 1500
SO4 (ueq/L)
CD -
in -
I
D"
CD CO -
5"
<=
§
8-
o _
CN
O _
[tin,
0.3 0.4 0.5 0.6 0.7
Q (rrfyr)
50 100 150 200
N03 (ueq/L)
Figure C-27. Region 8.1.1 Water Quality Data Summary
C-27
-------
Region 8.1.3 Northern Allegheny Plateau
The Northern Allegheny Plateau is made up of horizontally bedded, erodible shales and
siltstones, and moderately resistant sandstones of Devonian age. It is generally lower and less
forested than the adjacent unglaciated North Central Appalachians (5.3.3). Its rolling hills, open
valleys, and low mountains are covered by till from Wisconsinan Age glaciation and the
landscape is a mosaic of cropland, pastureland, and woodland. Historically, the natural
vegetation was primarily Appalachian oak forest dominated by white oak and red oak, with some
northern hardwood forest at higher elevations. The Northern Allegheny Plateau has more level
topography and more fertile, arable land than the more rugged and forested North Central
Appalachians (5.3.3).
Region
8.1.3
o _
In n
0 500 1500 2500
ANC (ueq/L)
n i i
I I I I I I I
0 500 1500 2500
nn nnnn
i
15
BCo (ueq/L)
I
10
DOC (mg/L)
I
20
200 400 600 800
8-
8-
o _
0.35
0.45
0.55
0.65
SO4 (ueq/L) Q (rrfyr)
Figure C-28. Region 8.1.3 Water Quality Data Summary
100 200 300 400
N03 (ueq/L)
C-28
-------
Region 8.1.4 North Central Hardwood Forests
The North Central Hardwood Forests ecoregion is transitional between the predominantly
forested Northern Lakes and Forests (5.2.1) to the north and the agricultural ecoregions to the
south. Land use/land cover in this ecoregion consists of a mosaic forests, wetlands and lakes,
cropland agriculture, pasture, and dairy operations. The growing season is generally longer and
warmer than that of the Northern Lakes and Forest and the soils are more arable and fertile,
contributing to the greater agricultural component of land use. Lake trophic states tend to be
higher here than in the Northern Lakes and Forests, with higher percentages in eutrophic and
hypereutrophic classes.
Region
8.1.4
I I I I I I I
0 1000 3000 5000
u
ANC(ueq/L)
I I I I I I I
0 2000 4000 6000
BCo (ueq/L)
I I I I I
5 10 15 20 25
DOC (mg/L)
J L
i i i i i i
0 500 1500 2500
fr
I
D"
0
8-
o J
-------
Region 8.1.5 Driftless Area
The hilly uplands of the Driftless Area easily distinguish it from surrounding ecoregions. Much
of the area consists of a deeply dissected, loess-capped, bedrock dominated plateau. The region
is also called the Paleozoic Plateau because the landscape's appearance is a result of erosion
through rock strata of Paleozoic age. Although there is evidence of glacial drift in the region, its
influence on the landscape has been minor compared to adjacent ecoregions. In contrast to
adjacent ecoregions, the Driftless Area has few lakes, most of which are reservoirs with
generally high trophic states. Livestock and dairy farming are major land uses and have had a
major impact on stream quality.
Region
8.1.5
CO
o
>- CO
§ °~
IS
a-
0 ^~
it o
o
o
o
1 1 1 1 1 1
1000 3000 5000
ANC (ueq/L)
CO
o
CO
o
q
o
I I I
I
1000 3000 5000 7000
BCo (ueq/L)
in
o
I I I I I I I
1.5 2.0 2.5 3.0 3.5 4.0 4.5
DOC (rrg/L)
CO
o
-------
Region 8.1.6 Southern Michigan/Northern Indiana
Bordered by Lake Michigan on the west, this ecoregion is less agricultural than the Central Corn
Belt (8.2.3) and Eastern Corn Belt (8.2.4) to the south, it is better drained and contains more
lakes than the flat agricultural Huron/Erie Lake Plains (8.2.2) to the east, and its soils are not as
nutrient poor as Northern Lakes and Forests (5.2.1) to the north. The region is characterized by
many lakes and marshes as well as an assortment of landforms, soil types, soil textures, and land
uses. Broad till plains with thick and complex deposits of drift, paleobeach ridges, relict dunes,
morainal hills, kames, drumlins, meltwater channels, and kettles occur. Oak-hickory forests,
northern swamp forests, and beech forests were typical. Feed grain, soybean, and livestock
farming as well as woodlots, quarries, recreational development, and urban-industrial areas are
now common.
Region
8.1.6
1000 3000
ANC (ueq/L)
5000
I I I I I
1000 3000 5000
BCo (ueq/L)
I
10
\ \ \
15 20 25
n
8-
n film
0 500 1000 1500 0.20 0.25 0.30 0.35
SO4 (ueq/L) Q (rrfyr)
Figure C-31. Region 8.1.6 Water Quality Data Summary
50 100 150 200
NO3 (ueq/L)
C-31
-------
Region 8.1.7 Northeastern Coastal Zone
Similar to the Northeastern Highlands (5.3.1), the Northeastern Coastal Zone contains relatively
nutrient poor soils and concentrations of continental glacial lakes, some of which are sensitive to
acidification; however, this ecoregion contains considerably less surface irregularity and much
greater concentrations of human population. Landforms in the region include irregular plains,
and plains with high hills. Appalachian oak forests and northeastern oak-pine forests are the
natural vegetation types. Although attempts were made to farm much of the Northeastern Coastal
Zone after the region was settled by Europeans, land use now mainly consists of forests,
woodlands, and urban and suburban development, with only some minor areas of pasture and
cropland.
Region
8.1.7
8-
0firm fin n rrfL
mm n nil rnn H
I I I I I
0 500 1000 1500 2000
I I I I I
0 500 1000 1500 2000
I I I I I I I
0 5 10 15 20 25 30
ANC(ueq/L)
BCo (ueq/L)
DOC (mg/L)
hn rrmn
>. o
o o H
0
8-
kh_
0 200 400 600 800 0.50 0.60 0.70
SO4 (ueq/L) Q (rrfyr)
Figure C-32. Region 8.1.7 Water Quality Data Summary
50 100 150 200
N03 (ueq/L)
C-32
-------
Region 8.1.8 Acadian Plains and Hills
This mostly forested region, with dense concentrations of continental glacial lakes, is less rugged
than the Northeastern Highlands (5.3.1) to the west and considerably less populated than
Northeastern Coastal Zone (8.1.7) to the south. Vegetation here is mostly spruce-fir on the
lowlands with some patches of maple, beech, and birch on the hills. Soils are predominantly
frigid Spodosols. By contrast, the forests in the Northeastern Coastal Zone (8.1.7) to the south
are mostly Appalacian oak or northeastern oak-pine and the soils are generally mesic Inceptisols
and Entisols.
Region
8.1.8
8-
9-
8-
O _
mm
I I I I I
0 500 1000 1500 2000
ANC(ueq/L)
500 1000 1500
BCo (ueq/L)
I
10
I
15
I
20
DOC (mg/L)
n
8-
I I -
0
LL
8-
50 100 150 0.55 0.65 0.75
SO4 (ueq/L) Q (rrfyr)
Figure C-33. Region 8.1.8 Water Quality Data Summary
5 10 15
N03 (ueq/L)
C-33
-------
Region 8.1.10 Erie Drift Plain
Once largely covered by a maple-beech-birch forest in the west and northern hardwoods in the
east, much of the Erie Drift Plain is now in farms, many associated with dairy operations. The
Eastern Corn Belt Plains (8.2.4), which border the region on the west, are flatter, more fertile,
and therefore more agricultural. The glaciated Erie Drift Plain is characterized by low rounded
hills, scattered end moraines, kettles, and areas of wetlands, in contrast to the adjacent
unglaciated ecoregions (5.3.3, 8.4.3) to the south and east that are more hilly and less
agricultural. Areas of urban development and industrial activity occur locally. Lake Erie's
influence substantially increases the growing season, winter cloudiness, and snowfall in the
northernmost areas bordering the strip of the Eastern Great Lakes Lowland (8.1.1) which fringes
the lake.
Region
8.1.10
1000 2000 3000 4000 5000
ANC (ueq/L)
in
o
q
o
1000
3000
BCo (ueq/L)
5000
in
o
q
o
234567
DOC (rrg/L)
q
CO
in
-------
Region 8.2 Central USA Plains
Reg_ll_8.2
USJ.3NAME
| | Central Corn Belt Plains
^j Eastern Corn Belt Plains
_J Huron/Erie Lake Plains
^| Southeastern Wisconsin Till Plains
• Reg_ll_8.2_ANC
» Reg_ll_B.2_CL
Figure C-35. Region 8.2
C-35
-------
Region 8.2.1 Southeastern Wisconsin Till Plains
The Southeastern Wisconsin Till Plains support a mosaic of vegetation types, representing a
transition between the hardwood forests and oak savannas of the ecoregions to the west and the
tallgrass prairies of the Central Corn Belt Plains (8.2.3) to the south. Like Ecoregion 54, land use
in the Southeastern Wisconsin Till Plains is mostly cropland, but the crops are largely forage and
feed grains to support dairy operations, rather than corn and soybeans for cash crops. The
ecoregion has a higher plant hardiness value and a different mosaic of soils than ecoregions to
the north and west.
Region
8.2.1
q
-------
Region 8.2.2 Huron/Erie Lake Plains
The Huron/Erie Lake Plains ecoregion is a broad, fertile, nearly flat plain punctuated by relic
sand dunes, beach ridges, and end moraines. Originally, soil drainage was typically poorer than
in the adjacent Eastern Corn Belt Plains (8.2.4), and elm-ash swamp and beech forests were
dominant. Oak savanna was typically restricted to sandy, well-drained dunes and beach ridges.
Today, most of the area has been cleared and artificially drained and contains highly productive
farms producing corn, soybeans, livestock, and vegetables; urban and industrial areas are also
extensive. Stream habitat and quality have been degraded by channelization, ditching, and
agricultural activities.
Region
8.2.2
CO
o
CD
O
q
o
1 1 1 1 1 1
1
5000 5200 5400 5600
ANC (ueq/L)
CO
o
CD
O
q
o
cq
o
CD
O
I I I I I I I
5150 5250 5350 5450
q
o
BCo (ueq/L)
I I
4567
DOC (mg/L)
cq
o
CD
O
CN
O
q
o
cq
o
s
CN
o
q
o
cq
o
s
CN
o
q
o
2000 6000 10000 0.20 0.22 0.24 0.26 0.28
SO4 (ueq/L) Q (rrfyr)
Figure C-37. Region 8.2.2 Water Quality Data Summary
100 200 300 400 500
N03 (ueq/L)
C-37
-------
Region 8.2.3 Central Corn Belt Plains
Extensive prairie communities intermixed with oak-hickory forests were native to the glaciated
plains of the Central Corn Belt Plains; they were a stark contrast to the hardwood forests that
grew on the drift plains of Ecoregions 8.2.4 and 8.1.6 to the east. Ecoregions 9.2.4 and 9.2.3 to
the west were mostly treeless except along larger streams. Beginning in the nineteenth century,
the natural vegetation was gradually replaced by agriculture. Farms are now extensive on the
dark, fertile soils of the Central Corn Belt Plains and mainly produce corn and soybeans; cattle,
sheep, poultry, and, especially hogs, are also raised, but they are not as dominant as in the drier
Western Corn Belt Plains (9.2.3) to the west. Agriculture has affected stream chemistry,
turbidity, and habitat.
Region
8.2.3
q
CO
in
-------
Region 8.2.4 Eastern Corn Belt Plains
The Eastern Corn Belt Plains ecoregion is primarily a rolling till plain with local end moraines; it
had more natural tree cover and has lighter colored soils than the Central Corn Belt Plains
(8.2.3). The region has loamier and better drained soils than the Huron/Erie Lake Plain (8.2.2),
and richer soils than the Erie Drift Plain (8.1.10). Glacial deposits of Wisconsinan age are
extensive. They are not as dissected nor as leached as the pre-Wisconsinan till which is restricted
to the southern part of the region. Originally, beech forests were common on Wisconsinan soils
while beech forests and elm-ash swamp forests dominated the wetter pre-Wisconsinan soils.
Today, extensive corn, soybean, and livestock production occurs and has affected stream
chemistry and turbidity.
Region
8.2.4
q
CO
in
CN
q
CN
in
o
q
o
I I I I I
2000 4000 6000
ANC(ueq/L)
q
CO
in
CN
q
CN
in
o
q
o
I I I I I
2000 4000 6000
BCo (ueq/L)
q
CO
in
CN
q
CN
in
o
q
o
4 6
DOC (mg/L)
in
o _
nnnn
n
0 1000 2000 3000 4000 0.30 0.35
SO4 (ueq/L) Q (rrfyr)
Figure C-39. Region 8.2.4 Water Quality Data Summary
0.40
100 200 300 400
N03 (ueq/L)
C-39
-------
Region 8.3 Southeastern USA Plains
Reg_ll_8.3
USJ.3NAME
^ East Central Texas Plains
^ Interior Plateau
^ Interior River Valleys and Hills
| Mississippi Valley Loess Plains
| Northern Piedmont
Piedmont
| | South Central Plains
^ Southeastern Plains
• Reg_ll_8.3_ANC
• Reg_ll_8.3_CL
Figure C-40. Region 8.3
C-40
-------
Region 8.3.1 Northern Piedmont
The Northern Piedmont is a transitional region of low rounded hills, irregular plains, and open
valleys in contrast to the low mountains of Ecoregions 5.3.1, 8.4.4, and 8.4.1 to the north and
west and the flatter coastal plains of Ecoregions 8.3.5 and 8.5.1 to the east. It is underlain by a
mix of metamorphic, igneous, and sedimentary rocks, with soils that are mostly Alfisols and
some Ultisols. Potential natural vegetation here was predominantly Appalachian oak forest as
compared to the mostly oak-hickory-pine forests of the Piedmont (8.3.4) ecoregion to the
southwest. The region now contains a higher proportion of cropland compared to the Piedmont.
Region
8.3.1
I I I I I I I
0 500 1500 2500
(D
in
(D
in
I I
0 500
ANC (ueq/L)
1500 2500
BCo (ueq/L)
\ I I I
6 8 10 12
DOC (mg/L)
200 400 600 800
SO4 (ueq/L)
Inn n m
0.40
\ \
0.50
Q (rrfyr)
\ I
0.60
200 400 600
N03 (ueq/L)
Figure C-41. Region 8.3.1 Water Quality Data Summary
C-41
-------
Region 8.3.2 Interior River Valleys and Hills
The Interior River Lowland is made up of many wide, flat-bottomed terraced valleys, forested
valley slopes, and dissected glacial till plains. In contrast to the generally rolling to slightly
irregular plains in adjacent ecological regions to the north (8.2.3), east (8.2.4) and west (9.2.4,
9.2.3), where most of the land is cultivated for corn and soybeans, a little less than half of this
area is in cropland, about 30 percent is in pasture, and the remainder is in forest. Bottomland
deciduous forests and swamp forests were common on wet lowland sites, with mixed oak and
oak-hickory forests on uplands. Paleozoic sedimentary rock is typical and coal mining occurs in
several areas.
Region
8.3.2
q
CN
in
o
q
o
ra
III
1000
3000
I I
5000
q
CO
in
CN
q
CN
in
o
o
o
I I
1000 3000
I I I
5000 7000
q
CO
in
CN
q
CN
in
o
q
o
ANC(ueq/L)
BCo (ueq/L)
23456789
DOC (mg/L)
ill fl fl 1
I I I I
0 2000 4000 6000
SO4 (ueq/L)
o
CO -
s-
1 «-
D"
0
Ul ^ -
CN —
O —
ffl
nnnnJ i
0.20
\ I \
0.30 0.40 0.50
nm n
i i i i i i
0 100 300 500
Q (rrfyr)
N03 (ueq/L)
Figure C-42. Region 8.3.2 Water Quality Data Summary
C-42
-------
Region 8.3.3 Interior Plateau
The Interior Plateau is a diverse ecoregion extending from southern Indiana and Ohio to northern
Alabama. Rock types are distinctly different from the coastal plain sediments and alluvial
deposits of ecoregions to the west, and elevations are lower than the Appalachian ecoregions
(8.4.4, 8.4.1, 8.4.9) to the east. Mississippian to Ordovician-age limestone, chert, sandstone,
siltstone and shale compose the landforms of open hills, irregular plains, and tablelands. The
natural vegetation is primarily oak-hickory forest, with some areas of bluestem prairie and cedar
glades. The region has a diverse fish fauna.
Region
8.3.3
q
CO
in
CN
q
CN
0 in
a- ^-
in
o
I I I I I
1000 2000 3000 4000 5000
I I I I I
1000 2000 3000 4000 5000
q
o
I
10
ANC(ueq/L)
BCo (ueq/L)
DOC (mg/L)
j?
I
i i i i i i i i i i i n
0 500 1500 2500 0.35 0.45 0.55
SO4 (ueq/L) Q (rrfyr)
Figure C-43. Region 8.3.3 Water Quality Data Summary
g-
I
n mn nn n
n
0.65
0 50 100 150 200
N03 (ueq/L)
C-43
-------
Region 8.3.4 Piedmont
Considered the non-mountainous portion of the old Appalachians Highland by physiographers,
the northeast-southwest trending Piedmont ecoregion comprises a transitional area between the
mostly mountainous ecoregions of the Appalachians to the northwest and the relatively flat
coastal plain to the southeast. It is a complex mosaic of Precambrian and Paleozoic metamorphic
and igneous rocks, with moderately dissected irregular plains and some hills. The soils tend to be
finer-textured than in coastal plain regions (8.5.1, 8.3.5). Once largely cultivated, much of this
region has reverted to successional pine and hardwood woodlands, with an increasing conversion
to an urban and suburban land cover.
Region
8.3.4
8-1
0 in
is •<-
a-
I I I I I
0 500 1000 1500 2000
_n
ANC(ueq/L)
500 1000 1500
BCo (ueq/L)
4 6
DOC (mg/L)
10
0 50 100 200
SO4 (ueq/L)
I \
0.3
0.5 0.7
Q (rrfyr)
0.9
Figure C-44. Region 8.3.4 Water Quality Data Summary
8-I
S-
I 8-I
8-
0 200 600 1000
N03 (ueq/L)
C-44
-------
Region 8.3.5 Southeastern Plains
These irregular plains have a mosaic of cropland, pasture, woodland, and forest. Natural
vegetation was predominantly longleaf pine, with smaller areas of oak-hickory-pine and
Southern mixed forest. The Cretaceous or Tertiary-age sands, silts, and clays of the region
contrast geologically with the older metamorphic and igneous rocks of the Piedmont (8.3.4), and
with the Paleozoic limestone, chert, and shale found in the Interior Plateau (8.3.3). Elevations
and relief are greater than in the Southern Coastal Plain (8.5.3), but generally less than in much
of the Piedmont. Streams in this area are relatively low-gradient and sandy-bottomed.
Region
8.3.5
8-
S? 3
0
| 8'
LJ_
O
0 1000 2000 3000 4000
ANC(ueq/L)
in
co n
8-
in
-------
Region 8.3.6 Mississippi Valley Loess Plains
This ecoregion stretches from near the Ohio River in western Kentucky to Louisiana. It consists
primarily of irregular plains, some gently rolling hills, and near the Mississippi River, bluffs.
Thick loess is one of the distinguishing characteristics. The bluff hills in the western portion
contain soils that are deep, steep, silty, and erosive. Flatter topography is found to the east, and
streams tend to have less gradient and siltier substrates than in the Southeastern Plains ecoregion
(8.3.5). To the east, upland forests dominated by oak, hickory, and both loblolly and shortleaf
pine, and to the west on bluffs some mixed and southern mesophytic forests, were the dominant
natural vegetation. Agriculture is now the typical land cover in the Kentucky and Tennessee
portion of the region, while in Mississippi there is a mosaic of forest and cropland.
Region
8.3.6
CO
o
CD
O
q
o
\
\
in
o
q
o
0 500 1500 2500
ANC (ueq/L)
0 500 1500 2500
BCo (ueq/L)
CO
o
CD
O
q
o
1
1 I
2 4 6 8 10 12
DOC (mg/L)
in
o
q
o
100 150 200 250 300
in
o
q
o
0.46
0.50
0.54
SO4 (ueq/L) Q (rrfyr)
Figure C-46. Region 8.3.6 Water Quality Data Summary
q
CO
in
r-i
in
o
q
o
5 10
N03 (ueq/L)
15
C-46
-------
Region 8.3.7 South Central Texas Plains
Locally termed the "piney woods", this region of mostly irregular plains represents the western
edge of the southern coniferous forest belt. Once blanketed by a mix of pine and hardwood
forests, much of the region is now in loblolly and shortleaf pine plantations. Only about one sixth
of the region is in cropland, primarily within the Red River floodplain, while about two thirds of
the region is in forests and woodland. Lumber, pulpwood, oil and gas production are major
economic activities.
Region
8.3.7
o
I D
2000 4000
ANC(ueq/L)
6000
I I I I
2000 4000
BCo (ueq/L)
I I
6000
I I I I I I
0 5 10 15 20 25
DOC (mg/L)
j?
I
mnfl n
nn
nn
0 200 400 600 0.2 0.3 0.4 0.5
SO4 (ueq/L) Q (rrfyr)
Figure C-47. Region 8.3.7 Water Quality Data Summary
10 20 30
N03 (ueq/L)
C-47
-------
Region 8.3.8 East Central Texas Plains
Also called the Post Oak Savanna or the Claypan Area, this region of irregular plains was
originally covered by post oak savanna vegetation, in contrast to the more open prairie-type
regions to the north, south, and west and the pine forests to the east. The boundary with
Ecoregion 8.3.7 is a subtle transition of soils and vegetation. Many areas have a dense,
underlying clay pan affecting water movement and available moisture for plant growth. The
bulk of this region is now used for pasture and range.
Region
8.3.8
CO
o
>- CD
£ ci-
0
a-
it o
CN
O
O
O
50
0
1
1500
I
2500
35C
q
-------
Region 8.4 Ozark/Ouchita-Appalachian Forests
Reg_ll_8.4
USJ.3NAME
^] Arkansas Valley
| | Blue Ridge
^\ Boston Mountains
~^\ Central Appalachians
"^ Ouachita Mountains
"^ Ozark Highlands
^] Ridge and Valley
^] SouthwesternAppalachians
^] V\festern Allegheny Plateau
• Reg_ll_S.4_ANC
• Reg_ll_8.4_CL
Figure C-49. Region 8.4
C-49
-------
Region 8.4.1 Ridge and Valley
This northeast-southwest trending, relatively low-lying, but diverse ecoregion is sandwiched
between generally higher, more rugged mountainous regions with greater forest cover. As a
result of extreme folding and faulting events, the region's roughly parallel ridges and valleys
have a variety of widths, heights, and geologic materials, including limestone, dolomite, shale,
siltstone, sandstone, chert, mudstone, and marble. Springs and caves are relatively numerous.
Present-day forests cover about 50% of the region. The ecoregion has a great diversity of aquatic
habitats and species offish.
Region
8.4.1
8-,
8-
I I I I I
2000 4000 6000
ANC(ueq/L)
2000 4000 6000
BCo (ueq/L)
8-
i-
J
I
4
DOC (mg/L)
8-
8-
ILrlVi
I I I I I I I I I I I I
0 2000 4000 6000 0.4 0.6 0.8 1.0
SO4 (ueq/L) Q (rrfyr)
Figure C-50. Region 8.4.1 Water Quality Data Summary
§-,
o
8-
0
-
1 I
1000
0 200 600
N03 (ueq/L)
1400
C-50
-------
Region 8.4.2 Central Appalachians
The Central Appalachian ecoregion, stretching from central Pennsylvania to northern Tennessee,
is primarily a high, dissected, rugged plateau composed of sandstone, shale, conglomerate, and
coal. The rugged terrain, cool climate, and infertile soils limit agriculture, resulting in a mostly
forested land cover. The high hills and low mountains are covered by a mixed mesophytic forest
with areas of Appalachian oak and northern hardwood forest. Bituminous coal mines are
common, and have caused the siltation and acidification of streams.
Region
8.4.2
8-
8-
8-
£ 9-
0
2 ° ~
o _
o _
o -
JhJ
Jkn™
I I I I I I I
-2000 0 1000 3000
ANC (ueq/L)
0 2000
6000
BCo (ueq/L)
I I \
234
DOC (mg/L)
8-
0 2000 6000 10000
SO4 (ueq/L)
j?
I
s-
I
0.4
0.6 0.8
Q (rrfyr)
1.0
100 200 300 400 500
N03 (ueq/L)
Figure C-51. Region 8.4.2 Water Quality Data Summary
C-51
-------
Region 8.4.3 Western Allegheny Plateau
The hilly and wooded terrain of the Western Allegheny Plateau was not muted by glaciation and
is more rugged than the agricultural till plains of Ecoregions 8.1.10 and 8.2.4 to the north and
west, but is less rugged and not as forested as Ecoregion 8.4.2 to the east and south. Extensive
mixed mesophytic forests and mixed oak forests originally grew in the Western Allegheny
Plateau and, today, most of its rounded hills remain in forest; dairy, livestock, and general farms
as well as residential developments are concentrated in the valleys. Horizontally-bedded
sedimentary rock underlying the region has been mined for bituminous coal.
Region
8.4.3
0 1000 3000
ANC (ueq/L)
I I I I
2000 4000 6000 8000
BCo (ueq/L)
I I \
345
DOC (mg/L)
n n
s-
I co
&
I
I I I I I I I
0.35 0.45 0.55 0.65
0 2000 4000 6000 8000
SO4 (ueq/L) Q (rrfyr)
Figure C-52. Region 8.4.3 Water Quality Data Summary
100 200 300 400
N03 (ueq/L)
C-52
-------
Region 8.4.4 Blue Ridge
The Blue Ridge extends from southern Pennsylvania to northern Georgia, varying from narrow
ridges to hilly plateaus to more massive mountainous areas, with high peaks reaching over 6600
feet. The mostly forested slopes, high-gradient, cool, clear streams, and rugged terrain occur
primarily on metamorphic rocks, with minor areas of igneous and sedimentary geology. Annual
precipitation of over 100 inches can occur in the wettest areas, while dry basins can average as
little as 40 inches. The southern Blue Ridge is one of the richest centers of biodiversity in the
eastern U.S. It is one of the most floristically diverse ecoregions, and includes Appalachian oak
forests, northern hardwoods, and, at the highest elevations, Southeastern spruce-fir forests.
Shrub, grass, and heath balds, hemlock, cove hardwoods, and oak-pine communities are also
significant.
Region
8.4.4
8-
8-
i i i
500 1000 1500
8-
i-
o _
ANC (ueq/L)
500 1000 1500 2000
BCo (ueq/L)
\
10
15
DOC (rrg/L)
0 200 400 600 800 0.4 0.6 0.8 1.0 1.2
SO4 (ueq/L) Q (rrfyr)
Figure C-53. Region 8.4.4 Water Quality Data Summary
§-,
8-
=i o
a- o
0 i-
8-
flh*~
0 20 40 60 80 100
NO3 (ueq/L)
C-53
-------
Region 8.4.5 Ozark Highlands
The Ozark Highlands ecoregion has a more irregular physiography and is generally more
forested than adjacent regions, with the exception of the Boston Mountains (8.4.6) to the south.
Soils are mostly derived from cherty carbonate rocks. Cambrian and Ordovician dolomite and
sandstone comprise the dominant bedrock in the interior of the region with Mississippian
limestone underlying the western outer regions. Karst features, including caves, springs, and
spring-fed streams are found throughout most of the Ozark Highlands. The majority of the region
is forested; oak is the predominant forest type but mixed stands of oak and pine are also
common, with pine concentrations greatest to the southeast. Less than one fourth of the core of
this region has been cleared for pasture and cropland, but half or more of the periphery, while not
as agricultural as bordering ecological regions, is in cropland and pasture.
Region
8.4.5
q
CO
in
-------
Region 8.4.6 Boston Mountains
In contrast to the nearby Ouachita Mountains (8.4.8) region which comprises folded and faulted
linear ridges mostly covered by pine forests, the Boston Mountains ecological region consists of
a deeply dissected sandstone and shale plateau, originally covered by oak-hickory forests. Red
oak, white oak, and hickory remain the dominant vegetation types in this region, although
shortleaf pine and eastern red cedar are found in many of the lower areas and on some south- and
west-facing slopes. The region is sparsely populated and recreation is a principal land use.
Region
8.4.6
I I I I
0 200 600
1000
ANC(ueq/L)
1400
q
-------
Region 8.4.7 Arkansas Valley
A region of mostly forested valleys and ridges, the physiography of the Arkansas Valley is much
less irregular than that of the Boston Mountains (8.4.6) to the north and the Ouachita Mountains
(8.4.8) to the south, but is more irregular than the ecological regions to the west and east. About
one fourth of the region is grazed and roughly one tenth is cropland. In the Arkansas Valley,
even streams that have been relatively unimpacted by human activities have considerably lower
dissolved oxygen levels, and hence support different biological communities, than those of most
of the adjacent regions.
Region
8.4.7
CN _
O _
CO -
I I I I I I
0 1000 3000 5000
ANC(ueq/L)
in -
CO -
CN -
1
0
III 1
1 1 1 1 1 1
500 1500 2500
BCo (ueq/L)
CD
in
46
DOC (mg/L)
\ \ \
8 10 12
uency
in
o
q
o
100 200 300 400 0.2 0.3 0.4
SO4 (ueq/L) Q (rrVyr)
Figure C-56. Region 8.4.7 Water Quality Data Summary
0.5
0 !£ -
In II
m n
I I I I I I
0 50 100 150 200 250
N03 (ueq/L)
C-56
-------
Region 8.4.8 Ouchita Mountains
The Ouachita Mountains ecological region is made up of sharply defined east-west trending
ridges, formed through erosion of compressed sedimentary rock formations. The Ouachitas are
structurally different from the Boston Mountains (8.4.6), more folded and rugged than the
lithologically distinct Ozark Highlands (8.4.5), and physiographically unlike the Arkansas Valley
(8.4.7), South Central Plains (8.3.7), and Mississippi Alluvial Plain (8.5.2). Potential natural
vegetation is oak-hickory-pine forest, which contrasts with the oak-hickory forest that dominates
Ecoregion 8.4.5 and the northern part of the Boston Mountains (8.4.6). Most of this region is
now in loblolly and shortleaf pine. Commercial logging is the major land use in the region.
Region
8.4.8
o
I I I I
0 500 1000 1500
i i r
0 200
ANC(ueq/L)
600
BCo (ueq/L)
1 1 1 1
1000 1400
DOC (rr^/L)
j?
I
D"
0
mn
50
150 250 350 0.35 0.45
SO4 (ueq/L) Q (rrfyr)
Figure C-57. Region 8.4.8 Water Quality Data Summary
0.55
5 10 15
N03 (ueq/L)
C-57
-------
Region 8.4.9 Southwestern Appalachians
Stretching from Kentucky to Alabama, these open low mountains contain a mosaic of forest and
woodland with some cropland and pasture. The eastern boundary of the ecoregion, along the
more abrupt escarpment where it meets the Ridge and Valley (8.4.1), is relatively smooth and
only slightly notched by small, eastward flowing streams. Much of the western boundary, next to
the Interior Plateau (8.3.3), is more crenulated, with a rougher escarpment that is more deeply
incised. The mixed mesophytic forest is restricted mostly to the deeper ravines and escarpment
slopes, and the upland forests are dominated by mixed oaks with shortleaf pine. Ecoregion 8.4.9
has less agriculture than the adjacent Ecoregion 8.3.3. Coal mining occurs in several parts of the
region.
Region
8.4.9
0 1000 3000
ANC (ueq/L)
5000
i i
0 1000 3000 5000
BCo (ueq/L)
4 6 8 10 12
DOC (mg/L)
m
0 5000 15000 25000 0.60 0.65 0.70
SO4 (ueq/L) Q (rrfyr)
Figure C-58. Region 8.4.9 Water Quality Data Summary
0.75
50 100 150
N03 (ueq/L)
200
C-58
-------
Region 8.5 Mississippi Alluvial and Southeastern USA Coastal Plains
Reg_ll_8.5
USJ.3NAME
_J Atlantic Coastal Pine Barrens
| | Middle Atlantic Coastal Plain
^| Mississippi Alluvial Plain
^] Southern Coastal Plain
• Reg_ll_8.5_ANC
• Reg_ll_8.5_CL
Figure C-59. Region 8.5
C-59
-------
Region 8.5.1 Middle Atlantic Coastal Plain
The Middle Atlantic Coastal Plain ecoregion stretches from Delaware to the South
Carolina/Georgia border and consists of low elevation flat plains, with many swamps, marshes,
and estuaries. Forest cover in the region, once dominated by longleaf pine in the Carolinas, is
now mostly loblolly and some shortleaf pine, with patches of oak, gum, and cypress near major
streams, as compared to the mainly longleaf-slash pine forests of the warmer Southern Coastal
Plain (8.5.3). Its low terraces, marshes, dunes, barrier islands, and beaches are underlain by
unconsolidated sediments. Poorly drained soils are common, and the region has a mix of coarse
and finer textured soils compared to the mostly coarse soils in the majority of Ecoregion 8.5.3.
The Middle Atlantic Coastal Plain is typically lower, flatter, more poorly drained, and more
marshy than Ecoregion 8.3.5. Less cropland occurs in the southern portion of the region than in
the central and northern parts.
Region
8.5.1
I I I I
0 500 1000 1500
q
CO
in
-------
Region 8.5.2 Mississippi Alluvial Plain
This riverine ecoregion extends from southern Illinois, at the confluence of the Ohio River with
the Mississippi River, south to the Gulf of Mexico. It is mostly a broad, flat alluvial plain with
river terraces, swales, and levees providing the main elements of relief. Soils are typically finer-
textured and more poorly drained than the upland soils of adjacent Ecoregions 8.3.7 and 8.3.6,
although there are some areas of coarser, better-drained soils. Winters are mild and summers are
hot, with temperatures and precipitation increasing from north to south. Bottomland deciduous
forest vegetation covered the region before much of it was cleared for cultivation. Presently,
most of the northern and central parts of the region are in cropland and receive heavy treatments
of insecticides and herbicides. Soybeans, cotton, and rice are the major crops.
Region
8.5.2
q
CO
in
-------
Region 8.5.3 Southern Coastal Plains
The Southern Coastal Plain consists of mostly flat plains, but it is a heterogeneous region
containing barrier islands, coastal lagoons, marshes, and swampy lowlands along the Gulf and
Atlantic coasts. In Florida, an area of discontinuous highlands contains numerous lakes. This
ecoregion is lower in elevation with less relief and wetter soils than the Southeastern Plains
(8.3.5). It is warmer, more heterogeneous, and has a longer growing season and coarser textured
soils than the Middle Atlantic Coastal Plain (8.5.1). Once covered by a variety of forest
communities that included trees of longleaf pine, slash pine, pond pine, beech, sweetgum,
southern magnolia, white oak, and laurel oak, land cover in the region is now mostly slash and
loblolly pine with oak-gum-cypress forest in some low lying areas, citrus groves in Florida,
pasture for beef cattle, and urban.
Region
8.5.3
8-
8-
nrfflrnn n n
I I I I
0 1000 2000 3000
ANC (ueq/L)
o _
nil nn
i i i i i i
0 500 1500 2500
BCo (ueq/L)
I I I
50 100 150
DOC (rrg/L)
o in
0 *~
>. o
o o H
0
Inn nn m
0 500 1500 2500 0.2 0.3 0.4 0.5 0.6 0.7
SO4 (ueq/L) Q (rrfyr)
Figure C-62. Region 8.5.3 Water Quality Data Summary
8-
100 200 300 400 500
NO3 (ueq/L)
C-62
-------
Region 8.5.4 Atlantic Coastal Pine Barrens
This is a transitional ecoregion, distinguished from the coastal ecoregion (8.5.1) to the south by
its coarser-grained soils, cooler climate, and Northeastern oak-pine potential natural vegetation.
The climate is milder than the coastal ecoregion (8.1.7) to the north that contains Appalachian
oak forests and some northern hardwoods forests. The physiography of this ecoregion is not as
flat as that of the Middle Atlantic Coastal Plain (8.5.1), but it is not as irregular as that of the
Northeastern Coastal Zone (8.1.7). The shore characteristics of sandy beaches, grassy dunes,
bays, marshes, and scrubby oak-pine forests are more like those to the south, in contrast to the
more rocky, jagged, forested coastline found to the north.
Region
8.5.4
n nrm n
I I I I I
-500 0 500 1000 1500
I I I I I I I
0 100 300 500
I I I I I I I
0 5 10 15 20 25 30
ANC(ueq/L)
BCo (ueq/L)
DOC (mg/L)
h- -
CD -
in -
£ --
§
D"
0 CO -
LL
CN -
I
&
I CO
D"
0
8-
8-
llnih™
100 300 500 0.50 0.55 0.60
SO4 (ueq/L) Q (rrfyr)
Figure C-63. Region 8.5.4 Water Quality Data Summary
0 100 300 500
N03 (ueq/L)
C-63
-------
Region 9.2 Temperate Prairies
Reg_ll_9.2
USJ.3NAME
^| Central Irregular Plains
^J Lake Agassiz Plain
^J Northern Glaciated Plains
[ 1 Western Corn Belt Plains
• Reg_ll_92_ANC
• Reg_ll_9.2_CL
Figure C-64. Region 9 2
C-64
-------
Region 9.2.1 Northern Glaciated Plains
The Northern Glaciated Plains ecoregion is characterized by a flat to gently rolling landscape
composed of glacial drift. The subhumid conditions foster a grassland transitional between tall
and shortgrass prairie. High concentrations of temporary and seasonal wetlands create favorable
conditions for waterfowl nesting and migration. Although the till soils are very fertile,
agricultural success is subject to annual climatic fluctuations.
Region
9.2.1
J D
(D
in
>,
i ^
0
D
0> CO
LJ_
-------
Region 9.2.2 Lake Agassiz Plains
Glacial Lake Agassiz was the last in a series of proglacial lakes to fill the Red River valley in the
three million years since the beginning of the Pleistocene. Thick beds of lake sediments on top of
glacial till create the extremely flat floor of the Lake Agassiz Plain. The historic tallgrass prairie
has been replaced by intensive row crop agriculture. The preferred crops in the northern half of
the region are potatoes, beans, sugar beets, and wheat; soybeans, sugar beets, and corn
predominate in the south.
Region
9.2.2
q
CO
in
-------
Region 9.2.3 Western Corn Belt Plains
Once mostly covered with tallgrass prairie, over 80 percent of the Western Corn Belt Plains is
now used for cropland agriculture and much of the remainder is in forage for livestock. A
combination of nearly level to gently rolling glaciated till plains and hilly loess plains, an
average annual precipitation of 26 to 37 inches, which occurs mainly in the growing season, and
fertile, warm, moist soils make this one of the most productive areas of corn and soybeans in the
world. Agricultural practices have contributed to environmental issues, including surface and
groundwater contamination from fertilizer and pesticide applications as well as concentrated
livestock production.
Region
9.2.3
i r
2000
\ r
4000
ANC(ueq/L)
6000
I I I I
2000 4000 6000 8000
o _
I I I I I I I
0 10 20 30 40 50 60
BCo (ueq/L)
DOC (mg/L)
o ^
0 5000 10000 15000 0.05 0.10 0.15 0.20
SO4 (ueq/L) Q (rrfyr)
Figure C-67. Region 9.2.3 Water Quality Data Summary
500 1000 1500
N03 (ueq/L)
C-67
-------
Region 9.2.4 Central Irregular Plains
The Central Irregular Plains have a mix of land use and are topographically more irregular than
the Western Corn Belt Plains (9.2.3) to the north, where most of the land is in crops. The region,
however, is less irregular and less forest covered than the ecoregions to the south and east. The
potential natural vegetation of this ecological region is a grassland/forest mosaic with wider
forested strips along the streams compared to Ecoregion 9.2.3 to the north. The mix of land use
activities in the Central Irregular Plains includes mining operations of high-sulfur bituminous
coal. The disturbance of these coal strata in southern Iowa and northern Missouri has degraded
water quality and affected aquatic biota.
Region
9.2.4
q
CO
in
CN
q
CN
in
o
q
o
I I I I
1000 2000 3000 4000
ANC(ueq/L)
I I I I I
1000 2000 3000 4000 5000
q
CO
in
CN
q
CN
in
o
q
o
BCo (ueq/L)
234567
DOC (mg/L)
I I I I I I
0 2000 6000 10000
SO4 (ueq/L)
s-
I
D"
0
m _
j?
I
0
I I I I I I I
0.16 0.20 0.24 0.28
I I I I I I
0 50 150 250
Q (rrfyr)
N03 (ueq/L)
Figure C-68. Region 9.2.4 Water Quality Data Summary
C-68
-------
Region 9.3 West-Central Semi-Arid Prairies
Reg_ll_9.3
US_L3NAME
j^B Nebraska Sand Hills
Northwestern Glaciated Plains
Northwestern Great Plains
RegJI_9.3_ANC
Reg_ll_9.3_CL
Figure C-69. Region 9.3
C-69
-------
Region 9.3.1 Northwestern Glaciated Plains
The Northwestern Glaciated Plains ecoregion is a transitional region between the generally more
level, moister, more agricultural Northern Glaciated Plains (9.2.1) to the east and the generally
more irregular, dryer, Northwestern Great Plains (9.3.3) to the west and southwest. The western
and southwestern boundary roughly coincides with the limits of continental glaciation. Pocking
this ecoregion is a moderately high concentration of semi-permanent and seasonal wetlands,
locally referred to as Prairie Potholes.
Region
9.3.1
0 20000 60000
ANC (ueq/L)
q
CO
in
CN
in
o
o
o
2000 4000 6000 8000 10000
BCo (ueq/L)
i
10
20 30
DOC (mg/L)
40
^-
CN
O
£ CO
s
O-
0 CD
8-1
i-
o _
0 20000 60000 0.00 0.05 0.10 0.15 0.20
SO4 (ueq/L) Q (rrfyr)
Figure C-70. Region 9.3.1 Water Quality Data Summary
0 10 20 30 40 50 60 70
N03 (ueq/L)
C-70
-------
Region 9.3.3 Northwestern Great Plains
The Northwestern Great Plains ecoregion encompasses the Missouri Plateau section of the Great
Plains that is mostly unglaciated. It is a semiarid rolling plain of shale, siltstone, and sandstone
punctuated by occasional buttes and badlands. Rangeland is common, but spring wheat and
alfalfa farming also occur; native grasslands, persist in areas of steep or broken topography.
Agriculture is restricted by the erratic precipitation and limited opportunities for irrigation.
Region
9.3.3
8-
o _
I I I I
0 20000 40000 60000
ANC(ueq/L)
2000 4000 6000 8000
BCo (ueq/L)
hmn
J
0 10 20 30 40
DOC (mg/L)
8-
i-
8-
o _
-------
Region 9.3.4 Nebraska San Hills
The Nebraska Sandhills comprise one of the most distinct and homogenous ecoregions in North
America. One of the largest areas of grass stabilized sand dunes in the world, this region is
generally devoid of cropland agriculture and except for some riparian areas in the north and east,
and the region is treeless. Large portions of this ecoregion contain numerous lakes and wetlands
and have a lack of streams. The area is sparsely populated; however, large cattle ranches are
found throughout the region.
Region
9.3.4
0 20000 60000
ANC (ueq/L)
q
CO
in
CN
q
CN
in
o
q
o
I I I I
2000 4000 6000 8000
q
CO
in
CN
q
CN
in
o
q
o
BCo (ueq/L)
0 10 20 30 40 50 60 70
DOC (mg/L)
o _
m n
uency
3
1
m _
linn
0 5000 15000 25000 0.02 0.04 0.06 0.08 0.10
SO4 (ueq/L) Q (rrfyr)
Figure C-72. Region 9.3.4 Water Quality Data Summary
0 10 20 30 40 50 60
N03 (ueq/L)
C-72
-------
Region 9.4 South Central Semi-Arid Prairies
Reg_ll_9.4
US_L3NAME
| | Central Great Plains
| Cross Timbers
^] Edwards Plateau
| | Flint Hills
^\ High Plains
^] Southwestern Tablelands
^| Texas Blackland Prairies
• Reg_ll_9.4_ANC
• Reg_ll_9.4_CL
Figure C-73. Region 9.4
C-73
-------
Region 9.4.1 High Plains
Higher and drier than the Central Great Plains (9.4.2) to the east, and in contrast to the irregular,
mostly grassland or grazing land of the Northwestern Great Plains (9.3.3) to the north, much of
the High Plains is characterized by smooth to slightly irregular plains having a high percentage
of cropland. Grama-buffalo grass is the potential natural vegetation in this region as compared to
mostly wheatgrass-needlegrass to the north, Trans-Pecos shrub savanna to the south, and taller
grasses to the east. The northern boundary of this ecological region is also the approximate
northern limit of winter wheat and sorghum and the southern limit of spring wheat.
Region
9.4.1
q
CO
in
-------
Region 9.4.2 Central Great Plains
The Central Great Plains are slightly lower, receive more precipitation, and are somewhat more
irregular than the High Plains (9.4.1) to the west. Once grassland, with scattered low trees and
shrubs in the south, much of this ecological region is now cropland, the eastern boundary of the
region marking the eastern limits of the major winter wheat growing area of the United States.
Subsurface salt deposits and leaching contribute to high salinity found in some streams.
Region
9.4.2
(D -
in -
co -
-------
Region 9.4.3 Southwestern Tablelands
The southwestern Tablelands flank the High Plains (9.4.1) with red hued canyons, mesas,
badlands, and dissected river breaks. Unlike most adjacent Great Plains ecological regions, little
of the Southwestern Tablelands is in cropland. Much of this region is in sub-humid grassland and
semiarid range land. The potential natural vegetation is grama-buffalo grass with some mesquite-
buffalo grass in the southeast, juniper-scrub oak-midgrass savanna on escarpment bluffs, and
shinnery (midgrass prairie with open low and shrubs) along the Canadian River.
Region
9.4.3
q
CO
in
-------
Region 9.4.4 Flint Hills
The Flint Hills is a region of rolling hills with relatively narrow steep valleys, and is composed
of shale and cherty limestone with rocky soils. In contrast to surrounding ecological regions that
are mostly in cropland, most of the Flint Hills region is grazed by beef cattle. The Flint Hills
mark the western edge of the tallgrass prairie, and contain the largest remaining intact tallgrass
prairie in the Great Plains.
Region
9.4.4
q
-------
Region 9.4.5 Cross Timbers
The Cross Timbers ecoregion is a transition area between the once prairie, now winter wheat
growing regions to the west, and the forested low mountains or hills of eastern Oklahoma and
Texas. The region does not possess the arability and suitability for crops such as corn and
soybeans that are common in the Central Irregular Plains (9.2.4) to the northeast. Transitional
"cross-timbers" (little bluestem grassland with scattered blackjack oak and post oak trees) is the
native vegetation, and presently rangeland and pastureland comprise the predominant land cover,
with some areas of woodland. Oil extraction has been a major activity in this region for over
eighty years.
Region
9.4.5
1000 3000
ANC(ueq/L)
5000
1000
3000
BCo (ueq/L)
5000
I I I I
6 8 10 12
DOC (mg/L)
1000 2000 3000 4000
SO4 (ueq/L)
s-
I
D"
0
s-
I
D"
0
0.05
0.10 0.15
Q (rrfyr)
0.20
0 10 20 30 40
N03 (ueq/L)
Figure C-78. Region 9.4.5 Water Quality Data Summary
C-78
-------
Region 9.4.6 Edwards Plateau
This ecoregion is largely a dissected limestone plateau that is hillier in the south and east where
it is easily distinguished from bordering ecological regions by a sharp fault line. The region
contains a sparse network of perennial streams, but due to karst topography and resultant
underground drainage they are relatively clear and cool compared to those of surrounding areas.
Originally covered by juniper-oak savanna and mesquite-oak savanna, most of the region is used
for grazing beef cattle, sheep, goats, and wildlife. Hunting leases are a major source of income.
Region
9.4.6
cq
o
c o
0
D
a-
q
o
3000 4000 5000
ANC (ueq/L)
-
r
1 1 1 1 1
3000 4000 5000
BCo (ueq/L)
q
CN
in
o
q
o
I
2.0
I
3.0
I
4.0
5.0
DOC (mg/L)
CO
o
CD
O
CN
O
q
o
s-
0
"
in
o
q
o
q
CO
in
CN
in
o
q
o
280 300 320 340 360 380 0.02 0.06 0.10
SO4 (ueq/L) Q (rrfyr)
Figure C-79. Region 9.4.6 Water Quality Data Summary
246
N03 (ueq/L)
C-79
-------
Region 9.4.7 Texas Blackland Prairies
The Texas Blackland Prairies form a disjunct ecological region, distinguished from surrounding
regions by its fine-textured, clayey soils and predominantly prairie potential natural vegetation.
This region now contains a higher percentage of cropland than adjacent regions, and pasture and
forage production for livestock is common. Large areas of the region are being converted to
urban and industrial uses.
Region
9.4.7
CO
o
CD
O
q
o
1 1
I I I I
1500 2500 3500 450
ANC(ueq/L)
cq
o
CD
o
•^
o
q
o
\ \ \ i i i
1500 2500 3500 4500
BCo (ueq/L)
CO
o
CD
O
•^
o
CN
O
q
o
3456789 10
DOC (mg/L)
in
o
q
o
in
o
q
o
0 1000 2000 3000 4000 0.05 0.10 0.15 0.20 0.25
SO4 (ueq/L) Q (rrfyr)
Figure C-80. Region 9.4.7 Water Quality Data Summary
0 10 20 30 40 50
N03 (ueq/L)
C-80
-------
Region 9.5 Texas-Louisiana Coastal Plain
USJ.3NAME
| | V\festern Gulf Coastal Plain
« Reg_ll_9.5_ANC
• Reg_ll_9.5_CL
Figure C-81. Region 9.5
C-81
-------
Region 9.5.1 Western Gulf Coastal Plain
The principal distinguishing characteristics of the Western Gulf Coastal Plain are its relatively
flat coastal plain topography and mainly grassland potential natural vegetation. Inland from this
region the plains are older, more irregular, and have mostly forest or savanna-type vegetation
potentials. Largely because of these characteristics, a higher percentage of the land is in cropland
than in bordering ecological regions. Urban and industrial land uses have expanded greatly in
recent decades, and oil and gas production is common.
Region
9.5.1
CO
o
c o
0
D
a-
q
o
1000 3000
ANC(ueq/L)
5000
CO
o
CO
o
q
o
1000 3000 5000
BCo (ueq/L)
q
-------
Region 9.6 Tamaulipas-Texas Semi-Arid Plain
USJ.3NAME
^ Southern Texas Plains
* Reg_ll_9.B_CL
Figure C-83. Region 9.6
C-83
-------
Region 9.6.1 Southern Texas Plains
This rolling to moderately dissected plain was once covered with grassland and savanna
vegetation that varied during wet and dry cycles. Following long continued grazing and fire
suppression, thorny brush, such as mesquite, is now the predominant vegetation type. Also
known as the Tamualipan Thornscrub, or the "brush country", as it is called locally, the
subhumid to dry region has its greatest extent in Mexico. It is generally lower in elevation with
warmer winters than the Chihuahuan Deserts (10.2.10) to the northwest, and it contains a high
and distinct diversity of plant and animal life. Oil and natural gas production activities are
widespread.
Region
9.6.1
CO
o
c
0
D
a-
q
o
I I I I I
1600 1800 2000 2200 2400
ANC (ueq/L)
I I I I I
3000 4000 5000
BCo (ueq/L)
CO
o
(Q
O
q
o
I I
5.0 6
I I I I
.0 7.0 8.0
DOC (mg/L)
CO
o
,
c
0 q _
0
it
o ~
o
•^ ~
CO
o
>, (D
i 0
§
D"
0 •^•
it 0 ~
-------
Region 10.1 Cold Deserts
Reg_ll_10.1
US_L3NAME
"^ Arizona/New Mexico Plateau
"^ Central Basin and Range
^J Colorado Plateaus
^| Columbia Plateau
^| Northern Basin and Range
^| Snake River Plain
| Vtyoming Basin
* Reg_ll_10.1_ANC
* Reg_ll_1D.1_CL
f* /101.2
*
Figure C-85. Region 10.1
C-85
-------
Region 10.1.2 Columbia Plateau
The Columbia Plateau is an arid sagebrush steppe and grassland, surrounded on all sides by
moister, predominantly forested, mountainous ecological regions. This region is underlain by
basalt up to two miles thick. It is covered in some places by loess soils that have been
extensively cultivated for wheat, particularly in the eastern portions of the region where
precipitation amounts are greater. During the glaciation of the Pleistocene era, parts of the area
were scoured to bedrock by huge floods from breached ice dams.
Region
10.1.2
I I I I
0 5000 10000 15000
q
CO
in
-------
Region 10.1.3 Northern Basin and Range
This ecoregion contains tablelands, dissected lava plains, valleys, alluvial fans, and scattered
mountains. Overall, it is higher and cooler than the Snake River Plain (10.1.8) to the east and has
more available moisture and a cooler climate than the Central Basin and Range (10.1.5) to the
south. The region has more extensive basins and fewer mountain ranges than the Central Basin
and Range. Non-mountain areas have sagebrush steppe vegetation; cool season grasses and
Mollisols are more common than in the hotter-drier basins of the Central Basin and Range where
Aridisols are dominated by sagebrush, shadscale, and greasewood. Ranges are covered in
mountain sagebrush, mountain brush, and Idaho fescue at lower and mid-elevations; Douglas-fir,
and aspen are common at higher elevations. Soils are less suitable for agriculture than those in
the Columbia Plateau (10.1.2) and the Snake River Plain. Rangeland is common and dryland and
irrigated agriculture occur in eastern basins.
Region
10.1.3
0 2000 6000
ANC (ueq/L)
\ \
10000
2000 6000
BCo (ueq/L)
0 5 10 15 20 25 30
JL
0 2000 6000 10000 0.05 0.10 0.15 0.20 0.25
SO4 (ueq/L) Q (rrfyr)
Figure C-87. Region 10.1.3 Water Quality Data Summary
0 100 300 500 700
NO3 (ueq/L)
C-87
-------
Region 10.1.4 Wyoming Basin
This ecoregion is a broad intermontane basin interrupted by hills and low mountains and
dominated by arid grasslands and shrublands. Nearly surrounded by forest covered mountains,
the region is somewhat drier than the Northwestern Great Plains (9.3.3) to the northeast and does
not have the extensive cover of pinyon-juniper woodland found in the Colorado Plateaus (10.16)
to the south. Much of the region is used for livestock grazing, although many areas lack
sufficient vegetation to support this activity. The region contains major producing natural gas
and petroleum fields. The Wyoming Basin also has extensive coal deposits along with areas of
trona, bentonite, clay, and uranium mining.
Region
10.1.4
I
2000 4000 6000
ANC(ueq/L)
0 2000 6000
BCo (ueq/L)
I
10
DOC (mg/L)
ll
fr
I
D"
0
10000 20000 30000
fr
I
D"
0
O _
0.0 0.1 0.2 0.3 0.4 0.5
iJ
n_
SO4 (ueq/L) Q (rrfyr)
Figure C-88. Region 10.1.4 Water Quality Data Summary
0 10 20 30 40 50 60
N03 (ueq/L)
c-s
-------
Region 10.1.5 Central Basin and Range
The Central Basin and Range ecoregion is internally drained and is characterized by a mosaic of
xeric basins, scattered low and high mountains, and salt flats. It has a hotter and drier climate,
more shrubland, and more mountain ranges than the Northern Basin and Range (10.1.3)
ecoregion to the north. Basins are covered by Great Basin sagebrush or saltbush-greasewood
vegetation that grow in Aridisols; cool season grasses are less common than in the Mollisols of
the Snake River Plain (10.1.8) and Northern Basin and Range. The region is not as hot as the
Mojave Basin and Range (10.2.1) ecoregion to the south and it has a greater percent of land that
is grazed.
Region
10.1.5
8-
I I
0 20000
60000
ANC(ueq/L)
o
0 2000 6000
BCo (ueq/L)
10000
I I
4 6
DOC (mg/L)
10
0 2000 4000 6000 8000
SO4 (ueq/L)
j?
1
D"
0
LJ_
8-1
8-
O _
CN
^~
O _
in -
o —1
111 [IfJn 11 Jin n nn m H
8-
8-
o _
0.0 0.2 0.4 0.6 0.8
Q (rrfyr)
100 200 300 400 500
N03 (ueq/L)
Figure C-89. Region 10.1.5 Water Quality Data Summary
C-89
-------
Region 10.1.6 Colorado Plateaus
Ecoregion 10.1.6 is an uplifted, eroded, and deeply dissected tableland. Its benches, mesas,
buttes, salt valleys, cliffs, and canyons are formed in and underlain by thick layers of
sedimentary rock. Precipitous side-walls mark abrupt changes in local relief, often from 1,000 to
2,000 feet. The region contains a greater extent of pinyon-juniper and Gambel oak woodlands
than the Wyoming Basin (10.1.4) to the north. There are also large low lying areas containing
saltbrush-greasewood (typical of hotter drier areas), which are generally not found in the higher
Arizona/New Mexico Plateau (10.1.7) to the south where grasslands are common. Summer
moisture from thunderstorms supports warm season grasses not found in the Central Basin and
Range (10.1.5) to the west. Many endemic plants occur and species diversity is greater than in
Ecoregion 10.1.5. Several national parks are located in this ecoregion and attract many visitors to
view their arches, spires, and canyons.
Region
10.1.6
ff
i i i
5000 10000 15000
ANC (ueq/L)
2000 4000 6000 8000
BCo (ueq/L)
5 10
DOC (rrg/L)
\
15
n n
n
2 ° -
0 10000 30000 0.0 0.1 0.2 0.3 0.4 0.5
SO4 (ueq/L) Q (rrfyr)
Figure C-90. Region 10.1.6 Water Quality Data Summary
HILn
mm n n n
0 20 40 60 80 100
NO3 (ueq/L)
C-90
-------
Region 10.1.7 Arizona/New Mexico Plateau
The Arizona/New Mexico Plateau represents a large transitional region between the semiarid
grasslands and low relief tablelands of the Southwestern Tablelands (9.4.3) in the east, the drier
shrublands and woodland covered higher relief tablelands of the Colorado Plateau (10.1.6) in the
north, and the lower, hotter, less vegetated Mojave Basin and Range (10.2.1) in the west and
Chihuahuan Deserts (10.2.10) in the southeast. Higher, forest-covered, mountainous ecoregions
border the region on the northeast (6.2.14) and south (13.1.1). Local relief in the region varies
from a few feet on plains and mesa tops to well over 1000 feet along tableland side slopes.
Region
10.1.7
CO
o
C 0
0
D
a-
q
o
I I I I
1000 2000 3000 4000
q
CN
in
o
q
o
q
-------
Region 10.1.8 Snake River Plain
This portion of the xeric intermontane western United States is considerably lower and more
gently sloping than the surrounding ecoregions. Mostly because of the available water for
irrigation, a large percent of the alluvial valleys bordering the Snake River are in agriculture,
with sugar beets, potatoes, alfalfa, and vegetables being the principal crops. Cattle feedlots and
dairy operations are also common in the river plain. Except for the scattered barren lava fields,
most of the plains and low hills in the ecoregion have sagebrush-grassland vegetation, now used
mostly for cattle grazing.
Region
10.1.8
q
CN
in
o
q
o
I I I I
1000 2000 3000 4000
cq
o
(Q
O
q
o
q
CN
I I I I
1000 2000 3000 4000
in
o
q
o
ANC(ueq/L)
BCo (ueq/L)
\ I F
468
DOC (mg/L)
I
10
cq
o
CD
o
CN
O
q
o
j?
I
in
o
q
o
0.6
0 200 400 600 800 0.0 0.2 0.4
SO4 (ueq/L) Q (rrfyr)
Figure C-92. Region 10.1.8 Water Quality Data Summary
0 20 40 60 80
N03 (ueq/L)
C-92
-------
Region 10.2 Warm Deserts
Reg_II_10.2
USJ.3NAME
^] Chihuahuan Deserts
^] Mojave Basin and Range
| Sonoran Basin and Range
• Reg_IIJ0.2_ANC
• RegJM0.2_CL
10.2.2
Figure C-93. Region 10.2
C-93
-------
Region 10.2.1 Mojave Basin and Range
This ecoregion contains broad basins and scattered mountains that are generally lower, warmer,
and drier, than those of the Central Basin and Range (10.1.5). Its creosote bush-dominated shrub
community is distinct from the saltbush-greasewood and sagebrush-grass associations that occur
to the north in the Central Basin and Range (10.1.5) and Northern Basin and Range (10.1.3); it is
also differs from the palo verde-cactus shrub and saguaro cactus that occur in the Sonoran Basin
and Range (10.2.2) to the south. Most of this region is federally owned and grazing is
constrained by the lack of water and forage for livestock. Heavy use of off-road vehicles and
motorcycles in some areas has made the soils susceptible to wind and water erosion.
Region
10.2.1
cq
o
CD
O
q
o
1
1 1 1
2000 3000 4000
ANC (ueq/L)
CO
o
CD
O
q
o
1 1 1 1 1
3000 5000 7000
BCo (ueq/L)
CO
o
CD
O
q
o
1.0 1.5 2.0 2.5 3.0
DOC (mg/L)
CO
o
CD
O
O
CN
O
O
<->
1 1 1 1 1
0 5000 10000 20000
SO4 (ueq/L)
q
CO
in
CN
in
o
q
o
0.000
CO
o
s
CN
O
q
o
0.010 0.020
Q (rrfyr)
0.030
10 20 30 40 50
N03 (ueq/L)
Figure C-94. Region 10.2.1 Water Quality Data Summary
C-94
-------
Region 10.2.2 Sonoran Basin and Range
Similar in topography to the Mojave Basin and Range (10.2.1) to the north, this ecoregion
contains scattered low mountains and has large tracts of federally owned land, a large portion of
which is used for military training. However, the Sonoran Basin and Range is slightly hotter than
the Mojave and contains large areas of palo verde-cactus shrub and giant saguaro cactus,
whereas the potential natural vegetation in the Mojave is largely creosote bush. Winter rainfall
decreases from west to east, while summer rainfall decreases from east to west.
Region
10.2.2
q
-------
Region 10.2.10 (10.2.4) Chihuahuan Deserts
This desert ecoregion extends from the Madrean Archipelago (12.1.1) in southeastern Arizona to
the Edwards Plateau (9.4.6) in south-central Texas. The physiography is generally a continuation
of basin and range terrain that is typical of the Mojave Basin and Range (10.2.1) and the Central
Basin and Range (10.1.5) to the west and northwest, although the patterns of alternating
mountains and valleys is not as pronounced as in Ecoregions 10.1.5 and 10.2.1. Vegetative cover
is predominantly desert grassland and shrubland, except on the higher mountains where oak,
juniper, and pinyon woodlands occur. The extent of desert shrubland is increasing across
lowlands and mountain foothills due to the gradual desertification caused in part by historical
grazing pressure.
Region
10.2.4
CO
o
(Q
O
q
o
I I I I
2000 2500 3000 3500
cq
o
(Q
O
q
o
ANC(ueq/L)
I I I I I
4000 4500 5000 5500 6000
BCo (ueq/L)
CO
o
(Q
O
q
o
1.5 2.0 2.5 3.0 3.5
DOC (mg/L)
in
o
q
o
5000 15000 25000 35000
cq
o
(Q
o
-------
Region 11.1 Mediterranean California
11.1 X3TS^
USJ.3NAME
^| Central California Valley ^
^| Southern California Mountains
^\ Southern and Central California Chaparral and Oak Woodlands
* Reg_ll_11.1_ANC
* Reg_ll_11.1_CL
Figure C-97. Region 11.1
C-97
-------
Region 11.1.1 Southern and Central California Chaparral and Oak Woodlands
The primary distinguishing characteristic of this ecoregion is its Mediterranean climate of hot
dry summers and cool moist winters, and associated vegetative cover comprising mainly
chaparral and oak woodlands; grasslands occur in some lower elevations and patches of pine are
found at higher elevations. Most of the region consists of open low mountains or foothills, but
there are areas of irregular plains in the south and near the border of the adjacent Central
California Valley ecoregion. Large parts of the region are grazed by domestic livestock;
relatively little land has been cultivated, although some valleys are or were important agricultural
centers.
Region
11.1.1
q
CO
in
-------
Region 11.1.2 Central California Valley
This region is characterized by flat, intensively farmed plains having long, hot dry summers and
mild winters, distinguish the Central California Valley from its neighboring ecoregions that are
either hilly or mountainous, forest or shrub covered, and generally nonagricultural. Nearly half of
the region is in cropland, about three fourths of which is irrigated. Environmental concerns in the
region include salinity due to evaporation of irrigation water, groundwater contamination from
heavy use of agricultural chemicals, wildlife habitat loss, and urban sprawl.
Region
11.1.2
cq
o
c o
0
D
a-
q
o
cq
o
(Q
o
I I I I I I I
1000 3000 5000 7000
q
o
cq
o
(Q
o
I I I I I I I
1000 3000 5000 7000
q
o
ANC(ueq/L)
BCo (ueq/L)
I I I I I
12345
DOC (mg/L)
in
o
q
o
in
o
q
o
CD -
in -
^- -
s-
c
S CO -
D"
0
LL
CN -
o -
1
100 200 300 400 500 0.0 0.2 0.4 0.6
SO4 (ueq/L) Q (rrfyr)
Figure C-99. Region 11.1.2 Water Quality Data Summary
I I I
0 50 150 250
N03 (ueq/L)
C-99
-------
Region 11.1.3 Southern California Mountains
Like the other ecoregions in central and southern California, the Southern California Mountains
has a Mediterranean climate of hot dry summers and moist cool winters. Although
Mediterranean types of vegetation such as chaparral and oak woodlands predominate in this
region, the elevations are considerably higher, the summers are slightly cooler, and precipitation
amounts are greater than in adjacent ecoregions, resulting in more dense vegetation and some
large areas of coniferous woodlands. Severe erosion problems are common where the vegetation
cover has been destroyed by fire or overgrazing.
Region
11.1.3
1000 3000 5000 7000
ANC (ueq/L)
I I I I
2000 4000 6000 8000
BCo (ueq/L)
5 10
DOC (mg/L)
I
15
o
I
a- o
m n
5000 10000 15000
j?
I CO
0.00
0.10
0.20
SO4 (ueq/L) Q (rrfyr)
Figure C-100. Region 11.1.3 Water Quality Data Summary
j?
I
8-i
8-
in _
o _
50 100 150 200 250
N03 (ueq/L)
C-100
-------
Region 12.1 Western Sierra Madre Piedmont
USJ.3NAME
f Madrean Archipelago
• RegJI_12.1_ANC
« Reg II 12.1 CL
Figure C-101. Region 12.1
C-101
-------
Region 12.1.1 Madrean Archipelago
Also known as the Sky Islands in the United States, this is a region of basins and ranges with
medium to high local relief, typically 3,000 to 5,000 feet. Native vegetation in the region is
mostly grama-tobosa shrubsteppe in the basins and oak-juniper woodlands on the ranges, except
at higher elevations where ponderosa pine is predominant. The region has ecological significance
as both a barrier and bridge between two major Cordilleras of North America, the Rocky
Mountains and the Sierra Madre Occidental.
Region
12.1.1
q
CN
in
o
q
o
1000 3000 5000
ANC (ueq/L)
q
CN
in
o
q
o
I I I I I
1000 3000 5000
BCo (ueq/L)
\ \ \
468
DOC (mg/L)
10
q
CO
in
CN
in
o
q
o
n
CD -
in -
1
D"
0 CO-
LL
CN -
O —
I
n
III 1 1
500 1000 1500 2000
j?
I
0.00
0.04
0.08
0.12
SO4 (ueq/L) Q (rrfyr)
Figure C-102. Region 12.1.1 Water Quality Data Summary
20 40 60
N03 (ueq/L)
80
C-102
-------
Region 13.1 Upper Gila Mountains
-'•
J3.1.1
'* •
USJ.3NAME
^ Arizona/New Mexico Mountains
* Reg_l!_13.1_ANC
• RegJIJ3.1_CL
Figure C-103. Region 13.1
C-103
-------
Region 13.1.1 Arizona/New Mexico Mountains
The Arizona/New Mexico Mountains are distinguished from neighboring mountainous
ecoregions by their lower elevations and associated vegetation indicative of drier, warmer
environments, which is due in part to the region's more southerly location. Forests of spruce, fir,
and Douglas-fir, that are common in the Southern Rockies (6.2.14) and the Uinta and Wasatch
Mountains (6.2.13), are only found in a few high elevation parts of this region. Chaparral is
common on the lower elevations, pinyon-juniper and oak woodlands are found on lower and
middle elevations, and the higher elevations are mostly covered with open to dense ponderosa
pine forests. These mountains are the northern extent of some Mexican plant and animal species.
Region
13.1.1
I I I I I I
0 1000 3000 5000
ANC(ueq/L)
I I I I I I I
0 2000 4000 6000
BCo (ueq/L)
o _
nm
10
DOC (mg/L)
I
20
>. o _
o ^ ^
i
n n m
500 1000 1500
j?
I
D"
0
0.00 0.05 0.10 0.15 0.20
SO4 (ueq/L) Q (rrfyr)
Figure C-104. Region 13.1.1 Water Quality Data Summary
8-
o !£
I
D"
0 O
n-
n mn n
5 10
N03 (ueq/L)
15
C-104
-------
Region 15.4 Everglades
USJ.3NAME
^J Southern Florida Coastal Plain
* Reg_ll_15.4_CL
Figure C-105. Region 15.4
C-105
-------
Region 15.4.1 Southern Florida Coastal Plain
The frost free climate of the Southern Florida Coastal Plain makes it distinct from other
ecoregions in the conterminous United States. This region is characterized by flat plains with wet
soils, marsh and swamp land cover with everglades and palmetto prairie vegetation types.
Relatively slight differences in elevation and landform have important consequences for
vegetation and the diversity of habitat types. Although portions of this region are in parks, game
refuges, and Indian reservations, a large part of the region has undergone extensive hydrological
and biological alteration.
Region
15.4.1
CO
o
c o
0
D
a-
q
o
cq
o
CO
o
I I I I I
2600 2650 2700 2750 2800
ANC(ueq/L)
q
o
cq
o
(Q
o
q
o
2800 2850 2900 2950 3000
BCo (ueq/L)
10
I
12
I
14
I
16
I
18
20
DOC (mg/L)
CO
o
CO
o
-------
References
US EPA. 2010. Primary Distinguishing Characteristics Of Level III Ecoregions Of The
Continental United States, July, 2010. U.S. Environmental Protection Agency,
Washington, DC. Available at
http://www.epa. gov/wed/pages/ecoregions/level_iii_iv. htm.
C-107
-------
C-108
-------
Appendix D
Maps and calculation procedures for alternative standards
This appendix includes supplemental information that is referred to in section 7.5,
Considerations associated with alternative standards. Section D. 1 provides the calculation
procedures used in calculating the AAI values; D.2 provides maps of ecoregions not likely
meeting alternative standards and summary tables of all related calculations; D.3 provides the
critical loads for all acid sensitive ecoregions, calculated AAI values and critical loads for
relatively non-acid sensitive regions, and CMAQ deposition values and transference ratios for
each region; and D.4 provides further explanation of calculations in the context of combinations
of NOy and SOx, concentrations and associated deposition, as well as illustrating the effect of
Neco on calculations.
D.I Analytical approach and data sources
Results of this assessment are based on calculating AAIs for alternative standards using
the range of levels (20, 35, 50, 75) and nth percentiles (70, 75, 80, 85, 90) discussed in section
7.4. Because we have modeled data only as a source of deposition and concentration fields,
CMAQ output for concentration fields, transference ratios, and deposition values enables
applications for calculating AAI values using equation 7-11.
ANCcalc = {ANClim + CLr/Qr} - NHx/Qr - TNOy [NOy]/Qr - TSOx[SOx]/Qr (7-11)
Calculations procedures:
1. Assemble current deposition and concentration fields using the annual average value of
each grid cell and then averaged over the grid cells of the ecoregion of interest.
Ndep = NHxdep + NOydep
All depositions are in values of meq/m2-yr
2. Calculate CLr(ANciim,%);
CLr(ANC,%) = ([BC*o]andim,% - [ANCi,m])QANCiim,% + Neco
Where the ANClim,% refers to the specified target ANC level and the specific water
body representing the nth percentile critical load in the ecoregion.
D-l
-------
([BC]o,% is calculated with water quality data (major cations, NO3 and SO4) as
described in Appendix B)
Neco = X (Ndep - Nleachj/n
3. Determine if Neco > Ndep
4. Calculate deposition exceedance, and check for Neco conditions:
DEPex = Ndep + Sdep - (dBC]0>%* - [ANCiim])Qo/0 + Neco%),, Ndep > Neco
DEPex = Sdep - ([BC]0,%* - [ANCiim])Q%, Ndep < Neco
5. Calculate an AAI value, which essentially is a design value at the ecoregion level:
Calculated AAI = Level (as ANClim) - DEPex/Qr
As described in section 7.2, this calculation procedure using deposition exceedances is
the basis for deriving equation 7-11, and the AAI calculations using this approach is identical,
algebraically, to using equation 7-11 directly.
Data sources used in the exceedance analyses.
The AAI calculations require estimates of current time frame N and S deposition,
observed water quality major cations (CA, Mg, K) and strong anions (NOs, SO/t) and runoff
rates, summarized in Table D-l. The 2005 and emissions sensitivity CMAQ simulations were
applied at 12 km horizontal grid cell resolution and provided all deposition and concentration
estimates. Water quality data were based on TIME/LTM and other data bases as described in
chapter 2. The representative critical loads for each ecoregion are provided in Table D.2.
D-2
-------
Table D-l. Data sources for calculating AAIs.
Parameter
Ndep
NHx
Sdep
Nleach
BC o,%
Q
Qr
Description
Sum of wet and dry NOy and NHx
deposition
Sum of wet and dry NHx
deposition
Sum of wet and dry SO2 and SO4
deposition
Outflow water column
concentration of NO3
Preindustrial base cation levels
Annual runoff rate
Ecoregion representative
Annual runoff rate
Data source
2005 12 CMAQ
base case
2005 12 CMAQ
base case
2005 12 CMAQ
base case
TIME/LTM and
STORET data bases
(Table 2-4)
TIME/LTM and
STORET data bases
(Table 2-4)
USGS (Table 2-4)
USGS (Table 2-4)
Notes
meq/(m2-yr), average of all 12
km grid cells within ecoregion
meq/(m2-yr), average of all 12
km grid cells within ecoregion
meq/(m2-yr), average of all 12
km grid cells within ecoregion
ueq/1; average of all water bodies
in an ecoregion
ueq/1; based on the % water body
m/yr; based on the % water body
m/yr; based on the 50 % water
body
D-3
-------
D.2 Maps and tables of ecoregions not meeting alternative standards.
Figures D-l through D-9 are illustrative maps of ecoregions likely not meeting alternative
standards based on current (2005) conditions and an emissions sensitivity simulation of
approximately 42% and 48% SOx and NOx emission reductions across the U.S. Table D-2a -D-
2d include all of the calculated AAI values for the range alternative standards considered.
ueq/L
196
- 176
- 157
137
- 117
98
78
59
39
20
0
ueq/L
y
103
93
83
72
62
52
41
31
21
10
Figure D-l. Acid sensitive ecoregion level III areas likely not meeting the standard when the
level = 35 jieq/L and the form is based on the 70th percentile water body; based on CMAQ 2005
D-4
-------
(top) and emissions sensitivity(bottom) simulations. The legend reflects the magnitude of the
exceedance relative to the specified level of the standard.
ueq/L
209
ueq/L
117
- 105
- 94
- 82
- 70
- 59
- 47
35
23
12
0
Figure D-2. Acid sensitive ecoregion level III areas likely not meeting the standard when the
level = 35 jieq/L and the form is based on the 80th percentile water body; based on CMAQ 2005
(top) and emissions sensitivity(bottom) simulations. The legend reflects the magnitude of the
exceedance relative to the specified level of the standard.
D-5
-------
ueq/L
228
- 205
- 182
- 160
- 137
- 114
- 91
68
46
23
ueq/L
136
- 122
- 109
- 95
- 81
- 68
- 54
- 41
27
14
0
Figure D-3. Acid sensitive ecoregion level III areas likely not meeting the standard when the
level = 35 jieq/L and the form is based on the 90th percentile water body; based on CMAQ 2005
(top) and emissions sensitivity(bottom) simulations. The legend reflects the magnitude of the
exceedance relative to the specified level of the standard.
D-6
-------
ueq/L
217
- 195
- 173
- 152
- 130
- 108
87
65
43
22
ueq/L
125
- 112
- 100
- 87
- 75
62
50
37
25
12
0
Figure D-4. Acid sensitive ecoregion level III areas likely not meeting the standard when the
level = 50 jieq/L and the form is based on the 70th percentile water body; based on CMAQ 2005
(top) and emissions sensitivity(bottom) simulations. The legend reflects the magnitude of the
exceedance relative to the specified level of the standard.
D-7
-------
ueq/L
225
- 202
• 180
- 157
135
- 112
90
67
45
22
ueq/L
133
- 119
106
93
80
66
y
53
40
27
13
Figure D-5. Acid sensitive ecoregion level III areas likely not meeting the standard when the
level = 50 jieq/L and the form is based on the 80th percentile water body; based on CMAQ 2005
(top) and emissions sensitivity(bottom) simulations. The legend reflects the magnitude of the
exceedance relative to the specified level of the standard.
D-8
-------
ueq/L
243
- 218
194
- 170
- 146
121
97
73
49
24
ueq/L
151
- 135
- 120
- 105
- 90
- 75
- 60
- 45
30
15
0
Figure D-6. Acid sensitive ecoregion level III areas likely not meeting the standard when the
level = 50 jieq/L and the form is based on the 90th percentile water body; based on CMAQ 2005
(top) and emissions sensitivity(bottom) simulations. The legend reflects the magnitude of the
exceedance relative to the specified level of the standard.
D-9
-------
ueq/L
242
- 218
- 193
- 169
- 145
- 121
- 97
- 73
48
24
0
ueq/L
150
- 135
- 120
- 105
- 90
- 75
60
45
30
15
Figure D-7. Acid sensitive ecoregion level III areas likely not meeting the standard when the
level = 75 jieq/L and the form is based on the 70th percentile water body; based on CMAQ 2005
(top) and emissions sensitivity(bottom) simulations. The legend reflects the magnitude of the
exceedance relative to the specified level of the standard.
D-10
-------
ueq/L
r 251
- 226
- 201
- 176
- 151
- 126
101
75
50
25
ueq/L
159
- 143
- 127
- 111
- 96
- 80
64
48
32
16
Figure D-8. Acid sensitive ecoregion level III areas likely not meeting the standard when the
level = 75 jieq/L and the form is based on the 80th percentile water body; based on CMAQ 2005
(top) and emissions sensitivity(bottom) simulations. The legend reflects the magnitude of the
exceedance relative to the specified level of the standard.
D-ll
-------
ueq/L
276
- 249
- 221
- 193
166
138
- 111
83
55
28
0
ueq/L
184
166
- 147
129
- 110
92
y
74
55
37
18
Figure D-9. Acid sensitive ecoregion level III areas likely not meeting the standard when the
level = 75 jieq/L and the form is based on the 90th percentile water body; based on CMAQ 2005
(top) and emissions sensitivity(bottom) simulations. The legend reflects the magnitude of the
exceedance relative to the specified level of the standard.
D-12
-------
Table D-2a. Calculated AAI values for sensitive ecoregions across the range of nth percentiles for a
Level of 20 (ieq/L: pairs of 2005 base and emissions sensitivity (42% and 48%NOx and SOx reduction).
Highlighted cell pairs: Yellow-Green (likely not meeting 2005; likely meeting with emissions
reductions); Yellow-Red (likely not meeting 2005 and after emissions reductions)
6.2.4
6.2.3
6.2.7
8.5.4
5.3.1
6.2.10
8.1.3
8.1.7
5.3.3
8.1.8
6.2.5
5.2.1
6.2.15
8.4.1
8.4.2
8.4.3
6.2.13
8.5.1
6.2.12
6.2.14
8.4.4
8.3.5
8.3.4
8.4.9
8.4.6
8.4.7
8.5.3
8.4.8
8.3.7
Canadian Rockies
Columbia
Mountains/Northern
Rockies
Cascades
Atlantic Coastal Pine
Barrens
Northern Appalachian
and Atlantic Maritime
Highlands
Middle Rockies
Northern Appalachian
Plateau and Uplands
Northeastern Coastal
Zone
North Central
Appalachians
Maine/New Brunswick
Plains and Hills
North Cascades
Northern Lakes and
Forests
Idaho Batholith
Ridge and Valley
Central Appalachians
Western Allegheny
Plateau
Wasatch and Uinta
Mountains
Middle Atlantic Coastal
Plain
Sierra Nevada
Southern Rockies
Blue Ridge
Southeastern Plains
Piedmont
Southwestern
Appalachians
Boston Mountains
Arkansas Valley
Southern Coastal Plain
Ouachita Mountains
South Central Plains
70th
933.4
353.5
90.2
-154.6
58.3
180.0
227.0
42.0
-60.8
89.7
138.4
51.4
66.8
-72.3
-78.3
412.4
297.8
-17.2
49.1
120.6
-65.5
-51.2
131.6
35.3
65.1
90.1
-31.2
89.3
287.1
935.8
359.5
92.7
-62.5
102.9
187.9
342.3
117.5
79.5
110.1
139.8
97.9
70.3
76.2
109.2
694.3
323.8
75.4
59.2
129.4
68.8
21
266.5
131.9
119.3
129.6
102.6
140.5
346.3
75th
740.3
267.3
72.2
-172.7
49.0
122.1
173.4
22.6
-74.4
84.4
130.7
38.7
62.0
-95.1
-109.0
280.7
255.3
-29.5
38.2
98.5
-73.5
-59.2
102.7
-18.5
65.1
82.5
-61.8
74.6
279.6
742.7
273.2
74.7
-80.6
93.6
130
288.7
98.1
66
104.7
132.1
85.2
65.4
53.4
78.4
562.6
281.4
63.1
48.2
107.4
60.8
13
237.5
78.1
119.3
122
71.9
125.9
338.8
80th
685.6
190.3
46.2
-174.6
33.6
99.0
165.7
9.3
-87.4
71.0
112.9
25.9
59.3
-117.6
-147.0
47.4
230.6
-64.0
28.1
85.7
-83.3
-73.1
72.7
-29.4
27.7
66.2
-105.1
64.6
213.3
688
196.3
48.6
-82.5
78.2
106.9
281
84.8
52.9
91.4
114.2
72.4
62.8
30.9
40.5
329.3
256.7
28.5
38.2
94.6
51
-0.9
207.6
67.2
81.9
105.7
28.6
115.8
272.6
85th
551.0
136.5
31.3
-182.4
21.3
81.6
120.4
-4.4
-97.8
65.5
93.8
14.1
48.0
-143.7
-169.5
-20.8
174.6
-131.6
22.2
67.6
-93.1
-91.0
45.7
-69.8
8.7
50.7
-143.2
51.3
136.4
553.4
142.5
33.8
-90.2
65.9
89.5
235.7
71.1
42.6
85.9
95.2
60.6
51.5
"•"
17.9
261.1
200.6
-39.1
32.2
76.4
41.3
-18.9
180.6
26.7
62.8
90.2
-9.4
102.6
195.7
90th
84.5
106.3
19.1
-193.6
6.4
69.4
85.6
-23.6
-112.5
48.5
65.8
3.8
41.6
-177.8
-186.2
-97.9
136.6
-169.4
12.6
50.8
-104.9
-106.7
11.8
-121.5
-24.4
-1.0
-154.9
-3.3
47.3
86.9
112.2
21.6
-101.5
51
77.3
200.9
51.9
27.9
68.9
67.2
50.3
45.1
"•"
•
184
162.6
-76.9
22.6
59.7
29.4
-34.5
146.7
-25
29.7
38.5
-21.2
47.9
106.6
D-13
-------
Table D-2b. Calculated AAI values for sensitive ecoregions across the range of nth percentiles
for a Level of 35 (ieq/L: pairs of 2005 base and emissions sensitivity (42% and 48% NOx and SOx
reduction). Highlighted cell pairs: Yellow-Green (likely not meeting 2005; likely meeting with
emissions reductions); Yellow-Red (likely not meeting 2005 and after emissions reductions)
6.2.4
6.2.3
6.2.7
8.5.4
5.3.1
6.2.10
8.1.3
8.1.7
5.3.3
8.1.8
6.2.5
5.2.1
6.2.15
8.4.1
8.4.2
8.4.3
6.2.13
8.5.1
6.2.12
6.2.14
8.4.4
8.3.5
8.3.4
8.4.9
8.4.6
8.4.7
8.5.3
8.4.8
8.3.7
Canadian Rockies
Columbia
Mountains/Northern
Rockies
Cascades
Atlantic Coastal Pine
Barrens
Northern Appalachian
and Atlantic Maritime
Highlands
Middle Rockies
Northern Appalachian
Plateau and Uplands
Northeastern Coastal
Zone
North Central
Appalachians
Maine/New Brunswick
Plains and Hills
North Cascades
Northern Lakes and
Forests
Idaho Batholith
Ridge and Valley
Central Appalachians
Western Allegheny
Plateau
Wasatch and Uinta
Mountains
Middle Atlantic Coastal
Plain
Sierra Nevada
Southern Rockies
Blue Ridge
Southeastern Plains
Piedmont
Southwestern
Appalachians
Boston Mountains
Arkansas Valley
Southern Coastal Plain
Ouachita Mountains
South Central Plains
70%
934.4
342.8
92.1
-160.7
58.8
179.1
225.5
42.3
-59.8
89.5
144.4
52.7
60.2
-72.3
-80.3
415.2
287.3
-16.2
47.4
120.2
-65.4
-55.5
131.2
31.1
65.1
89.3
-29.1
89.5
291.9
936.8
348.8
94.6
-68.5
103.5
187
340.7
117.8
80.6
109.9
145.8
99.2
63.6
76.2
107.1
697.1
313.4
76.4
57.4
129
69
•
266
127.7
119.3
128.8
104.6
140.8
351.1
75%
736.5
257.3
74.7
-173.7
46.2
121.7
173.7
23.2
-71.8
84.6
125.1
39.6
56.7
-94.0
-107.3
281.4
243.0
-32.0
40.5
102.3
-71.6
-63.4
96.1
-12.2
65.1
85.7
-61.7
78.9
275.4
738.9
263.2
77.1
-81.5
90.8
129.6
289
98.8
68.6
105
126.4
86.1
60.1
54.5
80.1
563.3
269
60.5
50.5
111.2
62.7
"•"
231
84.3
119.3
125.2
•
130.1
334.6
80%
692.4
188.0
47.8
-174.3
32.1
99.2
163.9
8.9
-88.7
72.1
110.3
25.7
53.7
-116.3
-143.2
46.2
221.8
-61.4
30.6
87.3
-82.4
-72.1
72.6
-29.1
28.9
71.4
-106.5
67.4
210.4
694.8
194
50.3
-82.1
76.7
107.1
279.2
84.4
51.7
92.5
111.7
72.2
57.2
"•"
44.2
328.1
247.9
"•"
40.7
96.2
52
"•"
207.4
67.5
83
110.9
"•"
118.7
269.7
85%
556.9
122.5
30.7
-180.7
20.1
81.0
121.4
-6.1
-95.3
65.9
86.5
14.3
49.6
-144.8
-173.0
-22.1
184.4
-133.2
21.8
66.4
-93.6
-96.2
43.7
-71.4
8.7
50.6
-138.6
49.3
133.1
559.2
128.5
33.2
-88.5
64.7
88.9
236.7
69.4
45.1
86.3
87.8
60.8
53.1
"•"
•
259.8
210.5
"•"
31.8
75.3
40.7
~B~
178.6
•
62.9
90.1
"•"
100.5
192.4
90%
80.9
106.8
21.4
-192.9
3.4
63.3
87.3
-24.2
-114.3
47.1
71.6
1.0
39.8
-175.4
-182.0
-94.1
126.3
-163.4
13.8
53.1
-104.6
-107.9
14.0
-121.3
-21.2
-2.0
-150.8
-4.8
47.3
83.2
112.8
23.9
-100.7
48
71.2
202.6
51.3
26.1
67.5
73
47.5
43.2
"•"
"•"
187.8
152.4
-70.9
"•"
62
"•"
"•"
148.9
"•"
•
37.5
~B~
46.5
106.6
D-14
-------
Table D-2c. Calculated AAI values for sensitive ecoregions across the range of nth percentiles
for a Level of 50 (ieq/L: pairs of 2005 base and emissions sensitivity (42% and 48% NOx and SOx
reduction).
Highlighted cell pairs: Yellow-Green (likely not meeting 2005; likely meeting with emissions
reductions); Yellow-Red (likely not meeting 2005 and after emissions reductions)
6.2.4
6.2.3
6.2.7
8.5.4
5.3.1
6.2.10
8.1.3
8.1.7
5.3.3
8.1.8
6.2.5
5.2.1
6.2.15
8.4.1
8.4.2
8.4.3
6.2.13
8.5.1
6.2.12
6.2.14
8.4.4
8.3.5
8.3.4
8.4.9
8.4.6
8.4.7
8.5.3
8.4.8
8.3.7
Canadian Rockies
Columbia
Mountains/Northern
Rockies
Cascades
Atlantic Coastal Pine
Barrens
Northern Appalachian and
Atlantic Maritime Highlands
Middle Rockies
Northern Appalachian
Plateau and Uplands
Northeastern Coastal Zone
North Central Appalachians
Maine/New Brunswick
Plains and Hills
North Cascades
Northern Lakes and Forests
Idaho Batholith
Ridge and Valley
Central Appalachians
Western Allegheny Plateau
Wasatch and Uinta
Mountains
Middle Atlantic Coastal
Plain
Sierra Nevada
Southern Rockies
Blue Ridge
Southeastern Plains
Piedmont
Southwestern Appalachians
Boston Mountains
Arkansas Valley
Southern Coastal Plain
Ouachita Mountains
South Central Plains
70%
935.5
327.0
93.0
-166.7
57.0
178.2
223.9
42.1
-58.7
90.1
148.3
54.0
59.3
-75.2
-82.0
418.0
276.9
-15.2
44.9
124.1
-65.3
-57.5
125.8
26.9
65.1
88.8
-27.5
89.8
296.7
937.8
333
95.4
"•"
101.6
186.1
339.2
117.6
81.7
110.5
149.7
100.5
62.8
73.3
105.5
699.9
303
"•"
55
133
69
"•"
260.6
123.4
119.2
128.3
106.2
141
355.9
75%
732.7
262.1
72.5
-174.0
44.0
119.5
174.1
23.9
-69.2
84.8
123.8
39.8
53.6
-94.4
-105.5
282.2
230.6
-34.6
39.0
112.6
-72.9
-65.2
99.6
-6.0
65.1
88.4
-55.8
83.1
271.2
735
268.1
74.9
"•"
88.7
127.4
289.4
—^~
71.1
105.2
125.2
86.3
57.1
54.1
•
564.1
256.7
58
•
121.5
1
234.5
90.5
119.2
127.9
77.9
134.4
330.5
80%
699.2
192.3
51.7
-174.7
30.8
97.5
162.2
8.7
-87.5
73.4
116.8
24.7
43.9
-116.8
-140.5
43.2
199.4
-58.8
32.0
85.2
-82.7
-77.2
74.2
-28.8
30.0
76.6
-101.9
70.3
209.4
701.6
198.2
54.2
"•"
75.4
105.4
277.5
84.2
52.9
93.8
118.2
"•"
"•"
"•"
325.1
225.5
"•"
"•"
94.1
"•"
209.1
67.7
84.2
116.1
"•"
121.5
268.7
85%
562.7
132.6
29.2
-176.5
17.0
81.6
122.4
-7.2
-95.5
65.1
93.0
14.1
40.6
-144.0
-172.1
-23.4
194.2
-134.8
24.3
67.9
-92.5
-102.6
40.1
-73.1
8.7
50.5
-134.7
47.2
129.8
565.1
138.5
"•"
"•"
89.5
237.7
68.3
"•"
85.5
94.4
60.6
— r
"•"
"•"
258.5
220.3
"•"
"•"
76.7
"•"
"•"
175
"•"
62.9
90
r
98.5
189.1
90%
79.4
106.4
23.9
-192.7
1.1
56.9
89.0
-24.8
-115.8
45.8
69.7
-0.2
37.5
-172.8
-182.1
-90.3
109.5
-157.4
13.7
58.1
-102.8
-112.7
14.8
-121.1
-17.9
0.7
-151.7
-6.2
47.3
81.8
112.4
"•"
•
64.8
204.3
50.7
~*~
66.2
71.1
~H~
40.9
"•"
"•"
191.6
135.6
"•"
66.9
"•"
"•"
149.7
"•"
"•"
"•"
"•"
•
106.6
Table D-2d. Calculated AAI values for sensitive ecoregions across the range of nth percentiles for a
Level of 75 (ieq/L: pairs of 2005 base and emissions sensitivity (42% and 48% NOx and SOx reduction).
Highlighted cell pairs: Yellow-Green (likely not meeting 2005; likely meeting with emissions
D-15
-------
reductions); Yellow-Red (likely not meeting 2005 and after emissions reductions)
70%
75%
85%
90%
6.2.4
Canadian Rockies
937.2
939.5
726.3
728.7
710.5
712.9
572.5
574.9
63.5
77.1
6.2.3
Columbia
Mountains/Northern
Rockies
316.6
322.6
268.7
274.7
207.7
130.5
136.5
40.3
103.6
6.2.7
5.3.1
6.2.10
8.1.3
Cascades
89.4
91.9
Atlantic Coastal Pine
Barrens
-166.7
Northern Appalachian and
Atlantic Maritime
Highlands
56.3
100.9
43.4
27.6
Middle Rockies
176.6
184.5
125.2
133.1
98.0
105.9
Northern Appalachian
Plateau and Uplands
225.3
340.6
174.6
289.9
159.3
274.5
128.5
243.7
-35.2
86.6
8.1.7
5.3.3
6.2.5
Northeastern Coastal Zone
43.0
118.5
25.0
100.5
8.1
83.6
-5.2
70.3
North Central Appalachians
-56.9
83.5
-73.1
67.3
-85.3
Maine/New Brunswick
Plains and Hills
87.9
108.2
83.2
103.6
71.7
North Cascades
145.6
147
125.1
126.5
116.0
92.1
117.4
95.4
96.7
5.2.1
6.2.13
Northern Lakes and Forests
54.9
Wasatch and Uinta
Mountains
259.6
285.6
210.6
236.7
186.2
212.3
168.0
194
71.9
102.8
8.5.1
8.3.4
Middle Atlantic Coastal
Plain
-13.5
79
-38.8
53.7
Piedmont
126.1
261
8.4.9
8.4.6
8.4.7
Southwestern
Appalachians
19.9
116.4
4.4
100.9
-28.4
Boston Mountains
65.0
119.2
65.0
119.2
31.9
Arkansas Valley
93.8
133.3
87.0
126.5
85.3
124.8
50.4
89.9
8.5.3
8.4.8
Southern Coastal Plain
-31.0
102.7
-60.7
Ouachita Mountains
90.2
141.5
90.2
141.5
-94.4
75.1
126.3
-132.0
43.8
95
-182.4
-44.9
8.3.7
South Central Plains
304.7
363.9
264.3
323.5
220.8
280.1
124.3
183.6
D-16
-------
D.3 Data summary tables
Tables D-3a and b includes critical load values for all acid sensitive areas; Table D-4
includes 2005 calculated AAI values and critical loads for relatively non-acid sensitive areas;
Table D-5 includes CMAQ deposition values and transference ratios for each ecoregion.
Table D-3a. Representative critical loads for acid sensitive ecoregions - Level values of 35 and
20 |ieq/L.
EcoRegion Level III
Numbe
r
5.2.1
5.3.1
5.3.3
6.2.10
6.2.12
6.2.13
6.2.14
6.2.15
6.2.3
6.2.4
6.2.5
6.2.7
8.1.3
8.1.7
8.1.8
8.3.4
8.3.5
8.3.7
8.4.1
8.4.2
8.4.3
8.4.4
8.4.6
8.4.7
8.4.8
Description
No. Lakes and
Forests
No. Appalachian/
Atlantic Maritime
Highlands
No. Central
Appalachians
Middle Rockies
Sierra Nevada
Wasatch and
Uinta Mountains
Southern Rockies
Idaho Batholith
Columbia Mtns/
No. Rockies
Canadian Rockies
No. Cascades
Cascades
No. Appalachian
Plateau and
Uplands
NE Coastal Zone
Maine/NBrunswic
k
Plains and Hills
Piedmont
Southeastern
Plains
South Central
Plains
Ridge and Valley
Central
Appalachians
Western
Allegheny
Plateau
Blue Ridge
Boston Mountains
Arkansas Valley
Ouachita
Mountains
70%
ANC
35
79.1
120.4
132.4
82.0
55.0
90.5
75.4
41.2
172.9
893.1
227.7
131.0
256.8
168.7
112.8
208.4
87.6
213.4
133.0
161.6
445.6
122.9
137.3
139.0
138.3
75%
ANC
35
74.7
112.3
125.8
60.8
50.5
81.3
66.2
38.9
131.7
703.5
194.4
104.1
230.1
156.7
109.5
193.1
83.7
207.2
122.9
146.9
384.9
119.9
137.3
137.3
133.3
80%
ANC
35
70.1
103.4
116.5
52.5
44.1
76.9
58.4
36.9
98.4
661.2
169.0
62.8
225.1
147.6
101.0
182.8
79.4
182.7
112.6
127.4
278.0
114.8
121.4
130.4
127.9
85%
ANC
35
66.3
95.8
112.9
45.7
38.3
69.0
47.7
34.1
66.9
531.3
128.0
36.6
203.2
138.2
96.7
170.3
67.3
153.6
99.3
111.3
247.0
109.3
112.6
120.4
119.3
90%
ANC
35
61.9
85.1
102.4
39.1
33.1
56.9
40.8
27.5
59.4
75.2
102.4
22.3
185.6
126.8
84.0
157.3
61.5
121.2
85.1
106.4
214.3
104.0
99.5
95.0
93.8
70%
ANC
20
83.6
129.6
140.1
87.9
66.0
95.9
83.3
55.8
185.2
906.5
243.2
151.0
265.3
177.9
123.2
215.1
97.3
217.2
140.0
170.8
451.2
130.1
143.8
146.7
145.3
75%
ANC
20
79.4
123.6
132.6
66.5
58.8
87.0
71.9
52.5
143.8
721.5
230.0
123.5
237.7
165.7
119.5
202.5
93.3
214.4
129.4
154.1
391.3
126.2
143.8
143.0
138.4
80%
ANC
20
75.2
113.9
125.4
57.9
52.3
81.8
65.3
50.7
106.8
669.1
199.3
83.4
233.7
157.3
110.4
189.4
86.4
189.4
119.0
133.5
285.4
121.5
127.5
135.1
133.6
85%
ANC
20
71.3
106.0
119.7
51.5
48.4
70.1
56.0
43.1
80.9
540.1
166.5
60.6
210.4
148.7
106.7
177.7
77.4
160.5
106.9
121.3
254.4
116.8
119.1
127.6
127.4
90%
ANC
20
67.8
96.6
111.7
47.0
42.1
62.2
47.3
38.8
66.3
93.1
118.3
41.8
192.5
136.6
95.1
162.9
69.6
126.9
91.0
112.3
219.3
111.1
104.7
102.7
101.6
D-17
-------
Table D-3a. Representative critical loads for acid sensitive ecoregions - Level values of 35 and
20 |ieq/L.
8.4.9
8.5.1
8.5.3
8.5.4
Southwestern
Appalachians
Middle Atlantic
Coastal Plain
Southern Coastal
Plain
Atlantic Coastal
Pine
Barrens
177.6
145.9
100.9
85.4
149.3
138.6
92.5
78.4
138.4
125.0
81.0
78.1
110.8
91.8
72.7
74.7
78.3
77.8
69.6
68.1
190.1
152.4
104.2
96.6
155.0
146.7
96.4
86.9
148.0
130.7
85.2
85.9
121.6
99.5
75.4
81.8
88.0
82.0
72.4
75.8
D-18
-------
Table D-3b. Representative critical loads for acid sensitive ecoregions - Level values of
75 and 50 |ieq/L.
EcoRegion Level III
Numbe
r
5.2.1
5.3.1
5.3.3
6.2.10
6.2.12
6.2.13
6.2.14
6.2.15
6.2.3
6.2.4
6.2.5
6.2.7
8.1.3
8.1.7
8.1.8
8.3.4
8.3.5
8.3.7
8.4.1
8.4.2
8.4.3
8.4.4
8.4.6
8.4.7
8.4.8
8.4.9
8.5.1
8.5.3
8.5.4
Description
No. Lakes and
Forests
No. Appalachian/
Atlantic Maritime
Highlands
No. Central
Appalachians
Middle Rockies
Sierra Nevada
Wasatch and
Uinta
Mountains
Southern
Rockies
Idaho Batholith
Columbia
Mtns/
No. Rockies
Canadian
Rockies
No. Cascades
Cascades
No.
Appalachian
Plateau and
Uplands
NE Coastal
Zone
Maine/NBrunswick
Plains and Hills
Piedmont
Southeastern
Plains
South Central
Plains
Ridge and
Valley
Central
Appalachians
Western
Allegheny
Plateau
Blue Ridge
Boston
Mountains
Arkansas
Valley
Ouachita
Mountains
Southwestern
Appalachians
Middle Atlantic
Coastal Plain
Southern
Coastal
Plain
Atlantic
Coastal Pine
70%
ANC
75
66.5
93.4
112.0
66.3
27.9
76.4
58.3
11.4
141.0
857.4
161.0
65.3
236.1
143.9
84.5
188.7
65.5
203.1
114.1
139.1
430.9
103.9
119.7
121.9
119.8
144.2
128.7
90.1
60.8
75%
ANC
75
61.9
85.1
103.1
47.3
23.8
66.2
50.8
4.1
118.0
655.4
125.6
35.7
210.0
132.6
81.3
176.7
60.9
187.9
103.7
125.7
367.6
99.2
119.7
118.6
119.8
134.1
116.9
82.5
57.2
80%
ANC
75
57.0
75.1
96.3
37.2
18.0
61.1
40.3
0.0
85.8
640.2
110.0
4.6
202.1
121.9
73.5
164.4
50.8
171.5
93.9
108.6
257.5
95.1
105.2
117.8
112.6
112.8
109.7
73.8
55.6
85%
ANC
75
52.6
66.9
92.3
31.7
13.5
57.2
33.1
-5.9
51.5
508.0
74.4
-16.4
186.2
113.5
67.2
150.2
41.0
135.2
82.0
91.5
227.2
90.4
95.1
101.0
97.9
81.9
71.4
64.1
55.4
90%
ANC
75
47.4
55.7
81.4
20.4
6.1
43.6
26.2
-7.6
38.6
33.3
36.2
-30.9
164.6
100.9
54.1
139.5
32.2
106.2
69.3
85.6
200.7
84.7
85.8
79.1
73.1
52.6
66.3
57.8
42.3
70%
ANC
50
74.5
109.7
124.8
76.1
43.6
85.2
69.7
30.6
158.1
879.7
208.6
109.2
248.2
159.1
103.0
199.5
79.2
209.5
124.7
152.5
440.1
115.7
130.7
131.6
131.4
165.1
139.4
97.4
74.1
75%
ANC
50
69.8
101.4
119.0
54.4
39.8
75.6
63.7
26.7
126.8
685.4
166.4
77.7
222.6
147.6
99.4
188.1
75.3
199.9
115.8
139.8
378.4
112.1
130.7
131.4
128.2
143.6
130.5
90.2
70.2
80%
ANC
50
64.8
93.0
108.9
46.3
35.2
69.0
49.6
20.2
93.2
653.3
154.3
45.8
216.5
138.1
91.7
177.0
69.3
176.7
105.4
120.7
269.8
107.4
115.3
125.7
122.2
128.8
119.3
78.3
69.9
85%
ANC
50
61.3
84.3
104.5
40.4
30.2
68.0
40.6
18.0
64.5
522.6
113.4
11.2
196.0
128.0
86.0
162.2
56.6
146.7
92.8
103.6
239.6
102.7
106.0
113.1
111.3
100.0
84.1
69.9
68.9
90%
ANC 50
56.5
74.1
93.3
31.2
23.2
50.2
35.6
15.9
52.0
59.5
73.3
3.1
178.7
117.0
72.9
151.1
51.6
115.6
79.4
98.2
209.2
97.7
94.4
89.0
86.1
68.7
73.7
65.5
60.3
D-19
-------
Table D-4. Critical Loads (meq/(m2-yr) and calculated AAI values for relatively less non-acid
sensitive ecoregions (note all calculations based on nth %percentile = 50).
Region
10.1.2
10.1.3
10.1.4
10.1.5
10.1.6
10.1.7
10.1.8
10.2.2
11.1.1
11.1.2
11.1.3
12.1.1
13.1.1
6.2.11
6.2.8
6.2.9
7.1.7
7.1.8
8.1.1
8.1.10
8.1.4
8.1.5
8.1.6
8.2.1
8.2.3
8.2.4
8.3.1
8.3.2
8.3.3
8.4.5
8.5.2
9.2.1
9.2.2
9.2.3
9.2.4
9.3.1
9.3.3
9.3.4
Name
Columbia Plateau
Northern Basin and Range
Wyoming Basin
Central Basin and Range
Colorado Plateaus
Arizona/New Mexico Plateau
Snake River Plain
Sonoran Desert
California Coastal Sage, Chaparral, and Oak
Woodlands
Central California Valley
Southern and Baja California Pine-Oak
Mountains
Madrean Archipelago
Arizona/New Mexico Mountains
Klamath Mountains
Eastern Cascades Slopes and Foothills
Blue Mountains
Strait of Georgia/Puget Lowland
Coast Range
Eastern Great Lakes and Hudson Lowlands
Erie Drift Plain
North Central Hardwood Forests
Driftless Area
S. Michigan/N. Indiana Drift Plains
Southeastern Wisconsin Till Plains
Central Corn Belt Plains
Eastern Corn Belt Plains
Northern Piedmont
Interior River Valleys and Hills
Interior Plateau
Ozark Highlands
Mississippi Alluvial Plain
Aspen Parkland/Northern Glaciated Plains
Lake Manitoba and Lake Agassiz Plain
Western Corn Belt Plains
Central Irregular Plains
Northwestern Glaciated Plains
Northwestern Great Plains
Nebraska Sand Hills
CL75
112.8
60.2
54.0
79.3
147.2
90.2
113.9
100.0
281.6
174.3
386.4
71.0
152.6
755.3
184.9
209.5
271.0
881.0
637.2
888.1
396.2
943.6
919.3
878.9
1050.1
931.2
386.2
574.0
803.4
719.4
1039.4
154.1
261.0
492.9
495.7
145.4
173.9
354.4
CL50
113.5
64.4
70.3
86.1
149.2
90.6
116.1
100.6
283.0
184.0
392.0
71.3
153.0
766.3
196.7
212.7
282.7
947.7
653.3
896.3
401.2
948.7
926.3
884.5
1058.0
939.8
403.8
581.0
815.5
725.8
1052.2
154.8
261.7
495.5
500.4
145.6
174.5
355.8
CL35
113.8
66.1
77.2
89.6
150.4
90.9
117.4
101.0
283.8
189.9
393.9
71.4
153.3
772.9
201.6
222.3
301.0
957.6
662.9
901.2
402.6
951.8
930.4
887.8
1062.7
944.9
414.4
585.1
822.7
729.6
1059.9
155.3
262.1
497.0
503.2
145.7
174.9
356.6
CL20
114.2
66.9
87.1
92.2
151.6
91.1
118.8
101.3
284.6
195.7
395.9
71.6
153.6
782.4
230.2
223.1
325.1
992.1
672.6
906.1
404.0
954.9
934.6
891.1
1067.5
950.1
425.0
589.3
829.9
733.4
1067.7
155.7
262.5
498.6
505.9
145.8
175.3
357.5
AAI75
3618.4
877.6
133.6
1814.0
2083.9
969.1
1133.5
3284.1
2174.8
361.8
3253.8
3076.7
2032.3
1045.5
335.6
867.3
232.6
715.8
974.0
1314.6
1428.7
3951.7
2620.3
3705.1
3880.7
2295.2
401.3
1799.0
1357.2
2202.4
2069.5
6663.8
7748.4
3006.8
1800.5
9116.2
5810.6
4103.2
AAI50
3618.4
932.5
140.3
1989.3
2093.0
950.0
1140.1
3287.5
2162.5
366.0
3285.5
3073.1
2015.0
1035.2
329.3
855.6
217.0
741.9
978.8
1305.5
1426.9
3950.9
2618.9
3707.2
3889.9
2297.0
410.7
1804.5
1357.7
2200.1
2072.4
6690.8
7750.1
3002.3
1796.5
9105.5
5812.9
4096.8
AAI35
3618.4
950.5
139.0
2077.6
2098.5
938.6
1144.0
3289.5
2155.1
368.5
3290.0
3071.0
2004.6
1029.0
322.2
880.7
216.7
734.5
981.8
1300.1
1418.5
3950.4
2618.1
3708.4
3895.4
2298.2
416.4
1807.7
1357.9
2198.7
2074.2
6707.0
7751.1
2999.6
1794.0
9099.0
5814.2
4093.0
AAI20
3618.4
951.3
143.3
2141.6
2104.0
927.1
1147.9
3291.6
2147.7
371.1
3294.5
3068.8
1994.2
1026.7
352.7
869.0
221.2
746.0
984.7
1294.6
1410.1
3950.0
2617.2
3709.6
3900.9
2299.3
422.0
1811.0
1358.2
2197.3
2075.9
6723.1
7752.1
2996.9
1791.6
9092.6
5815.6
4089.2
D-20
-------
Table D-4. Critical Loads (meq/(m2-yr) and calculated AAI values for relatively less non-acid
sensitive ecoregions (note all calculations based on nth %percentile = 50).
Region
9.4.1
9.4.2
9.4.3
9.4.5
9.5.1
15.4.11
5.2.21
8.2.21
9.6.1
10.2.11
10.2.41
7.1.91
8.3.61
9.4.61
9.4.41
9.4.71
8.3.81
Name
High Plains
Central Great Plains
Southwestern Tablelands
Cross Timbers
Western Gulf Coastal Plain
Southern Florida Coastal Plain
Northern Minnesota Wetlands
Huron/Erie Lake Plains
Southern Texas Plains/Interior Plains and Hills
with Xerophytic Shrub and Oak Forest
Mojave Basin and Range
Chihuahuan Desert
Willamette Valley
Mississippi Valley Loess Plains
Edwards Plateau
Flint Hills
Texas Blackland Prairies
East Central Texas Plains
CL75
101.7
197.5
86.5
309.4
738.9
281.6
281.6
281.6
281.6
281.6
281.6
281.6
281.6
281.6
281.6
281.6
281.6
CL50
102.0
198.9
86.7
312.6
750.9
283.0
283.0
283.0
283.0
283.0
283.0
283.0
283.0
283.0
283.0
283.0
283.0
CL35
102.2
199.7
86.9
314.5
758.0
301.0
301.0
301.0
301.0
301.0
301.0
301.0
301.0
301.0
301.0
301.0
301.0
CL20
102.4
200.5
87.0
316.5
765.1
316.5
316.5
316.5
316.5
316.5
316.5
316.5
316.5
316.5
316.5
316.5
316.5
AAI75
1632.4
2340.2
5535.8
1916.1
1416.0
931.9
1166.4
547.1
1086.2
1291.0
1265.2
1109.3
783.1
1087.3
939.4
835.5
881.2
AAI50
1618.2
2341.2
5538.1
1918.2
1416.0
913.4
1147.9
528.6
1067.7
1272.5
1246.7
1090.9
764.6
1068.8
920.9
817.0
862.7
AAI35
1609.7
2341.8
5539.5
1919.4
1416.0
986.0
1220.5
601.3
1140.3
1345.1
1319.3
1163.5
837.2
1141.4
993.5
889.7
935.3
AAI20
1601.3
2342.4
5540.8
1920.7
1416.0
1046.1
1280.6
661.4
1200.4
1405.2
1379.4
1223.6
897.3
1201.5
1053.6
949.7
995.4
1 - indicates relatively non-acidless- sensitive regions with data from less than 10 water bodies; CL and Qr values are based on median values of
all the relatively non-acid sensitive regions with data from > 10 water bodies with data.
D-21
-------
Table D-5. CMAQ 2005 annual average depositions, transference ratios, and Neco vales. All
depositions and Neco in meq/(m2-yr); T ratios in m/yr; Ndep = NOydep + NHxdep.
EcoRgnlll
3111
10.1.2
10.1.3
10.1.4
10.1.5
10.1.6
10.1.7
10.1.8
10.2.1
10.2.2
10.2.4
11.1.1
11.1.2
11.1.3
12.1.1
13.1.1
15.4.1
5.2.2
6.2.11
6.2.8
6.2.9
7.1.7
7.1.8
7.1.9
8.1.1
8.1.10
8.1.4
8.1.5
8.1.6
8.2.1
8.2.2
8.2.3
8.2.4
8.3.1
8.3.2
8.3.3
8.3.6
8.3.8
8.4.5
Name
Columbia Plateau
Northern Basin and Range
Wyoming Basin
Central Basin and Range
Colorado Plateaus
Arizona/New Mexico Plateau
Snake River Plain
Mojave Basin and Range
Sonoran Desert
Chihuahuan Desert
California Coastal Sage, Chaparral, and Oak Woodlands
Central California Valley
Southern and Baja California Pine-Oak Mountains
Madrean Archipelago
Arizona/New Mexico Mountains
Southern Florida Coastal Plain
Northern Minnesota Wetlands
Klamath Mountains
Eastern Cascades Slopes and Foothills
Blue Mountains
Strait of Georgia/Puget Lowland
Coast Range
Willamette Valley
Eastern Great Lakes and Hudson Lowlands
Erie Drift Plain
North Central Hardwood Forests
Driftless Area
S. Michigan/N. Indiana Drift Plains
Southeastern Wisconsin Till Plains
Huron/Erie Lake Plains
Central Corn Belt Plains
Eastern Corn Belt Plains
Northern Piedmont
Interior River Valleys and Hills
Interior Plateau
Mississippi Valley Loess Plains
East Central Texas Plains
Ozark Highlands
Total
NDep
19.2
14.2
15.7
15.6
21.3
19.3
33.1
26.1
23.6
24.5
45.8
68.8
59.6
23.4
24.6
64.6
39.8
20.9
16.3
15.9
50.3
25.2
51.8
82.1
111.6
70.8
80.8
90.0
81.0
95.0
89.2
104.0
110.0
87.1
90.5
80.6
73.7
76.4
NHx
Dep
8.9
5.6
4.7
4.4
5.8
4.7
21.6
4.9
7.1
7.8
11.4
34.4
12.1
8.6
6.5
10.1
19.5
6.1
4.9
6.0
22.7
7.5
27.8
35.3
37.6
38.5
46.5
32.7
37.6
38.1
33.8
40.0
37.9
32.3
27.4
27.7
30.4
32.2
SOx
Dep
3.6
4.3
8.3
4.7
9.0
7.5
6.2
5.4
5.7
12.4
14.6
9.8
10.9
9.1
8.6
40.8
17.4
9.7
4.5
4.1
25.4
20.9
17.0
70.4
140.1
34.2
39.5
78.5
51.2
89.5
79.6
116.8
108.6
91.6
106.5
55.4
42.1
47.0
TSOx
8.9
15.3
13.3
14.3
19.3
14.1
14.8
8.6
6.7
11.6
9.8
7.0
11.8
12.2
18.1
17.6
21.5
30.8
16.6
15.5
15.5
39.6
15.7
20.1
20.6
17.4
16.1
16.5
14.6
16.7
14.2
17.3
13.4
15.3
16.8
15.5
13.9
16.0
TNOy
7.1
14.9
11.9
13.4
16.4
12.2
7.6
11.2
8.5
15.3
7.5
5.5
13.5
13.9
18.1
12.9
13.9
17.6
15.2
13.5
5.0
14.4
4.6
10.8
10.9
7.9
8.3
8.3
6.9
7.9
7.2
8.5
6.8
9.7
11.5
11.5
11.8
14.0
Neco
17.1
13.1
14.8
15.1
19.8
18.7
27.3
NA
NA
24.4
32.7
67.9
57.7
23.2
24.4
NA
NA
18.9
15.7
15.5
42.7
17.2
NA
78.7
90.0
69.2
0.9
86.7
51.7
NA
-36.1
87.3
39.1
81.5
80.2
NA
NA
75.6
D-22
-------
Table D-5. CMAQ 2005 annual average depositions, transference ratios, and Neco vales. All
depositions and Neco in meq/(m2-yr); T ratios in m/yr; Ndep = NOydep + NHxdep.
EcoRgnlll
3111
8.5.2
9.2.1
9.2.2
9.2.3
9.2.4
9.3.1
9.3.3
9.3.4
9.4.1
9.4.2
9.4.3
9.4.4
9.4.5
9.4.6
9.4.7
9.5.1
9.6.1
5.2.1
5.3.1
5.3.3
6.2.10
6.2.12
6.2.13
6.2.14
6.2.15
6.2.3
6.2.4
6.2.5
6.2.7
8.1.3
8.1.7
8.1.8
8.3.4
8.3.5
8.3.7
8.4.1
8.4.2
8.4.3
Name
Mississippi Alluvial Plain
Aspen Parkland/Northern Glaciated Plains
Lake Manitoba and Lake Agassiz Plain
Western Corn Belt Plains
Central Irregular Plains
Northwestern Glaciated Plains
Northwestern Great Plains
Nebraska Sand Hills
High Plains
Central Great Plains
Southwestern Tablelands
Flint Hills
Cross Timbers
Edwards Plateau
Texas Blackland Prairies
Western Gulf Coastal Plain
Southern Texas Plains/Interior Plains and Hills with
Xerophytic Shrub and Oak Forest
Northern Lakes and Forests
Northern Appalachian and Atlantic Maritime Highlands
North Central Appalachians
Middle Rockies
Sierra Nevada
Wasatch and Uinta Mountains
Southern Rockies
Idaho Batholith
Columbia Mountains/Northern Rockies
Canadian Rockies
North Cascades
Cascades
Northern Appalachian Plateau and Uplands
Northeastern Coastal Zone
Maine/New Brunswick Plains and Hills
Piedmont
Southeastern Plains
South Central Plains
Ridge and Valley
Central Appalachians
Western Allegheny Plateau
Total
NDep
73.4
46.3
50.7
86.9
73.8
29.1
23.4
43.6
45.1
61.6
32.6
70.6
66.3
50.6
81.7
62.9
46.2
46.1
56.7
80.4
20.9
35.9
28.4
22.6
18.5
18.5
19.6
27.2
30.0
76.2
82.4
36.4
84.5
73.2
72.1
86.0
90.9
98.1
NHx
Dep
26.4
29.2
33.6
53.9
33.4
15.7
10.2
23.4
23.3
30.2
12.1
31.5
24.2
14.7
33.1
20.4
13.3
18.4
17.9
18.8
8.6
11.3
9.1
6.0
7.4
6.8
7.1
8.0
10.1
22.7
21.5
10.9
27.6
25.8
24.4
27.0
20.6
23.1
SOx
Dep
48.9
15.9
15.0
34.5
40.2
11.2
10.3
12.4
12.1
19.4
10.9
33.3
25.0
22.8
43.5
37.0
27.4
27.1
48.5
104.2
7.8
11.1
9.4
8.9
5.8
6.4
11.7
12.2
13.2
82.4
81.7
39.3
82.0
59.7
44.5
96.9
133.3
174.8
TSOx
14.2
17.1
18.5
15.4
14.4
15.5
17.4
18.7
14.1
15.0
13.7
15.1
11.2
17.0
11.6
14.9
15.7
20.2
25.1
20.9
25.4
30.5
22.6
27.0
26.7
19.6
17.3
35.8
36.1
19.2
17.9
29.7
15.3
15.6
16.7
16.3
23.1
20.5
TNOy
9.2
10.1
9.0
9.0
10.0
9.1
12.5
13.7
10.1
12.0
12.7
11.8
10.5
16.6
8.7
9.0
15.5
14.3
15.9
15.4
20.5
25.0
17.6
22.3
21.4
13.6
17.4
22.0
17.3
14.3
7.1
18.3
10.4
12.6
13.4
10.9
18.1
12.5
Neco
70.4
46.0
50.1
40.1
71.0
28.8
22.9
43.4
41.8
60.2
32.5
NA
65.9
NA
NA
60.8
NA
45.6
52.2
71.1
19.9
35.2
26.8
21.3
17.8
17.9
15.6
25.0
27.9
59.3
77.7
35.8
80.1
60.9
70.4
61.8
74.1
78.7
D-23
-------
Table D-5. CMAQ 2005 annual average depositions, transference ratios, and Neco vales. All
depositions and Neco in meq/(m2-yr); T ratios in m/yr; Ndep = NOydep + NHxdep.
EcoRgnlll
3111
8.4.4
8.4.6
8.4.7
8.4.8
8.4.9
8.5.1
8.5.3
8.5.4
Name
Blue Ridge
Boston Mountains
Arkansas Valley
Ouachita Mountains
Southwestern Appalachians
Middle Atlantic Coastal Plain
Southern Coastal Plain
Atlantic Coastal Pine Barrens
Total
NDep
84.2
81.9
75.5
72.0
88.9
87.0
66.2
91.5
NHx
Dep
24.4
36.0
35.2
27.1
30.4
33.9
15.3
22.1
SOx
Dep
87.1
42.2
37.3
40.6
91.2
82.6
51.2
98.3
TSOx
21.3
17.9
14.5
16.8
18.1
17.9
18.4
16.1
TNOy
17.8
17.7
12.2
16.8
14.1
11.7
13.6
6.8
Neco
79.5
79.0
66.0
71.0
72.5
4.4
62.9
68.1
1- NA indicates a lessrelatively non-acid sensitive area with less than lOo water bodies with water chemistry data.
D-24
-------
D.4 Relating exceedances to ambient air indicators: introducing trade-off curves
The exceedance calculations based on deposition reflect a straightforward approach to
determine if the AAI is met, and serves the purpose of explaining current conditions with respect
various combinations of alternative levels and forms of the standard. In practice, a state agency
would use observed ambient concentrations of NOy, SO2 and SO4 and calculate the AAI using
equation
AAI = Fl -F2 -F3[NOy] - F4[SOx]
An alternative standard would be met if the calculated AAI is equal to or above the alternative
standard level. An infinite amount of combinations of NOy and SOx concentrations can satisfy
the condition that the calculated AAI be greater than or equal to the level of the standard.
NOx/SOx tradeoff curves, using nitrogen and sulfur axes are used to illustrate the possible
combinations of nitrogen and sulfur that could meet a target AAI. The following discussion on
tradeoff curves illustrates how NHx and Neco affect the AAI calculations, and demonstrate the
range of tradeoff options between NOy and SOx for attaining a hypothetical standard.
NOx and SOx tradeoff curves are introduced using deposition axes to illustrate (Figure
D-10) the basic concepts of combinations of NOx and SOx deposition and the role of Neco and
NHx deposition. The tradeoff curve is based on the critical load equation described earlier,
= ([BC]o - [ANClim])Q + Neco (D-l)
expressed in terms of nitrogen (zero sulfur deposition) and sulfur (zero nitrogen deposition)
deposition only:
([BC]o - [ANClm])Q + Neco (D-2a)
([BC]0* - [ANClim])Q (D-2b)
D-25
-------
Equations D-2a and D-2b provide the x and y intercepts of the basic tradeoff curve (Figure D-
10). Neco reflects how much nitrogen deposition can be neutralized before it contributes to
acidification so, as the two equations imply, Neco can only affect nitrogen deposition.
Smax
tr
0)
NHxdep
2
Ndep (eq)
Neco
5
f
Nmax
Figure D-10. Hypothetical tradeoff diagram for acidifying nitrogen and sulfur deposition. The
solid horizontal and diagonal lines represent the combinations of nitrogen and sulfur deposition
to meet a target ANC. Deposition values within the area bounded by these lines do not exceed
the target critical load and values outside the borders exceed a critical load. When all units are
transformed to equivalent charge, there is a 1:1 slope in the region where nitrogen deposition
exceeds Neco which illustrates the area of tradeoffs between nitrogen and sulfur deposition.
Because, NHx is not treated as an ambient air indicator, NHx deposition is extracted from the
tradeoff curve by reducing the available pool of Neco and changing the total nitrogen to a an
NOy deposition axis (Figure D-l 1):
= ([BC]o - [ANCUm])Q + Neco-NRx
D-26
-------
Smax ^ 2
er
-------
Additional Explanation of Developing Tradeoff Curves
The "basic" acid-base charge balance equation,
CL(Acidify) = BCdep + BCw + Nup + Nde + Ni - BCup - ANClimit (D-3)
Where CL(Acidity) represents the deposition flux of nitrogen and sulfur beyond which a
prescribed ANClimit is not met. The BC terms represent the contribution of major cations due to
weathering (BCw) and atmospheric deposition (BCdep) and loss through plant uptake (BCup).
We assume that BCup is negligible and combine cation contributions due to weathering and
deposition to define BCo:
BCo = BCdep + BCw
BCo integrates at the watershed level of deposition of BC from the air and weathering processes.
The N terms represent the ecosystems capacity for neutralizing nitrogen deposition through
plant uptake, denitrification and immobilization, largely through adsorption, which collectively
is represented by the term, Neco.
Neco = Nup + Nde + Ni
Neco is estimated by subtracting measured surface water outflow nitrate (leached nitrogen) from
total nitrogen deposition (oxidized and reduced). Substituting the expressions for BCo and Neco
into equation (1) and accounting for catchment runoff, Q to retain mass consistency between
deposition fluxes and concentrations yields the basic critical load expression for a specific ANC
target at the catchment level:
CLANClim(N + S) = ({BC]0 - [ANClim])Q + Neco (D_4)
Equation (2) defines the maximum amount of acidifying deposition a watershed can handle to
meet a prescribed ANC. Recognizing that Neco only affects depositing nitrogen, the max S
deposition when N deposition < Neco is defined as:
CLANClm(S] = ([BC]0 - [ANC^Q (D-5)
The max N deposition when S deposition = 0
D-28
-------
CLANClm(N) = ([BC]0 - [ANClm])Q + Neco (D-6)
Tradeoff Curves
Equations (D-5) and (D-6) establish the y and x intercepts, respectively, of a hypothetical
Nitrogen/Sulfur tradeoff curve (Figure D-10). When all units are transformed to equivalent
charge, there is a 1 : 1 slope in the region where nitrogen deposition exceeds Neco which
illustrates the possibilities of combined nitrogen and sulfur deposition to meet a target ANC.
Note that the concept of tradeoff between S and N is best realized where nitrogen deposition
exceeds Neco. As long as nitrogen deposition is less than Neco, the amount of allowable sulfur
deposition remains constant at Smax. Any deposition value above the horizontal and diagonal
lines reflect levels greater than the targeted critical load. The terms Ndep and Sdep reflect
deposition of total nitrogen and SOx, respectively.
The N/S tradeoff diagram is based on the following conditions:
1 . CL(S) = Smax, when Ndep < Neco;
2. For a given nitrogen deposition , Ndep; the allowable sulfur deposition, CL(S) = Smax
- (Ndep - Neco), when Nmax> Ndep > Neco
The following conditions must be met in order to not exceed a CL;
Sdep is always <
and,
Sdep + Ndep < ([5C£ ~[ANCl]m])Q+ Neco
Note that Nmax - Neco = Smax
Delineating reduced nitrogen, NHx, deposition from total.
Because the standard addresses oxidized forms of nitrogen, it is useful to illustrate how reduced
nitrogen deposition conceptually is incorporated in the deposition tradeoff diagram. Figure
D-l 1 describes the tradeoff diagram in terms of reactive oxidized nitrogen (NOy) deposition and
sulfur deposition. The NHx deposition displayed in Figure D-l 1 effectively is "neutralized" by
D-29
-------
Neco, and therefore the remaining "pool" of Neco is available to neutralize NOy deposition. The
impact of NHx deposition impacts the "allowable" NOy deposition to meet a target critical load.
Since Nmax incorporates reduced and oxidized forms of nitrogen, the level of NHx directly
impacts the amount of allowable NOy deposition where:
Maximum NOy deposition = Nmax - NHx
In equation form, the removal of NHx to develop oxidized nitrogen and SOx deposition tradeoff
curves is represented as:
The max S deposition when N deposition < Neco
S max = CLANC]im (S max) = ({BCl - [ANC^ ])Q (D-7)
The max NOy deposition when S deposition is zero
NOymax = CLANC^ (M^max) = ([BC]0 - [ANC^ ])Q + Neco - NHx (D-8)
Or, NOy max = Nmax - NHx
Case where NHx > Neco
In cases where NHx is greater than Neco, the amount of maximum NOy deposition is decreased
by a quantity equal to the difference between NHx dep and Neco as described by equation (D-8)
(e.g., consider two cases of NHx = 1 and 4 with Neco a constant at 3). Equation (D-7) by
definition remains satisfied in that Smax is defined as the allowable amount of sulfur deposition
in the absence of any nitrogen deposition. However, because there always remains some residual
atmospheric nitrogen burden in the form of NHx when all NOy is removed, the allowable sulfur
deposition in concept also is reduced an amount equal to the difference between NHx dep and
Neco, which is consistent with the condition that:
CLANCiim(S + NOy) < ([BC]0* - [ANCHm])Q +Neco - NHx
Smaxa< ([BC]0* - [ANCHm])Q +Neco -NHx
NOymax < ([BC]0* - [ANCHm])Q +Neco - NHx
D-30
-------
Where Smaxa is the amount of allowable sulfur deposition in the presence of excess NHx (Figure
D-12).
Smax 2
\
CT
QJ
r 1
Smax"
OJ
T3
-1
0
1
2
NOydep(eq)
I
i
i
3
t
Neco
i
/
Nl
\
t
\
ix
5
f
Nmax
Figure D-12. NOy and SOx deposition tradeoff curve for hypothetical case where NHx
deposition exceeds Neco. The effect can be visualized as the system retaining a residual one
equivalent unit of NHx that can not be neutralized by Neco with the consequent reduction on
both allowable maximum amounts of SOx and NOy deposition.
Relating Concentrations of NOy and SOx to critical loads
Starting with:
1. CLANCiim(N+S) = (BC*0 - ANCHm)-Q + Neco;
2. Substituting CLANciim(N+S) = Ndep + Sdep = NOydep + NHx + Sdep (conceived as any
combinations NOy, S, NHx deposition that = CLANciim(N+S)
3. NOydep + NHx + Sdep = (BC*0 - ANCHm)-Q + Neco;
4. NOydep + Sdep = (BC*0 - ANCHm)-Q + (Neco - NHx);
5. Converting deposition to concentration using Sdep =[SOx]-TSox; NOydep = [NOy]-TNOy;
6. [SOx]-TSOx + [NOy]-TNOy = (BC*0 - ANCHm)-Q + Neco - NHx;
7. [NOy]CL = [((BC*0 - ANCHm)-Q + Neco - NHx) - Sdep]/TNOy = CLANCiim - NHx - Sdep)/
D-31
-------
defining the maximum allowable NOy concentration for any value of Sdep and a fixed value of
NHx.
8. [SOx]CL = [((BC*0 - ANClim)-Q + Neco - NHx) - NOydep]/TSOx = CLANCiim -NHx -
Sdep)/ TSOX ;;
defining the maximum allowable SOx concentration for any value of NOydep and a fixed value
of NHx.
With the condition that when Ndep < Neco;
9. [SOxJcL = [((BC o - ANCiim)-Q ) /Tsox, a constant value as all nitrogen deposition is
neutralized
Maximum allowable NOy and SOx concentrations defined as in the absence of the other
component are defined as:
[NOy]max =( Nmax - NHx)/ TNoy = NOymax/ TNoy in the absence of any SOx deposition
[SOx]max = Smax/ TSox ; in the absence of any N deposition
We also can define a maximum SOx concentration in cases where NHx exceeds Neco;
[SOx]amax = [Smax - (NHx-Neco)]/ TSOX ; in the absence of NOy deposition and excess NHx
beyond Neco.
D-32
-------
Appendix E
Derivation to use measured total nitrate as a surrogate for NOy
Our recommended measurement approach is to use NOy monitors to judge compliance
with a NOx/SOx secondary standard. This derivation illustrates how the AAI equation would be
modified to use total nitrate data from the CASTNET filter packs.
Two general approaches enable the use of measured total nitrate to estimate the level of
NOy consistent with the AAI formulation. The first approach introduces an aggregated dry
oxidized nitrogen deposition transference ratio to separate total nitrate and NOy and yield NOy
as a function of measured nitrate, based on CMAQ attributes consistent with the AAPI
formulation. The second approach would be to adjust measured nitrate upward to represent
NOy. This adjustment would be performed by multiplying measured total nitrate by the CMAQ
derived ratio of (ambient NOy/ambient total nitrate) for the acid sensitive area of interest.
1. Rearranging AAPI expression to show NOy as a function of measured total nitrate
AAI= —
[Q
Neco+[BCl
= NOywetdep 4
%eco ^~f
TII s~i amb
- TdryaNOy cNOy
Q M0y
= NOywetdep
r /"iamb , /"iamb -\
[CTNO3 + C(NOy_mo3)]
/~< amb rpj r /~
-------
2. Adjust measured total nitrate upward to estimate ambient NOy. This adjustment could be
performed multiplying by the CMAQ NOy/TNO3 ratio, or by using the ratio of
collocated NOy and nitrate measurements.
Conclusions
Nitrate does not account for a significant portion of the ambient air oxidized nitrogen budget and
can also miss a significant part of oxidized nitrogen deposition. These two concerns combined
with the dynamic response of nitrate to changes in emissions, as discussed in chapter 2, outweigh
any benefit to be derived from using an available, partial indicator like total nitrate.
E-2
-------
Appendix F
Evaluation of Variability, Sensitivity and Uncertainty in the Acidification Index
F.I INTRODUCTION AND PURPOSE
This appendix provides analyses and discussion of the relative uncertainty in the AAI
equation. This includes analyses of the individual and combined components of the linked
atmospheric-ecological effects system described in chapter 7, as well as important uncertainties
in the scientific evidence that should be considered in developing options for the standard. This
appendix is intended to integrate a variety of analyses related to the sensitivity of the models and
model components to uncertainty and variability, and place the results of those analyses within
the context of the conclusions that can be drawn regarding the components of the AAI. These
components include ecosystem effects; dose-response relationships; underlying ecosystem
sensitivity to acid deposition, biogeochemical, atmospheric and deposition processes; and
characterization of ecosystem services. While several processes are imbedded in the AAI
equation introduced in chapter 7, the level of the AAI, as in all NAAQS, is to include
consideration of information on uncertainty and variability. Consequently, knowledge of the
relative confidence and natural variability in the structural components of the AAI are considered
in staff conclusions on options for ranges of the level of the standard. These analyses are not
intended to be a comprehensive treatment of all uncertainties that exist relative to the overall
review of the standards, instead, it focused on those that are most relevant in evaluating choices
regarding the AAI form of the standard and options regarding the indicator, averaging time, and
ranges of levels of the AAI-based NOx and SOx standard.
Uncertainty and sensitivity analyses are used to inform the relative confidence in the
components and models that are used in defining the standard. Assessments of variability in the
data used to determine parameters of the standard increases the level of understanding about the
likelihood that alternative parameterizations of the standard will achieve targeted levels of
protection when applied to sensitive ecosystems across the U.S. Assessments of the sensitivity of
the overall AAI to the components of the equation proposed to calculate the AAI can help
demonstrate how important uncertainty and variability in those components are in assessing the
protection of ecosystems provided by an AAI standard. To evaluate the potential interactions
F-l
-------
between uncertain and/or variable AAI components, a multifactor sensitivity analysis is also
conducted. The ranges of component values evaluated in the multifactor sensitivity assessment
are guided by individual variability and uncertainty analyses of specific components. In addition
to informing considerations of the AAI level, an additional objective of these "confidence"
related analyses and discussions is to help guide research and data collection efforts intended to
reduce uncertainty for future NAAQS reviews and implementation efforts. Spatial and
temporal variability analyses of AAI components are especially useful to inform monitoring
network design, the spatial boundaries of acid sensitive regions, and averaging periods relevant
to NAAQS implementation.
Significant emphasis is placed on evaluations of CMAQ due to the unique role that
atmospheric models hold in the linked AAI system. The AAI as currently formulated relies on
CMAQ for both the initial characterization of reduced nitrogen deposition, and the deposition
transformation ratios (TNox and TSox) which characterize the relationships between atmospheric
concentrations of NOx and SOx and deposition of N and S. Included are interpretations of
model evaluation results from the REA (EPA, 2009) as well as more recent results related to wet
deposition and the treatment of ammonia deposition. Comparison of model results to
observations provides a general sense of the confidence we have that the models capture the
spatial, temporal and compositional texture of the relevant atmospheric and deposition species
that drive the linked atmospheric-ecosystem processes. Both model evaluation results and
assessments of spatial and temporal variability guide implementation strategies for monitoring
network design and emission inventory improvement. Sensitivity of CMAQ derived deposition
transformation ratios to changes in emissions, and treatment of chemistry and variability over
time provide insight into the stability of these parameters that are used in a relatively static
manner in the AAI, and into how well proposed averaging times capture the overall spatial and
temporal trends in the parameters.
We evaluate the sensitivity of critical load modeling components by comparing dynamic
(MAGIC) and hybrid steady state model results, looking at terminal results of MAGIC. This
approach was viewed as a test of the more reduced form approximations used in steady state
modeling relative to more sophisticated treatment in MAGIC. The MAGIC critical load
simulations also provide information on the temporal trajectory of ANC, including the expected
time necessary to reach a desired ANC, which can help inform the level of the AAI, recognizing
F-2
-------
that there may be additional consideration given to reaching a target ANC within a specific
timeframe, e.g. by 2030 or 2040.
For the purposes of this discussion, we characterize uncertainty regarding models and
their outputs as referring to the lack of knowledge regarding both the actual values of model
input variables (parameter uncertainty) and the model characterization of physical systems or
relationships (model uncertainty). In any application, uncertainty is, ideally, reduced to the
maximum extent possible, but significant uncertainty often remains. It can be reduced by
improved measurement and improved model formulation. Model evaluation results provide
some insight into the relative uncertainty associated with the ability of models to capture key
environmental state characteristics. Confidence regarding the fundamental science supporting
causal determinations about the effects of acid deposition, and the translation of those effects
into ecosystem services and values is less amenable to quantification. As a result, these
uncertainties are more difficult to explicitly account for in development of the standards. In the
case of the equation describing the AAI, a Monte Carlo style analysis (described in Appendix G)
was used to assess a the combined influences of parameter uncertainty. In addition, we
evaluated the sensitivity of the AAI to its components using an elasticity analysis. The results of
these assessments are addressed in section 7.6.
Sensitivity refers to the influence on modeled results due to perturbations in input
variables or change of process formulations. Sensitivity analysis can provide a sense of how
important different parameters and inputs might be to the outcomes of interest, e.g. the AAI
level, but cannot by themselves indicate how important specific parameters actually are, because
they do not incorporate information on the range of parameter values or the likelihood associated
with any specific parameter value. Sensitivity results in this PAD are intended to provide insight
into the relative stability of the AAI and associated NOx and SOx tradeoff curves and confidence
in modeled parameterizations. Sensitivity analyses are especially useful in the absence of
observed data to challenge models. For example, the NOy and SOx transference ratios are a
model construct that is difficult, if not impossible, to compare to observations. The sensitivity of
these ratios to changing meteorology, emissions and chemical mechanism treatments is evaluated
in reference to the stability of these ratios under changing conditions. Low sensitivity here
implies that the choice to use long-term averages of modeled ratios is justified. Sensitivity
analyses also are used to discern the relative influence (on AAI results) of AAI parameters.
F-3
-------
Toward that end, elasticity analyses were applied to determine the relative sensitivity of AAI
results associated with individual and combined AAI parameters.
Variability refers to the heterogeneity in a population or variable of interest that is
inherent and cannot be reduced through further data collection and research. In the context of
the AAI and trade-off curves, variability is considered in guiding the design of monitoring and
modeling analyses supporting implementation activities.
F.2 Uncertainty associated with ecosystem effects and dose - response relationships.
This section provides a brief summary of uncertainties based on the REA and is
reproduced here to centralize all uncertainty discussions. There are different levels of
uncertainty associated with relationships between deposition, ecological effects and ecological
indicators. In Chapter 7 of the REA, the case study analyses associated with each targeted effect
area were synthesized by identifying the strengths, limitations, and uncertainties associated with
the available data, modeling approach, and relationship between the ANC and atmospheric
deposition. The key uncertainties were characterized as follows to evaluate the strength of the
scientific basis for setting a national standard to protect against a given effect (REA 7.0):
• Data Availability: high, medium or low quality. This criterion is based on the
availability and robustness of data sets, monitoring networks, availability of data that
allows for extrapolation to larger assessment areas, and input parameters for modeling
and developing the ecological effect function. The scientific basis for the ecological
indicator selected is also incorporated into this criterion.
• Modeling Approach: high, fairly high, intermediate, or low confidence. This value is
based on the strengths and limitations of the models used in the analysis and how
accepted they are by the scientific community for their application in this analysis.
• Ecological Effect Function: high, fairly high, intermediate, or low confidence. This
ranking is based on how well the ecological effect function describes the relationship
between atmospheric deposition and the ecological indicator of an effect.
The REA concludes that the available data are robust and considered high quality. There is
high confidence about the use of these data and their value for extrapolating to a larger regional
population of lakes. The EPA TIME/LTM network represents a source of long-term,
F-4
-------
representative sampling. Data on sulfate concentrations, nitrate concentrations and ANC from
1990 to 2006 used for this analysis as well as EPA EMAP and REMAP surveys, provide
considerable data on surface water trends.
There is fairly high confidence associated with modeling and input parameters. Uncertainty in
water quality estimates (i.e., ANC) from MAGIC was derived from multiple site calibrations.
The 95% confidence interval for pre-acidification of lakes was an average of 15 ueq/L difference
in ANC concentrations or 10% and 8 ueq/L or 5% for streams (REA 7.1.2). The use of the
critical load model used to estimate aquatic critical loads is limited by the uncertainties
associated with runoff and surface water measurements and in estimating the catchment supply
of base cations from the weathering of bedrock and soils (McNulty et al, 2007). To propagate
uncertainty in the model parameters, Monte Carlo methods were employed to develop an inverse
function of exceedances. There is high confidence associated with the ecological effect function
developed for aquatic acidification. In calculating the ANC function, the depositional load for N
or S is fixed by the deposition of the other, so deposition for either will never be zero (Figure
F.I-6 REA).
Chapter 3 also reviews the basic evidence underlying effects on fish mortality, aquatic species
diversity and more extended food web disruptions leading to adverse impacts on birds associated
with aquatic acidification. There is high confidence associated with correlation between
acidification and these ecological effects. Also, there is high confidence in the relationship
between the ecological indicator, ANC, and the more direct chemical properties (lower pH and
increased Al) associated with acidification.
F.3 Uncertainty in benefits estimates
Descriptions of the current provision of ecosystem services presented for each of the
effect areas analyzed for this review followed by estimations of the damages incurred to selected
services due to nitrogen and sulfur deposition. The current services are presented to give the
reader a sense of the magnitude of the benefit the public receives from these ecosystems under
current conditions. The data used in these descriptive passages is generally derived from
government (either federal or state) sources we are reasonably certain to be of the highest
quality. Where monetary values are placed on the these services we have generally used widely
cited studies, particularly meta analyses that provide an average value that smoothes the variation
in WTP estimates. These estimates underestimate the total value of these services as they use
F-5
-------
benefit estimates for a marginal increase in these services. It is likely that the total benefits of
these services are greater because their marginal value likely is lower than the average value.
While reductions in sulfur and nitrogen emissions would increase the size of the benefits from
these services, for many of them it is unknown how significant the increase will be.
The analyses of damages incurred are more uncertain and are limited to those areas where
data and tools were available. Only some services were analyzed which in some cases meant
that the results were limited to one or two services and in the case of terrestrial nutrient
enrichment no services had sufficient data available to attempt an estimate of damage. This
means that the estimates presented are a very small part of the total damage incurred due to
deposition.
Aquatic Acidification
Recreational Fishing Model
The analysis of recreational fishing damages presented in Chapter 4 is subject to the
assumptions necessary to perform the analysis. The original analysis performed for the REA
was based on projecting future benefits of increased recreational fishing based on a complete
cessation of all nitrogen and sulfur emissions. These decisions under or over estimate the current
damages to public welfare incurred from nitrogen and sulfur deposition. The magnitude of the
bias in results is unknown in either direction however the majority of the assumptions influence
the estimates downward. These include the use of emissions estimates that include projected
decreases due to implementation of Title IV regulations in 2020. These emissions estimates are
lower than current emissions and therefore lead to underestimation of damages. Because the
models only value this improvement for New York residents (without accounting for out-of-state
visitors) the damages are underestimates of the benefits of these improvements in the
Adirondacks region.
The use of projected population in the REA analyses contributes to an overestimate of
current damages since current population is smaller than future population. Further, these
estimates are extrapolated from a 44 lake subset and applied to all Adirondack lakes. The
representativeness of this sample is unknown. This analysis also does not account for any
change in fishing demand (possible overestimate) and income (possible underestimate).
F-6
-------
Benefits Transfer
The approach using the WTP estimates from the Banzhaf et al. (2006) study is subject to
the same uncertainties described above and some additional considerations. Specifically there is
some uncertainty regarding which types of ecosystem services are reflected in the study's
estimates of the improvements in ecosystem services of reducing acidification, particularly
provisioning and regulating services. The values likely include recreational fishing services,
which mean they cannot be added to the recreational fishing model results, and other cultural
services including other recreation and nonuse services. The inclusion in the survey of other
ecosystem changes (birds, trees, etc.) leads to an overestimation of WTP for remediation of lake
acidification alone. Finally, assumptions were required to align the Banzhaf survey scenarios to
the likely results of complete removal of all nitrogen and sulfur emissions. These are reasonably
close but not exact and may not be applicable to another baseline.
Conclusion. While these estimates are subject to uncertainty we are reasonably confident that
they represent a good first-order approximation of the damages to recreational fishing due to
nitrogen and sulfur deposition. Additionally it should be noted that the Banzhaf survey results
represent a broader picture (though by no means complete) of the damages to ecosystem services
in the Adirondacks. Finally, we would again like to emphasize that these estimates represent
only a small sample of the damages incurred to a broad range of ecosystem services affected and
the areas of the nation where acidic deposition is an ongoing issue.
F.4 Uncertainty in the AAI related to component parameters
Uncertainty within the AAI is divided into four components of analysis. First, an
analysis of elasticity; second and third, analyses of uncertainty within individual components for
atmospheric and ecosystem modeling, respectively; fourth, a Monte Carlo style analysis
incorporating the uncertainty derived into the preceding section to assess the cumulative effect of
the uncertainty of the input parameters. This appendix concludes with a summary discussion
that includes more qualitative conclusions.
F.4.1 Elasticity Analyses
An elasticity analysis was applied to investigate sensitivity of the AAI to its components.
The means, medians and quartiles of the AAI component variables were based on the range
variable values across ecoregions that overlapped with the CMAQ domains. Elasticities measure
F-7
-------
the percent change in the AAI for a 1% change in the AAI parameters: Q, Neco, NHX, BC*,
TNOY, TSOX, NOy, and (862 + 804). We note that an earlier form of the expression is used where
Q is at the catchment level. AAI = —Neco + BC* - —NHx - — \TNOy • NOy + TSOx • (SO4 + SO2)]
In general, the formula for elasticity is:
^AAPI (jAAL
pAAfl _
J-^Y —
X dX] AAI
Where E^PI is the elasticity of AAI with respect to component Xj, and j is the number of
components. So, for AAI defined as
AAI = -Neco + BC; - —NHx - - \TNOv • NOy + TSOx • SOx]
^ eco U ^ ^-j L NUy ^ SUx J
The set of relevant elasticities are:
For runoff, Q:
Thrn NOy - TV, SOx x , which can be rewritten as
NOy ^ so* j ^4/.
O
,or
QL J AAI
EfPI = -1 + ^
Q Ml
For BC*,
r'AAPI
^n =
Q AAI
For Neco,
r-^AAPI 1
Q AAI
For NHx,
1 NHx
FAAPI
Q AAI
For TNOy,
E?API =---NOy^
N0y Q AAI
For
F-8
-------
Q AAI
For NOy,
__
Q N0y' AAI
For SOx,
FA4I __J_ T SOX
S0x Q ' S0x ' AAI
These elasticities can be evaluated at various points along the ranges of each component, as well
as along ranges of the AAI. We evaluate the elasticities at the sample means, medians, first
quartiles, and third quartiles. Elasticities are evaluated only for ecoregions that overlap the
CMAQ modeling domain which provides values for reduced nitrogen and the transformation
ratios (TNOX and TSOX). This will provide a reasonable assessment of the sensitivity of the AAI to
input components. Table F-l provides the estimated elasticities. Elasticities are summarized
across ecoregions using means, medians, minimums, and maximums.
Note that elasticities can be either positive or negative. A negative elasticity means that
the calculated AAI will decrease as a component increases. The magnitude of the elasticity
depends on the values of the components and the starting value of AAI.
Based on the calculated elasticities, AAI is most responsive to changes in Q, BCO, and
Neco with some responsiveness to reduced N. Note that for some components, such as Q, the
elasticities switch signs depending on the values of the variables for which the elasticity is
evaluated. This suggests potentially important interactions. AAI is not responsive to the
transformation ratios, TNOx and TSOx at mean values of the AAI components. However, when
the elasticities for TNOx and TSOx are evaluated at the first quartiles of the data, some locations
in the Eastern U.S. show higher responsiveness to changes in TNOx and TSOx, with elasticities
as high as 2.
F-9
-------
Table F-l. Summary of Elasticity Results
AAI
Component
to
o
f§ §
S £P
'3 '§
0 j§ ^
1 ^8
« £ a
0
o
(L>
£
T3 C
S
0 00
3 0
•8 £
rt £
S .2
5— < TJ
Ǥa~
X fi S 0
0 g .2 *
£ H ta b
SOx
Transformation
Ratio (Tsox)
Metric for
Which
Elasticities
are Evaluated
Mean
Median
1st Quartile
3rd Quartile
Mean
Median
1st Quartile
3rd Quartile
Mean
Median
1st Quartile
3rd Quartile
Mean
Median
1st Quartile
3rd Quartile
Mean
Median
1st Quartile
3rd Quartile
Mean
Median
1st Quartile
3rd Quartile
Mean
Elasticity
Across
Ecoregions
-0.1047
0.2561
-0.9283
0.2988
0.8953
1.2561
0.0717
1.2988
0.0179
0.3376
0.3543
0.3016
-0.0409
-0.1407
-0.0308
-0.1128
0.0089
-0.0061
-0.0191
-0.0154
0.0019
-0.0045
0.0040
-0.0058
Median
Elasticity
Across
Ecoregions
0.1221
0.1426
0.1303
0.1110
1.1221
1.1426
1.1303
1.1110
0.1464
0.2563
0.2596
0.1440
-0.0702
-0.1031
-0.1190
-0.0615
-0.0053
-0.0064
-0.0071
-0.0044
-0.0024
-0.0036
-0.0032
-0.0021
Minimum
Elasticity
Across
Ecoregions
-20.4572
-6.2481
-135.2544
-0.1810
-19.4572
-5.2481
-134.2544
0.8190
-13.8545
-4.7203
-115.1112
0.0137
-0.4708
-0.7957
-33.9063
-1.9167
-0.0597
-0.0506
-2.7061
-0.5028
-0.0185
-0.0413
-1.8023
-0.1534
Maximum
Elasticity
Across
Ecoregions
1.6005
2.4578
88.1063
7.5684
2.6005
3.4578
89.1063
8.5684
1.6051
2.5044
137.0526
6.5565
3.9332
1.0061
35.3783
-0.0050
1.0598
0.2751
1.7749
-0.0002
0.3608
0.1091
1.9860
-0.0002
* Elasticity is the percent change in AAI for a one percent change in the component variable.
For example, when evaluated at the means of all component variables, the mean elasticity of
AAI to the runoff variable Q is -0.1047, which means that for each 1 percent increase in Q, the
AAI is reduced by 0.1047 percent.
F-10
-------
Conclusions
Base cation weathering, BC*, and hydraulic flow rate, Q, exerted strong influence on
AAI, an expected result given the explicit dependency evident in the AAI expression. The
transference rations for NOy (T^oy) and SOx (Tsox) exhibited relatively less influence on AAI
calculations than all other parameters when evaluated at means of the variables. However, in
some locations, when evaluated at other values of the variables, AAI can be more sensitive to the
deposition transformation ratios.
These results suggest focusing on the uncertainties in the non-atmospheric inputs,
including base cation weathering and runoff rates, and the implications of those uncertainties in
setting an AAI that will have a high likelihood of providing the targeted level of protection.
F.4.2 Individual Components of the AAI
F.4.2.1 Overview of CMAQ model application
The CMAQ model is a comprehensive, peer-reviewed (Aiyyer et al, 2007), three-
dimensional grid-based Eulerian air quality model designed to simulate the formation and fate of
gaseous and particle (i.e., particulate matter or PM) species, including ozone, oxidant precursors,
and primary and secondary PM concentrations and deposition over urban, regional, and larger
spatial scales (Dennis et al., 1996; U.S. EPA, 1999; Bryun and Schere, 2006). CMAQ is run for
user-defined input sets of meteorological conditions and emissions. For this analysis, we are
using predictions from several existing CMAQ runs. These runs include annual simulations for
2002 using CMAQv4.6 and annual simulations for each of the years 2002 through 2005 using
CMAQv4.7 (Foley et al., 2010). CMAQv4.6 was released by the U.S. Environmental Protection
Agency's (EPA's) Office of Research and Development (ORD) in October 2007. CMAQv4.7
along with an updated version of CMAQ's meteorological preprocessor (MCIPv3.4, Otte and
Pleim, 2010)1 were released in October 20082. The 2002 simulation with CMAQv4.6 was
performed for both the Eastern and Western domains. The horizontal spatial resolution of the
CMAQ grid cells in these domains is 12 x 12 km. The 2002 through 2005 simulations with
CMAQv4.7 were performed for the eastern 12-km domain and for the continental United States
1 The scientific updates in CMAQ v4.7 and MCIP v3.4 can be found at the following web links:
http://www.cmascenter.org/help/model_docs/cmaq/4.7/RELEASE_NOTES.txt
http://www.cmascenter.org/help/model_docs/mcip/3.4/ReleaseNotes
2 The differences in nitrogen and sulfur deposition in the case study areas between CMAQ v4.6 and v4.7 for 2002
are small, as described in Chapter 3.
F-ll
-------
domain, which has a grid resolution of 36 x 36 km. The CMAQv4.6 and v4.7 annual simulations
feature year-specific meteorology, as well as year-specific emissions inventories for key source
sectors, such as utilities, on-road vehicles, nonroad vehicles, wild fires, and natural biogenic
sources. Emissions for other sectors of the inventory for each of the years modeled rely on
inventories for 2002. Details on the development of emissions, meteorology, and other inputs to
the 2002 CMAQv4.6 runs can be found in a separate report (U.S. EPA, 2008). Inputs for the
CMAQv4.7 runs for 2002 through 2005 were derived using procedures similar to those for the
CMAQv4.6 2002 runs.
Additional details of the modeling domain, emissions and meteorological inputs are
provided in EPA (2009; REA Appendices).
F.4.2.2 CMAQ Evaluation, Sensitivity and Variability Analyses
Past results A variety of comparisons of modeled estimates to observations were
included in the REA (EPA, 2009), and some of the highlights are summarized here in addition to
new work on ammonia characterization and wet deposition. Readers are encouraged to review
the earlier report. Ambient air concentrations and wet deposition observations are paired against
modeled estimates. In contrast, dry deposition is always a modeled value, either derived from
ambient or modeled ambient concentrations. Given the interest in relevant nitrogen and sulfur
species, CASTNET observations were used extensively. Comparisons of modeled annual
average total nitrate (sum of nitric acid and particulate nitrate), ammonium, sulfate, and sulfur
dioxide to observations for the 2002 base year are provided in Figures F-l through F-4.
Normalized mean bias statistics for 2002-2005 base years are provided in Table F-2.
CMAQ overpredicts 862 and underpredicts SO/t. Although model performance is good
for total SOx, the inclusion of co-located 862 and sulfate measurements required for future
secondary NOX/SOX NAAQS comparisons will help diagnose issues with the model's ability to
partition these two species. CMAQ generally overpredicts total nitrate and slightly underpredicts
ammonium and the model captures the monthly temporal patterns of sulfate, total nitrate and
ammonium when all sites are aggregated (Figures F-5 to F-7). There are some basic
incommensurabilities between model estimates and observations that complicate interpretation
of model to observation comparisons, most notably the representation of space as a model
represents a volume average of roughly 144 km2, which depends on the time varying vertical
F-12
-------
depth of the lowest modeled layer. Most surface based observations rely on point sampling and
the extent to which a point is representative of broader volume space varies with meteorology,
distribution of emissions and surface characteristics.
Table F-2. Normalized Mean Bias Statistics for Predicted and Observed Pollutant
Concentration.
Pollutant
Concentrations
SO2
SO42"
TNC-3
NH4+
2002
45%
-13%
22%
4%
2003
39%
-9%
26%
11%
2004
47%
-13%
22%
7%
2005
41%
-17%
24%
2%
F-13
-------
2002ac met2v33 12km ESO2 for 20020101 to 20021231
o _
1 -
o
CD -
n CASTNet (2002ac met2v33 12kmE) /
n /
(ug/m3) (%) /
IA = 0.92 NMB = 33.6 /
RMSE = 1.91 NME = 35.6 /
RMSEs = 1.38 NMdnB = 29.6 / D
RMSEu = 1.32 NMdnE = 30.2 / D
MB = 1.31 FB = 30.9 J
ME = 1.39 FE = 32.6 / /
MdnB =1.04 / /
MdnE =1.07 / /
'/' /
m / /
D A / D /
Q/ D / 2 predicted concentrations versus
observations at CASTNet sites in the eastern domain (note, units are in actual mass for
including oxygen).
F-14
-------
2002ac met2v33 12km ES04 for 20020101 to 20021231
o
in
in
o
IT)
10
in
CO
q
CO
in
c\i
q
c\i
in
in
6
n CASTNet (2002ac_met2v33_12kmE)
(ug/m3)
IA
RMSE =
RMSEs =
RMSEu =
MB
ME
0.96
0.51
0.44
0.26
-0.44
0.44
NMB =
NME =
NMdnB =
NMdnE =
FB =
FE =
MdnB = -0.41
MdnE = 0.41
Period Average
SO4 ( ug/m3)
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.J
Figure F-2. 2002 CMAQv4.6 annual average SO2"4 predicted concentrations versus
observations at CASTNet sites in the eastern domain (note, units are in actual mass for 864,
including oxygen).
F-15
-------
2002ac met2v33 12kmETNO3 for 20020101 to 20021231
n CASTNet(2002ac_met2v33_12kmE)
(ug/m3)
IA
RMSE =
RMSEs =
RMSEu =
MB
ME
MdnB =
MdnE =
NMB
NME
NMdnB =
NMdnE =
FB
FE
Period Average
TNO3 ug/m3 )
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0
Observation
Figure F-3. 2002 CMAQv4.6 annual average TNOs predicted concentrations versus
observations at CASTNet sites in the eastern domain (note, units are in actual mass for
including oxygen).
F-16
-------
I
o
2002ac met2v33 12kmE NH4 for 20020101 to 20021231
un
C\j
q
c\i
un
ci
q
ci
n CASTNet (2002ac_met2v33_12kmE)
IA
RMSE =
RMSEs =
RMSEu =
MB
ME
NMB
NME
NMdnB =
NMdnE =
FB
FE
MdnB = -0.03
MdnE = 0.17
Period Average
NH4 ( ug/m3
0.0
0.5
1.0
1.5
2.0
2.E
Figure F-4. 2002 CMAQv4.6 annual average NH4+ predicted concentrations versus
observations at CASTNet sites in the eastern domain (note, units are in actual mass for NH4,
including hydrogen).
F-17
-------
CDC PHASE
7.0 -
6.0 -
5.5 -
5.0 -
-. 4.5 -
Xl
E
|u
n 3 5 "
3.0 -
2.5 -
2.0 -
1.5 -
1.0 -
0.5 -
RUNS CASTNET SO4 for 20020101 to 20051231; Sta
-Ar- CASTNET
-*- CMAQ
Az
If
AX
J/
X
X
A
&
A ^
/ \
/
x
V
H
V
\
2002
i i i i i i
1 3 5
7 9
A
i
I
I
n
A /
> x
F
A /
714
A' x
f/X
/
V
y
•
2003
I 1 I I I 1 I
11 1 3 5 7
<
A
|
1
^
A
'
\
i i i i
A'X\
//\A
r\\
/x \\
^ *s
/ / 11
/ /x \\
// !
/X \
n I
r \ /
1 »
/
2004
1 I 1 I 1 1 1 I 1 1 1
A'
/,:
y
1
9 11 1 3 5 7 9 11 1
2-
Figure F-5. 2002-2005 Domain-wide average SC>4 " predicted concentrations and
observations by month at CASTNet Sites in the eastern domain(note, units are in actual mass for
864, including oxygen) .
F-18
-------
CDC_PHASE_RUNS CASTNET TNO3 for 20020101 to 20051231 ; Sti
5.5 -
4.0 -
n
E
B) 3.5 -
3
*•*
B
Z 3.0 -
2.5 -
2.0 -
-±- CASTNET
-x- CMAQ
Vi /\
* y \ / \
Ax \ 1 i
\ \ X X\,
\\ if
W 2
W A /
A \ A /
\ /
2002
i i i i i i i i i i i i i
1 3 5 7 9 11 '
X
I
XA
/ A
X
fi
\
x
1
'1 l'i X
A
in / \
X I / k< / \
6 Mil x x>-
H ] xf 1\ 7— J
\J
2003
3 5 7 9 11 1
\x x
4 / /
1 / A
V i /
f\x X ,A
A A
'A./
A
2004
3 5 7 9 11 1
X
I
// '
//
f
k
1
Figure F-6. 2002-2005 Domain-wide average TNOs predicted concentrations and
observations by month at CASTNet sites in the eastern domain (note, units are in actual mass for
s, including oxygen).
F-19
-------
CDC_PHASE_RUNS CASTNET NH4 for 20020101 to 20051231; Sta
2.0 -
CASTNET
CMAQ
X fl X \ X
I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
3579: i 3 5 7 9 3579 '
0.6 -
Figure F-7. 2002-2005 Domain-wide average NH4+ predicted concentrations and observations
by month at CASTNet sites in the eastern domain (note, units are in actual mass for
including hydrogen).
F-20
-------
Kenansville Ammonia July 2004
12-hour Averages: 6am-6pm
184 186 188 190 192 194 196 198 200 202 204 206 208 210 212 214
Julian Day: TickMark at Midnight (July 2004)
-Observations
Kenansville Ammonia August 2004
12 hour averages: 6am-6pm
214 216 218 220 222 224 226 228 230 232 234 236 238 240 242 244
Julian Day: TickMark at Midnight (August 2004)
-Observations
Millbrook (Raleigh) Ammonia July 2004
12-hour Averages: 6am-6pm
183.5 185.5 187.5 189.5 191.5 193.5 195.5 197.5 199.5 201.5 203.5 205.5 207.5 209.5 211.5 213.5
Julian Day: TickMark at Midnight (July 2004)
Millbrook (Raleigh) Ammonia August 2004
12 hour averages: 6am-6pm
214 216 218 220 222 224 226 228 230 232 234 236 238 240 242 244
Julian Day: TickMark at Midnight (August 2004)
-Observations
Figure F-8. Comparison of CMAQ predictions and measurements for 12-hour (6am-6pm)
average NH3 concentrations, with a monitoring cycle of 4 days on and 4days off, at a high
emission site (Kenansville) and a low emission urban site (Raleigh) in North Carolina compared
to CMAQ for July 2004 (top) and August 2004 (bottom), from Dennis et al, 2010 (note, units
are in actual mass for NHa, including hydrogen).
F-21
-------
Comparison with SEARCH Data
The SEARCH network (chapter 2) includes sites instrumented with continuously
operating SO2 monitors. Comparison of CMAQ estimated SO2 and SERACH data showed
similar over-predictions to CASTNET data sets. The SEARCH data is affords the ability to
diagnose the diurnal aspects of SO2 patterns, which is not possible with the weekly averaged
CASTNET data. Example model to observation plots follow for the SEARCH sites reflecting
regional air quality, located away from major sources.
F-22
-------
OLF : Hourly SO2 (ppb), 2006
observed
modeled
,F • °tF
012345678
10 II 12 13 14 15 16 17 18 19 20 21 22 23
hour (LSI)
Figure F-9. 2006 hour by hour comparisons of CMAQ and SEARCH SO2 data at the Ook Grove, MS site (source,
K. Foley, U.S. EPA-ORD)
CTR : Hourly SO2 (ppb), 2006
a
3
observed
modeled
01234567
9 10 II 12 13 14 15 16 17 18 19 20 21 22 23
hour (LSI)
Figure F-10. 2006 hour by hour comparisons of CMAQ and CSEARCH SO2 data at the Centerville, AL site
(source, K. Foley, U.S. EPA-ORD)
F-23
-------
YRK : Hourly SO2 (ppb), 2006
• observed
• modeled
8 0
12 13 14 15 16 17 18 19 20 21 22 23
hour (LSI)
Figure F- 11. 2006 hour by hour comparisons of CMAQ and CSEARCH SO2 data at the Yorkville, GA site
(source, K. Foley, U.S. EPA-ORD)
Wet deposition.
Modeled wet deposition in CMAQ is a function of the volume of predicted precipitation
within a grid cell and the pollutant concentrations scavenged from the atmosphere during
precipitation events. As a result, errors in modeled precipitation and in emission inputs can lead
to significant bias and error in the wet deposition predictions compared to observed values. EPA
(Dennis and Foley, 2010) has corrected CMAQ wet deposition predictions by scaling the model
output based on observation-based gridded precipitation data generated by the Parameter-
elevation Regressions on Independent Slopes Model (PRISM, 2004). The precipitation adjusted
deposition fields are more highly correlated with observed values for all wet deposited nitrogen
and sulfur species compared to the base model output (Figures F-12, F-13). In addition, the
adjusted fields are better able to capture the spatial heterogeneity of accumulated wet deposition
due to orographic effects on precipitation amounts.
Adjusting the wet deposition values to account for over-predictions in the model
precipitation inputs revealed compensating errors for nitrate and ammonium. The negative bias
seen in these species after the precipitation adjustment is believed to be due to missing emissions
F-24
-------
?
sources. A second bias adjustment was performed for nitrate and ammonium based on observed
levels at the NADP/NTN sites (Figure F-13). The final adjusted spatial fields of annual total wet
deposition values are more consistent with observed wet deposition values. Ongoing studies
suggest that much of this bias can be reduced in the Eastern half of the US by including nitrogen
oxide produced by lightning and accounting for the bi-directional flux of ammonia. Once these
model improvements are incorporated in CMAQ a second bias adjustment may not be needed in
the East.
In 2011, the EPA will deploy a prototype flux measurement package over a grass field at
the Duke Forest Blackwood Division. Dry deposition fluxes will include NOy, NO2, FINOS,
HONO, NO3 aerosol, NH3, NH4 aerosol, SO2, SO4 aerosol and ozone. Concentrations of
organic N in aerosol, but not fluxes, also will be measured. The availability of these flux
measurements will enable a more direct assessment of CMAQ treatment of sulfur and nitrogen
deposition processes.
MB = 2.4 kg/ha RMSE = 5.1 kg/ha
••**. •*•
MB = 0.7 kg/ha
N MB = 6 %
RMSE = 2.9 kg/ha
NME = 18 %
RMSEs = 0.8 kg/ha
RMSEu = 2.8 kg/ha j 51 % decrease }
observed SO4 wet deposition (kg/ha)
observed SO4 we! deposition (kg/ha)
Figure F-12. Unadjusted (left) and PRISM (right) adjusted CMAQ annual wet deposited
sulfate for 2002 (note, units are in actual mass for SO4, including oxygen).
F-25
-------
MB = -0.6 kg/ha
NMB = -6 %
RMSEs = 0.7 kg/ha
RMSEu = 2.8 kg/ha
RMSE = 2.9 kg/ha
NME a 21 %
MB = 0.1 kg/ha
NMB = 1 %
RMSE = 2.4 kgrlu
NME = 17 %
RMSEs = 0.1 kg/ha
RMSEu = 2.4 kg.'ha (17 % decrease )
observed NO3 wet deposition (kg/ha)
MS = -0,1 Kg/ha RMSE = 0.8 kgfha
NMB = -2% NME = 24%
RMSEs = 0.2 kgfha
RMSEu = 0.8 kg/ha
observed NO3 wel deposition (kg''h
MB = 0.1 kgy
NMB = 3 %
RMSE = 0.7 kg/ha
NME = 20%
RMSEs = 0.2 kg'ha
RMSEu = 0.7 kg/hia (17% decrease )
•t .'
'
observed NHd .vei deposition ikg'hai
NH4 wet deposition (
Figure F-13. Unadjusted (left) and PRISM and bias (right) adjusted CMAQ annual wet
deposition of nitrate (top) and ammonium (bottom) (note, units are in actual mass for MLt and
, including hydrogen).
F-26
-------
Ammonia.
The role of NHX deposition is incorporated in the AAPI expression as a parameter that
influences the level of allowable concentrations of NOy and SOX, due to its role as part of the
total reactive nitrogen budget which affects acidification. Characterizing ammonia deposition is
challenging due to the variety of surface and vegetation types that influence ammonia dry
deposition velocities as well the potential for bi-directional flux of ammonia. In addition,
ammonia emission estimates remain relatively more uncertain than emissions of NOX and 862
given the complexity of meteorology and agricultural practices that influence the spatial and
temporal patterns of ammonia releases. An exploration of the sensitivity of ammonia to three
different treatments of deposition processes in CMAQ was performed by EPA (Dennis et al,
2010) to test the inclusion of a bi-directional NHa flux algorithm and elucidate the relative
importance associated with advection, deposition and chemical transformation on ammonia
patterns. These treatments included a (1) base case of current CMAQ treatment using existing
ammonia deposition velocity schemes and uni-directional deposition, (2) modified the base case
by replacing ammonia deposition velocity calculations with 862 deposition velocities (862
interacts with surfaces and vegetation similarly to NH3, but with reduced velocity) as a lower
bound and (3) introducing a bi-directional flux algorithm to the base case (retaining NH3
deposition velocities). Based on modeled process analysis that delineates the effects of
deposition, chemical transformation and advection (horizontal and vertical) on emitted ammonia,
the results (Figure F-14) suggest that ammonia patterns, especially when a bi-directional flux
process is incorporated, are more indicative of a transported pollutant where emissions influence
can span hundreds of kilometers, markedly different from some earlier perspectives where
ammonia often was thought of as near source phenomenon due to high deposition velocities.
The process analysis illustrates the importance of vertical advection which enables the movement
of ammonia into traditional mesoscale flow patterns. The effect is enhanced by the
reintroduction of deposited ammonia through bi-directional flux into the ambient environment.
From a monitoring perspective, a design that addresses the regional characterization of
NOy and SOX would be consistent with characterizing NHX. Not only would ammonia and
ammonium measurements be useful for estimating dry deposition through deposition modeling
approaches such as those used in CASTNET, but they would serve as important diagnostic data
F-27
-------
to continually assess the effectiveness of NHX deposition processes in models like CMAQ. This
is especially important as we recognize a large uncertainty in the bi-directional formulation
associated with the estimation of F, the emissions potential due to the existence of compensation
points. Nonetheless, we can learn much about the NH3 budget in spite of these uncertainties.
High priority research is ongoing to improve the bi-directional parameterization and the
estimates of the leaf and soil gammas across different cropping regions and throughout the year.
We are developing a software tool to estimate the soil F associated with fertilizer application.
When we have a spatially and temporally varying Fg, we will investigate the emissions budgets
for fertilized fields, as well as reexamine the animal operation emission budgets, as this will be
of interest. Work to examine the seasonality of single cell budgets and their range of influence is
continuing. Current and future CMAQ applications to ecosystem deposition will incorporate bi-
directional flux treatment of ammonia.
F-28
-------
Base
— Advection
— Dry Deposition
Wet Deposition
VdSO2 Bi-directional
Advection -o— Advection
Dry Deposition -o—Dry Deposition
Wet Deposition -o—Wet Deposition
LU
M—
O
c
o
'o
0.8-3
~ 0.6-
0.4-
0.2-
ooooooooooooooooooooooooooo
100
200 300
Distance (km)
400
500
Figure F-14. Cumulative regional NH3 budget of advection, wet- and dry deposition,
calculated for an expanding box starting at the high-emitting Sampson County NC cell (from
Dennis et al, 2010)
F.4.2.3 Variability and sensitivity of CMAQ generated components.
Ambient Concentration to Deposition Transformation ratios.
Derivation
Atmospheric pollutants deposit onto land and water surfaces through at least two major
mechanisms: direct contact with the surface (dry deposition), and transfer into liquid
F-29
-------
precipitation (wet deposition). The magnitude of each deposition process is related to the
ambient concentration through the time-, location-, process- and species-specific deposition
velocity (Seinfeld and Pandis, 1998):
DePiDry = vtDry • CtAmb (1)
t f~, Wet /O\
•Ci (2)
where v, ry and v, e are the dry and wet deposition velocities, Depf ry and Depf e are the dry and
wet deposition fluxes, Cf is the ambient concentration, and the /' subscript indicates the
pollutant species under study. The total deposition of each pollutant is
T-X Tot T-N Dry T-N Wet /o \
Dep, = Dept v +Dept (3)
Substituting Equations 1 and 2 into Equation 3 yields
T-. Tot Dry /-, Amb Wet /-, Amb / A\
DePi =vi'7 -Ct + v; •Ci (4)
The total deposition of sulfur or nitrogen would therefore be:
j-x Tot X ^ / Dry Wet \ /-» Amb .---.
DePs\N = 2^(V, +v, )-™,-C, (5)
i
where m is the molar ratio of the atom (sulfur or nitrogen) of interest to the /'th pollutant.
Ambient sulfur- and nitrogen-containing pollutants include gases such as sulfur dioxide (SO2),
ammonia (NH3), various nitrogen oxides (NO, NO2, HONO, N2O5), nitric acid (HNO3), and
organic nitrates such as peroxyacetyl nitrates (PAN); as well as particulate species such as sulfate
(SC>42"), nitrate (N(V), and ammonium (NH4+). The species regulated by the SOx/NOx standard
will include the sulfur-containing species above and the above oxidized forms of nitrogen
(NOy); ammonia and ammonium are not currently included as regulated pollutants.
Aggregation Issues
Equation 5 provides a relationship for converting a sulfur or nitrogen deposition to an
"equivalent" airborne concentration. This is useful for setting the NAAQS, where we expect to
convert deposition critical loads developed by the ecosystem models to ambient concentrations,
which will then be the air quality standard. A major issue to consider during such conversion is
the spatial, temporal and chemical resolutions of the deposition data and the resulting standards.
Since the objective is to regulate total oxidized sulfur and nitrogen, and this is also the chemical
F-30
-------
resolution provided by the ecosystem models, it is convenient to collect the deposition velocities
and apply the mt values to the C,'s in Equation 5 and re-arrange it as
CTot Tr j~\ Amb //-\
SIAT =VSIN-DepSIN (6)
where VS/N is a constant that relates total deposition of sulfur or nitrogen to the total ambient
concentration. Since the deposition critical loads are annual total depositions, it is proposed that
the standard be an annual average concentration. Data used to derive annual VS/N values will
need to have the same spatial representativeness as the critical loads.
Air Quality Simulation Models
Ideally, VS/N values would be derived for each area of interest from concurrently collected
sulfur and nitrogen deposition and concentration measurements. However, no monitoring
network currently exists that can provide such information. We therefore propose using output
of the Community Multi-scale Air Quality (CMAQ) model (EPA, 1999) for initial calculation of
VS/N values.
CMAQ provides both concentrations and depositions of a large suite of pollutant species
on an hourly basis for 12 km grids across the continental U.S. Its comprehensive structure is
ideal for providing VS/N values that appropriately address the chemical and temporal aggregation
issues discussed above, and weighted spatial averages of the gridded data can be used for areas
that span multiple grid cells. The major potential drawback to using CMAQ output is that the
data is simulated rather than measured, which calls its accuracy into question (discussed further
below).
CMAQ does not directly calculate or use VS/N values; instead the following procedures
are used in the code to model deposition:
1) vdry values of gaseous pollutants are calculated in the CMAQ weather module called
the Meteorology-Chemistry Interface Processor (MCIP) through a complex function of
meteorological parameters (e.g. temperature, relative humidity) and properties of the geographic
surface (e.g. leaf area index, surface wetness).
2) vdry values for paniculate pollutants are calculated in the aerosol module of CMAQ,
which, in addition to the parameters needed for the gaseous calculations, also accounts for
properties of the aerosol size distribution.
3) vwet values are not explicitly calculated. Wet deposition is derived from the cloud
processing module of CMAQ, which performs simulations of mass transfer into cloud droplets
F-31
-------
and aqueous chemistry to incorporate pollutants into rainwater, all of which is conceptually
contained in the vwet parameter in Equation 2.
The deposition transference ratios^ introduced in Chapter 5 are referenced as TSOX and
TNoy, to distinguish these parameters from an exact linkage to deposition velocity, which is
uniquely associated with individual atmospheric species. Deposition transference ratios are
defined as the annual wet and dry deposition of all oxidized species (NOy for TNoy, 862 plus 864
for TSOX) divided by the average annual concentration of NOy, for TNoy, or 864 plus 862, for
TSOX. The units for TNoy and TSox are distance/time. Deposition transference ratios provide a
mechanism to associate ambient concentrations to deposition loads and to determine if an area's
air concentrations of NOy and SOX meet a NAAQS level using the AAI form. A deposition
transformation ratio is an aggregate representation of the deposition process generated through
modeling which does not lend itself to a traditional analysis relating observations and
predictions. Furthermore, there is an implicit assumption that the response of deposition
transformation ratios to changes in meteorology and emissions is relatively stiff, as these ratios
are an attribute of the system that channels ambient air response associated with decreases in
emissions of NOX and SOX to changes in deposition. The stiffness of the deposition transference
ratios would suggest that the relationship between ambient concentrations and deposition is
strictly a constant proportion, not impacted by the mixture and level of emissions or by changes
in meteorology. To better understand the implications of this assumption, we investigated the
relative variability of the modeled deposition transformation ratios across time and space, and the
stability of the ratios relative to emissions and meteorological inputs was conducted to guide
EPA in determining how uncertainties in this parameter may eventually impact AAI related
calculations.
Spatial and Interannual Variation of TS and TN.
Generally small spatial and inter-annual variability exist in the deposition transformation
ratios for the 2002 -2005 model years (Figure F-15). The inter-annual variability, calculated at
the grid cell level, as measured by the median coefficient of variation is around 10% and the
absolute values of the ratios remain stable, suggesting that year to year changes in meteorology
3 In the first draft of the Policy Assessment, the deposition transformation ratios were labeled VNOx and VSOx- For
this draft, based on recommendations from CASAC, we have renamed these ratios TNOx and TSOx.
F-32
-------
have minimal impact on the ratios. Spatial homogeneity of deposition transformation ratios
within the two acid sensitive areas we evaluated in the REA (Adirondacks and Shenendoah)
(Figure F-15) is consistent with a relatively homogeneous ambient concentration environment
overlaid upon a landscape of similar vegetation and surface conditions. Such spatial
homogeneity within case study areas provides confidence that an area wide application AAI will
not be strongly dependent on the exact boundaries chosen to define an acid sensitive area.
TSOX and TNoy Sensitivity to emission changes.
The response of TSOX and TNOY to emission changes was explored by analyzing available
base case 2005 and a CMAQ simulation emissions reduction scenario referred as the 2030 case.
The 2030 case represents Eastern U.S. domain wide NOx and SOx emission reductions of 48%
and 40%, respectively. Median changes in deposition transference ratios tended to be around
zero (Figure F-16), with the Adirondack region exhibiting slightly higher response than the
Shenandoah region and remainder of the Eastern U.S. domain.
F-33
-------
Conc:Dep Raliqi in ADR aullirw
»
, .
--
n
--
.• -
* *
* *
i ! i "
; " ! ; !
it * >
i it
i i i
III i
1 1 1 1 i i 1 1 i ..
2002 £003 2004 2005 XKff, 2030cp 2002 2003 2004 2003
Conc:D*p Ralifls in Sh.n Lake
• # - - ,
1 1
B . « t * •
3 " t * «
t . 1 1 1 . I
1 ! '
tti • ! t
t i
j i i i
j i j I I
..........
•
f ,
i
! i
..
2005qi 2030cp
-
;
-
1
1
i
m±
Sulfur
Nitrogen
Sulfur
Nitrogen
2002 2003 2004 2003
2002 2003 2004 2003 2MSep 2030cp
Figure F-15. Spatial and interannual variability of inverse deposition transference ratios, 1/Tsox
and I/TNOY, for Adirondack (top) and Shenandoah case study areas.
F-34
-------
Inter-year CV of Sulfur Conc:Dep Ratio
Inter-year CV of Nitrogen Conc:Dep Ratio
>
o
o
to
o
I
ADR
Other Shen
ADR
Other
Shen
en
c
CO
£
O
o
c
O
o
-------
Comparisons of CMAQ and CASTNET and NADP generated TSOx and TNOy
Transference ratios based on CASTNET sulfur and nitrogen measurements, CASTNET
derived dry deposition and NADP wet deposition were constructed to approximate ratios
developed using measurements where possible. These ratios were developed to compare with
CMAQ derived ratio. The limitations of the ratios constructed from observations include
availability of only part of the NOy species mix, total nitrate, as well as the use of a model to
calculate dry deposition velocities. Nevertheless, any comparison to measurements not only can
provide some assurance that CMAQ constructed ratios relate to observed data as well as serving
as a potential diagnostic metric for modeled deposition processes. Modeled and observed ratios
(Figure F-17) indicate closer agreement for TSOX relative to TNOY, an expected result given the
missing nitrogen species in the CASTNET data set. It is difficult to diagnose the cause of
differences as the dry deposition estimates from CASTNET are based on different modeling
approach relative to that used in CMAQ. However, as observation data sets become more
complete, these types of analyses can help elucidate what aspects of deposition characterization
require improvement. The next logical steps in developing comparisons between observed and
modeled ratios will be to replace CMAQ wet deposition with PRISM based wet deposition
results to allow for a better spatial pairing of observed and modeled data sets. This would
partially constrain the problem to differences in dry deposition calculation methods between
CASTNET and CMAQ. EPA's ORD will be conducting sulfur and nitrogen dry deposition flux
experiments in 2011. Those data will provide be extremely valuable diagnostic information to
improve CMAQ deposition processes, which in turn will lead to improved confidence in
CMAQ derived deposition ratios.
F-36
-------
2005 Sulfur T ratios
2005 Oxidized Nitrogen T ratios
o
2
Castnet
o o
o o
0 O
o
10
Castnet
15
Figure F-17. Comparison of observation based and CMAQ derived transference ratios. Each
point represents an annual average at the grid cell of the CASNET station. The Observed values
used NADP wet precipitation data, CASTNET air quality data and modeled dry deposition based
on the CASTNET observations and the dry deposition model specific to CASTNET.
F-37
-------
Comparisons between models
The variability in T ratios of SOx and NOx was explored by comparing their values from
a number of different modeling scenarios. Model runs were procured for the purpose of
analyzing variability from the following sources:
The variability in T ratios due to each of these sources is shown through boxplots specific
to pollutants (i.e. SOx and NOx) and selected study areas (the Adirondacks, Shenandoahs) in
Figures F-18 to F-21. The data in the boxplots are the T values of all grid cells whose center lies
within the specified study area. The center line of each box are the population medians, the box
edges are the 25th and 75th percentiles, the ends of the whiskers are the 10th and 90th
percentiles, and the points are data that lie outside the 10th and 90th percentiles. The notches in
the boxes indicate the significance differences in the medians: non-overlapping notches strongly
suggest that the populations are significantly different. The model scenario that produced each
box's data is indicated on the x-axis.
The boxplots show that, in general, there is not a large degree of inter-annual variability
in T ratios. The notches of T ratios for nitrogen appear to overlap for all 4 years of CDC 12km
and 36 km CMAQ runs in both study areas. The S values appear to be slightly high in the
Shenandoahs in 2003 and lower in 2005 in the 12 km run. The 36 km result show what appears
to be a decline in S values in both areas, especially when comparing the first 2 years to the latter
2 years. Results from the AURAMS 2005 runs appear to generally be similar to those of the
2005 CMAQ runs, which is surprising given the different mix of chemical species available for
calculations and the much larger (45 km) grid size of the AURAMS model.
F-38
-------
Figure F-18. Modeled generated TNOy ratios in the Shenadoah area showing CMAQ 12 and
36 km values 2002 - 2005 and AURAMS 2005.
Figure F-19. Modeled generated TSOx ratios in the Shenadoah area showing CMAQ 12 and 36 km values 2002
2005 and AURAMS 2005.
F-39
-------
J- -f-
I i
T
a
4-
Figure F-20. Modeled generated TNOy ratios in the Adirondacks area showing CMAQ 12 and 36 km values 2002
2005 and AURAMS 2005.
Figure F-21. Modeled generated TSOx ratios in the Adirondacks area showing CMAQ 12 and 36
km values 2002 - 2005 and AURAMS 2005.
F-40
-------
CMAQ uncertainties and the AAI
The AAI relies on CMAQ for the sulfur and nitrogen transference ratios and NHX
deposition. The model evaluation results, including the ammonia and wet precipitation
treatments, reflect a continual process of model improvement designed to ingest the latest
science within a framework links a myriad of atmospheric and surface processes across multiple
pollutant species. While this document focuses in on the more direct processes affecting N and S
deposition, these modifications are incorporated with the philosophy that the best science is
being adopted and they in turn support the overall improvement of the models' treatment of all
processes. The inclusion of better chemistry and physics of a particular process acts as an
internal diagnostic tool for other processes that are linked throughout the model framework
through basic conservation of mass principles. With respect to the AAI, the CMAQ model must
be relied on to provide the spatial flexibility attendant with a national standard. As the model
continually adopts the best science, confidence in relevant CMAQ generated AAI parameters is
raised for both near term and future scenarios.
F.4.3 Ecosystem Modeling
F.4.3.1 MAGIC modeling
An extensive uncertainty analysis of the MAGIC model was conducted as part of the
REA, and documented in Appendix 4 of the REF. This uncertainty analysis included
comparison of MAGIC outputs with observed water chemistry and ANC values. The uncertainty
analysis also included an approach for generating confidence intervals for predicted ANC, using
ensembles of model results based on alternative model calibration methods.
The model performance comparisons documented in Appendix 4 of the REA show close
correspondence between simulated and observed annual average surface water SO/t, NOs, and
ANC during the model calibration period for 44 lakes in the Adirondacks Case Study Area and
60 streams in the Shenendoah Case Study Area. These comparisons are reproduced in Figures F-
22 and F-23. Comparisons in the ability of MAGIC to reproduce the temporal pattern of ANC
for individual lakes was also assessed, and the model does reasonably well at matching the
pattern of ANC, although the fit is not as good as during the model calibration period.
The estimated confidence bounds on predicted ANC suggest that the 95 percent upper
confidence bound is on average 10 percent higher in lakes and 5 percent higher in streams. This
suggests relatively low uncertainty introduced by the MAGIC modeling assumptions. MAGIC
F-41
-------
modeling is used in developing the estimates of base cation weathering for comparison to the F-
factor approach described in Chapter 5.
F.4.3.2 SSWC modeling
As stated in Appendix 4 of the REA, uncertainties in some elements of the SSWC
modeling are not well understood. The version of the SSWC model used here uses the F-factor
approach to estimate the preindustrial base cation supply for a given catchment. While this
approach has been widely applied in Canada and Europe, it has only been used in a few cases
within the United States and its assumptions and parameters have not been fully evaluated for
aquatic systems. The natural or preindustrial catchment supply of base cations (i.e. weathering
rates) has the most influence on the critical load calculation and also has the largest uncertainty
(Li and McNulty, 2007). The uncertainty and ability to accurately estimate this parameter has
not fully been evaluated and its uncertainty is unknown. It is important to note that for the
United States, there is only one study for surface waters critical loads that compared steady-state
and dynamic models and different steady-state approaches (MAGIC and F-factor) (Holdren et al.
1992) other than what is presented in Chapter 5. Holdren et al. (1992) compared critical loads
calculated by the steady-state MAGIC and the SSWC F-factor model for lakes in the Northeast.
In this study, steady-state MAGIC model yielded critical load values that show the same general
trend and on average were 14 kg/(ha-yr) SC>4 higher than those from the SSWC F-factor
approach, which is consistent with results, presented in Chapter 5. The two models converge at
low critical, but diverge as the buffering potential for watersheds increase, as indicated by
increasing critical loads.
The REA conducted an uncertainty assessment using Monte Carlo simulation methods to
characterize the uncertainty in estimated critical loads using the SSWC, varying a number of
important inputs including runoff rates, water chemistry variables, and acid deposition. The
coefficients of variation (CV) for the estimated critical loads (standard deviation divided by the
mean) were calculated for each lake in the study as a measure of relative uncertainty.
The results of this uncertainty analysis show that the coefficients of variation are on
average very low for target ANC values within the range we are recommending (20 to 50 |ieq/L).
The CVs for critical loads are only 5% and 9% for critical load limits of 20 and 50 ueq/L,
respectively. Although the average CV is relatively small for the population of sites modeled,
F-42
-------
individual site CV can vary from 1% to 45%. This difference is due to the high degree of
uncertainty in site specific parameters for particular sites.
These analyses suggest that uncertainties introduced in the AAI directly by the SSWC
Factor model are likely to be moderate. Additional uncertainties are introduced by the
generalization of the F-factor approach to estimate critical loads in locations where F-factors
have not been developed.
150
g-
I 100-1
£
V)
SO4 fjeq/L
1:1
50 100
Observed Chemistry
150
5 10 15
Obser\*d Chemistry
20
300
£• 200 -
E
to
100 -
0 -
-100
ANC fjeq/L
k
l -
-100 0 100 200
Observed Chemistry
300
pH |jeq/L
1:1
567
Observed Chemistry
Figure F-22. Simulated versus observed annual average surface water SO42-, NO3-, ANC, and
pH during the model calibration period for each of the 44 lakes in the Adirondacks Case Study
Area. The black line is the 1:1 line. (Source: reproduced from REA, Appendix 4, Figure 1.1-1).
F-43
-------
150
~ 100-
o
300
£ 200 -
(/}
1
OJ
0 100 -
_
1
-100
SO42" |Jeq/L
1:1
Observed Chemistry
ANC |jeq/L
-100 0 100 200
Observed Chemistry
300
7 -
6 -
E
•i;
5-
10 20 30 40
Observed Chemistry
50
567
Observed Chemistry
Figure F-23. Simulated versus observed annual average surface water SO42-, NO3-, ANC, and
pH during the model calibration period for each of the 60 streams in the Shenandoah Case Study
Area. The black line is the 1:1 line. (Source: reproduced from REA, Appendix 4, Figure 1.1-2) .
F-44
-------
F.4.4 Cumulative uncertainty analysis
An analysis of the cumulative effects of uncertainty on the AAI was conducted and is
described in Appendix G. In summary, this included bootstrapping analyses of the parameters in
the AAI equation to translate error in individual measurements to the regional values used in the
equation. The parameters that are averages of grid-level CMAQ modeled values, Ndep and NHx,
had bootstrapped uncertainty values of approximately +20% (Figures G-l and G-5). The
transference ratios include two different grid-level CMAQ modeled values and had much higher
uncertainty, exceeding 100% (Figures G-3 and G-4). The calculation ofNeco also includes two
different input values, the CMAQ derived Ndep values and lake specific nitrogen leaching
values. The Neco results also had high uncertainty values, ranging from -65% to approximately
200% (Figure G-2). The critical load value for the region is affected by the Q and BCo values at
the individual lakes within a region. Uncertainty in these parameters gave a regional uncertainty
range for the critical load of ±35% (Figure G-6).
The results of the bootstrapping analyses were used to complete a cumulative analysis of
uncertainty in a subsequent Monte Carlo style analysis. This analysis was illustrated in the form
of the tradeoff curve for the concentrations of NOy and SOx (Figure G-7). The results in the two
regions analyzed were similar. There was a range of uncertainty, with 50% of the distribution
within +20% of the observed value. Most importantly, the mean value of the results was very
close to the observed value in both regions. This indicates that there is no systematic bias in the
results despite what can be relatively high levels of uncertainty in the input parameters.
F-45
-------
F.5 Modeling and Data Gaps
Deposition processes. Currently, there are efforts to improve a number of nitrogen and
sulfur deposition processes in CMAQ. Active areas of model process improvement are in the
treatment of lightning generated NOx and the transference of nitrogen between atmospheric and
terrestrial and aquatic media, often referred to as bi-directional flux. Lightning NOx potentially
provides a significant contribution to wet deposition as the resulting NOx is rapidly entrained
into aqueous cloud processes. Both the thermodynamics of soil processes and mass transfer of
nitrogen species across the surface-atmosphere interface is governed by an assortment of
temperature, moisture, advection and concentration patterns. These processes and mass transfer
relationships are coupled within the emissions, meteorological, and chemical simulation
processes and associated surface/vegetation and terrain information incorporated in or accessed
by the CMAQ. In addition to research activities to improve the characterization of nitrogen-
related processes in CMAQ, efforts are also underway to improve the general characterization of
ammonia emissions which remains as an area of large uncertainty due to limited source data and
the ubiquitous nature of these emissions. Another challenge for regional/national air quality
modeling is properly representing the effects on pollutant concentrations, precipitation and
therefore deposition of variable terrain features, particularly steep mountain-valley gradients and
the interfaces to wide open basins encountered in the Western United States.
Two important enhancements regarding the treatment of wet precipitation and the bi-
directional flux of ammonia were discussed in Chapter 2. In 2011, the EPA will deploy a
prototype flux measurement package over a grass field at the Duke Forest Blackwood Division.
Dry deposition fluxes will include NOy, NO2, HNO3, HONO, NO3 aerosol, NH3, NH4 aerosol,
SO2, SO4 aerosol and ozone. Concentrations of organic N in aerosol, but not fluxes, also will be
measured. The availability of these flux measurements will enable a more direct assessment of
CMAQ treatment of sulfur and nitrogen deposition processes.
The interest in deposition of sulfur and nitrogen raises the potential importance of occult
(cloud and fog related processes) deposition associated with mists and clouds, which may be
particularly relevant for aquatic acidification of high elevation watersheds. Occult deposition
currently is in the early stages of development within the CMAQ framework.
Lightning generated NOx emissions have been an active area of research over the last
decade and approaches that incorporate lightning count data and estimated NOx generation based
F-46
-------
on satellite measurements and aircraft campaigns have been tested in modern air quality models,
including CMAQ. Lightning NOx is hypothesized to increase upper tropospheric ozone levels
and wet nitrogen precipitation, with relatively negligible impact on near surface ambient nitrogen
patterns. It is anticipated that CMAQ will incorporate lightning NOx for EPA assessments in the
2012timeframe.
Interest in organic bound nitrogen has increased based on NADP measurements
suggesting that organic nitrogen contributes as much as 30% of the total nitrogen in precipitation
samples. Significant uncertainties regarding the origin and composition of organic nitrogen
(Altieri et al, 2009) suggest research efforts to improve our understanding of organic nitrogen
prior to developing parameterizations in air quality models. Concentrations of organic -N will
be measured as part of EPA's flux studies in 2011. Questions regarding the relative contribution
of anthropogenic or natural sources as well as the effects of re-entrainment from the surface
require attention.
Atmospheric Observations. Chapter 2 addresses the current state of atmospheric
observations relative to the NOx/SOx secondary standard and includes suggestions for enriching
the observational data base used to evaluate models. This new standard poses measurement
resource challenges as the current networks, with the exception of CASTNET and some National
Park Service (NPS) efforts, there is sparse spatial coverage relevant to anticipated acid sensitive
areas and the specific measurements related to NOy, speciated NOy and ammonia and
ammonium. Temporally resolved data based on co-located measurements of continuously
operating 862 and sulfate would help diagnose the partitioning of these species and support
evaluation of SOx deposition. As discussed above, deposition flux studies should be considered
in a wide variety of ecoregions given the influence surface, vegetation type and meteorology
imparts on deposition velocities.
Source emissions. Anthropogenic emissions of nitrogen oxides (NO and NO2) and
sulfur dioxide generally are believed to be well characterized as the major contributors of NOx
and SO2 from energy generation and transportation sectors have a history of continuous
improvements of emissions modeling as well as direct emission measurements for major power
generating units. Greater uncertainty resides in natural emissions of NOx from lightning
processes (discussed above) and soil and agricultural related phenomena. Both NOx and
ammonia emissions are subject to re-emission after deposition as part of the complex cycling of
F-47
-------
nitrogen in soils and biota. Characterizing the variety of agricultural practices that impact both
ammonia and NOx is complicated by the dispersed nature of agriculture processes as well as the
influence of various meteorological factors on relevant biogeochemical processes controlling
transformation and removal of nitrogen species.
Ecosystem processes and soil and surface water observations. The critical load
modeling approaches underlying the standard require a variety of input data depending on the
approach chosen. In general terms, the availability of watershed related deposition, soil and
vegetation characteristics and surface water chemistry determine the approach taken. There is a
relatively extensive source of data for critical load modeling using the steady state model for
most of the contiguous U.S. However, several ecoregions included relatively small sample sizes
of critical load estimates and an increased pool of data to increase the representativeness of water
quality variables should be pursued at the national level.
Data to support dynamic modeling like MAGIC has focused on the Adirondack and
Shenandoah regions. A more thorough characterization of nitrogen retention, dissolved organic
carbon, soil chemistry in all acid sensitive areas would lead to reduced uncertainties in applying
the AAI as well as future considerations for standards that incorporate terrestrial acidification
and nutrient enrichment effects. In looking forward, it may not be practical to use dynamic
modeling everywhere, but periodic application of dynamic models in different ecoregions
gradually will result in an information system to revisit and evaluate many of the assumptions
applied in national scale steady state modeling. This extension to previously undersampled areas
is especially important in the mountainous West where there is evidence of aquatic acidification,
yet many of the atmospheric attributes are markedly different than in Eastern systems, starting
with a nitrogen dominated atmosphere, the strong influence of mountain-valley terrain on local
meteorology and the closer proximity to transport of hemispheric pollution across the Pacific
Ocean.
In addition to developing the data necessary to support existing models such as MAGIC,
efforts should be made to continue develop dynamic models to simulate effects of acidic
deposition on soil, drainage waters and biota, to test these models and to apply these as tools in
determining critical loads. This research effort should address key uncertainties identified in this
qualitative uncertainty analysis, including pre-industrial sulfate, nitrate, base cation, and ANC
levels, as well as characterizing the relationships between naturally occurring organic acids and
F-48
-------
acid deposition. In parallel, long term research plans and monitoring efforts should be deigned
to compare results from steady-state and dynamic models.
It is essential that surface water monitoring programs be maintained and soil and
biological monitoring programs be strengthened, as they provide the information not only to
support development of ecosystem models but also to enable broader application nationally in
under sampled areas. A concerted effort to elevate and sustain water and soil observation
programs is key to most foreseeable reviews and development of secondary standards relevant to
the effects of atmospheric deposition. This is especially important as future reviews are likely to
broaden the focus of ecological endpoints beyond aquatic acidification.
Linking modeling across atmospheric and terrestrial/aquatic media.
Research leading to improved linkages between atmospheric and watershed models
would support assessments relating atmospheric deposition to ecosystem effects. The spatial
scales of application typically are vastly different as watershed models target more spatially
resolved catchment specific areas whereas air quality models such as CMAQ generally are
applied at 12 km horizontal scale resolution. For example, consideration to nesting grid layers
for finer scale resolution, similar to what is often performed to capture urban area gradients,
could add insight into sub grid scale variability of deposition and the relative consequences on
driving water quality models. However, the long time period averaging of water quality models
tends to reduce the relative variation in spatial allocation of deposition. Research into bi-
directional flux mechanisms is especially relevant to the model linkage between atmospheric and
terrestrial models. The continued improvement and parallel adoption of soil process models is
key to this linkage. Further efforts in characterizing the sources and processes related to reactive
organic nitrogen most likely is best addressed through a linked modeling approach, given the
likely dependence of organic nitrogen on a variety of ecosystem and atmospheric processes.
Consideration of incorporating and modifying available processes to characterize cation
deposition in air quality models is a potentially important step in linking atmospheric and
terrestrial/aquatic media.
F-49
-------
F.6 Summary and Conclusions
Uncertainty and natural variability exist in all of the components of the structure of the
NOx and SOx standard introduced in this PA, and should be considered in establishing the level
of the AAI. A summary of the relative uncertainties of these components is provided in Table F-
3. On balance, the confidence level in the information and processes associated with the
linkages from ecological effects to atmospheric conditions through deposition and ecosystem
modeling is very high. The considerable body of evidence is conclusive with regard to causality
between aquatic acidification and biological and ecological effects. Confidence in the linkage
associating aquatic acidification and ANC is extremely high, as the aquatic chemistry describing
this relationship, while nonlinear, is relatively simple with regard to chemical species and
reactions. The relationships between deposition and ANC, while complicated by a variety of
biogeochemical and hydrological processes and data requirements within watersheds, are well
established and the critical load models have been thoroughly vetted through the scientific
community with a demonstrated level of successful evaluation. The linkages between ambient
concentrations of relevant species and deposition are best handled through air quality modeling
systems like CMAQ. The relationship between concentrations and deposition loads is well
characterized by these models, which are constrained by mass balance principles. While much
of the physical and chemical processing that determine concentrations and consequent deposition
is interwoven with numerous fundamental processes characterizing mass transport and
atmospheric chemical oxidation, the science is relatively mature with years of applications and
continued evolution of the models. The specific processes guiding nitrogen and sulfur chemistry
and deposition are relatively simple. More challenging is the ability to parameterize processes at
the air-surface interface which guide the estimation of deposition velocities and the re-emission
of certain species, as well as many of the area wide natural processes and agricultural practices
which influence emissions of oxidized and reduced forms of nitrogen.
The variety of uncertainty, variability and sensitivity analyses included in this chapter
have been conducted under the assumption that the basic model construct is solid, as discussed
immediately above, and are used to inform conclusions regarding the level of the AAI that
incorporate consideration of uncertainty. These analyses are also useful in guiding
implementation efforts related to future monitoring, emissions and model process improvements.
The influence of uncertainty on the level of the AAI can be thought of as reducing or increasing
F-50
-------
relative stringency of the level to increase the likelihood that requisite protection of public
welfare is provided. Throughout these discussions there is no apparent directional bias in the
uncertainty regarding the biological, chemical and physical processes incorporated in the AAI.
From the perspective of valuation of ecosystem services, the estimates generally are believed to
be biased low, meaning the values of reaching a target level of protection are underestimated.
However, quantification of these values is perhaps the most uncertain of all aspects considered.
Consequently, the level of the AAI should be relatively high in a buffering context to account
for the existence of uncertainties in several components. In addition to, but related to these
uncertainties discussions, are considerations of time lag to reach a target level ANC due to
ecosystem response dynamics, as well the uncertainties in the severity and prevalence of episodic
events. Both of these considerations suggest support for an AAI that is somewhat higher than
the target ANC supported by the specific evidence and risk information.
F-51
-------
Table F-3. Summary of Qualitative Uncertainty Analysis of Key Elements Affecting the AAI form of the NOx/SOx Standards.
Source
Description
Potential influence of
uncertainty in element
Direction
(negative
implies
less
relative
protection)
Magnitude
Knowledge-
Base
uncertainty
Comments
Major elements (and sub-models) of the ecological effects to ambient concentration framework
Biological/ecosystem
response to
acidification
Linkage between
direct acidification
species and
ecological indicator
(ANC)
Linkage between
ecological indictor
and adverse
ecological effects
Deposition to ANC
linkage through
Critical Load
approach
Clear associations between
aquatic acidification (pH,
elevated Al) and adverse
ecosystem effects (fish
mortality, decreased species
diversity)
The relationships across
ANC, pH and dissolved Al
are controlled by well
defined aquatic equilibrium
chemistry
Direct nonlinear
associations between ANC
and fish mortality and
species diversity
Mass-balance Steady State
critical load model is
applied to determine critical
load values. MAGIC
Both
Both
Both
Both
Low
Low
Low-
medium
Low
Low
(regionally)
Low
Low
Low
The ecosystem level responses are well studied at regional
levels. The uncertainty increases at larger scales due to an
increasing number of factors influencing the patterns (e.g.
latitudinal species gradient, specie-area relationships, etc.).
ANC is the preferred ecosystem indicator as it has a direct
relationship with pH and the deposition species relevant to
the NOx/SOx standard.
Although the pH dependency on ANC is nonlinear, it is
always directionally consistent. In extremely low and
high ANC environments the relationship is of minimal
value as catchments are in relatively "less sensitive"
regimes due to natural conditions or extreme
anthropogenic influence (i.e., acid mine drainage). In
sensitive areas of concern the relationship essentially is
similar to the relationships between direct acidification
species and adverse effects.
The model formulation is well conceived and based on a
substantial amount of research and applications available
in the peer reviewed literature. There is greater
uncertainty associated with the availability of data to
F-52
-------
Source
Description
Potential influence of
uncertainty in element
Direction
(negative
implies
less
relative
protection)
Magnitude
Knowledge-
Base
uncertainty
Comments
model is used to validate
steady State model. The
Steady State critical load
model formulation is used
as the foundation for
deriving the AAI equation.
support certain model components.
Atmospheric
concentrations to
deposition
Deposition is a direct
function of ambient
concentration, influenced
by several processes, and
handled in the AAI through
air quality modeling.
Both
Low
Low
The model design is appropriate given the spatial and
temporal complexities that influence deposition velocity,
as well as the variety of atmospheric species that generally
are not measured. Greater uncertainty resides in the
information (e,g,, ammonia emissions) driving these
calculations and availability of observations to evaluate
model behavior.
Ecological indicator
to changes in the
value of ecosystem
services
Definitions of public
welfare may include
economic considerations,
based on the tradeoffs
people would make to
avoid the negative impacts
of acidification, through
effects on the values of
ecosystem services.
Empirical estimates of
valuation for limited
ecosystem service
categories are used to
Negative
Medium-
high
Low-medium
There are many studies that estimate the value of
increasing services that may be affected by changes in
acidification and eutrophication. However, few of these
studies focus on the particular impact of acidification and
eutrophication on the quality of these services and
preferences for avoiding these impacts.
Those studies that do are often limited to analyzing the
impacts on a narrow population or particular change in
environmental quality. The monetized benefits to fishers
and to New York residents for ecosystem improvements in
the Adirondacks associated with improvements to the
ecological indicator are significant underestimates of the
F-53
-------
Source
Description
inform the discussions of
adversity associated with
alternative ANC levels.
Potential influence of
uncertainty in element
Direction
(negative
implies
less
relative
protection)
Magnitude
Knowledge-
Base
uncertainty
Comments
total benefits in the U.S. This is because those living
outside New York would value improvements to the
Adirondacks and similar natural environments elsewhere.
The methodologies used in the studies that underlie the
estimates of the value of changes in ecosystem services in
the Adirondacks region are sound and have been subject to
peer review. The method of aligning the improvements
valued in the Banzhaf et al. study with estimates of
eliminating current damages leads to may lead to an over
or underestimate of the benefits. The range of this
difference is difficult to know a priori, but the total
improvements in the share of lakes that improve above an
ANC threshold of 20 ueq/L are consistent.
Sub-components and data of individual models
Atmospheric Components
D^
DepNo
Oy
Annual deposition of sulfur
mass from dry deposition of
(SO2 and SO4) and wet SO4
derived from CMAQ 12km
horizontal grid resolution
averaged over 5 years
Annual deposition of
oxidized nitrogen mass
from dry deposition of (all
NOy species) and wet NO3
both
both
low
low
low
low-medium
The treatment of SOx deposition in EPA air quality
models has evolved over the last two decades. There is
general consensus that the overall mass balance of S is
treated well with difficulties in spatial pairing of
observations and modeled results of wet deposition. This
spatial pairing has improved with the more recent PRISM
adjustments.
The treatment of oxidized nitrogen deposition in EPA air
quality models has evolved over the last two decades.
There is general consensus that the overall mass balance
of oxidized N is treated well. However, the broad range of
F-54
-------
Source
De HX
Hx
Wet deposition
(generically - N and
S species)
Description
derived from CMAQ 12 km
horizontal grid resolution
averaged over 5 years
Annual deposition of
reduced nitrogen mass
from dry deposition of
(NH3 and SO4) and wet
NH4 derived from CMAQ
12km horizontal grid
resolution averaged over 5
years
Wet component of total
deposition as described in
the Dep terms, above
Potential influence of
uncertainty in element
Direction
(negative
implies
less
relative
protection)
both
both
Magnitude
low
low
Knowledge-
Base
uncertainty
medium
low
Comments
deposition velocities across NOy species, and especially
uncertainties regarding the deposition of significant
species such as NO2 pose ongoing challenges. Similarly,
a shortage of NOy species measurements as well a lack of
techniques to directly measure dry deposition impede
progress on improving parameterization of N dry
deposition.
NHx deposition also is quantified through CMAQ
applications. The well dispersed nature of agricultural
based emissions that are influenced strongly by
meteorological and surface /soil characteristics continues
to challenge characterization of ammonia emissions.
Recent incorporation of a bi-directional flux process in
CMAQ improves consistency with available scientific
understanding and yields improved time and space pairing
of limited observations with model results. A lack of both
ammonia and ammonium ambient observations continues
to compromise our ability to characterize uncertainty in
our treatment of NHx. As with all dry deposition
estimates, technologies for direct measurements are not
available routinely. Both NHx deposition and NOx
deposition are assigned low values of magnitude based on
a general dominating role of sulfur deposition.
Wet deposition remains an attribute of relatively high
confidence based on the ability to directly measure
chemical components in precipitation samples.
However, given the stochastic nature of precipitation,
F-55
-------
Source
Description
Potential influence of
uncertainty in element
Direction
(negative
implies
less
relative
protection)
Magnitude
Knowledge-
Base
uncertainty
Comments
models have a difficult time in matching observations.
The use of 5 year averages and post-processing PRISM
adjustments have reduced uncertainty in spatial pairing of
observations and modeled estimates.
Dry deposition
(generically - N and
S species)
Dry component of total
deposition as described in
the Dep terms, above
both
medium
Medium-high
The absence of direct dry deposition measurements
combined with the significant variability in the parameters
that influence dry deposition velocity reduces the
confidence level in dry deposition relative to wet
deposition.
Deposition
Transference Ratios
CMAQ derived ratio of
total oxidized deposition to
concentration averaged
over one year
both
low
unknown
Transference ratios enable the connection between
deposition and the policy relevant ambient air indicators,
NOy and (SO2 + SO4). They are strictly a model
construct and cannot be evaluated in a traditional model to
observation context. The low sensitivity of these ratios
to emission changes and inter annual meteorology
combined with low spatial variability indicate that these
ratios are necessarily stable.
Ambient concentrations of
NOy through observations.
negative
low
Low-medium
Adequate spatial coverage of NOy observations does not
exist, but will be addressed in the proposed rule. The
monitoring technology only over the last 5 years has been
perceived as "routine" based on incorporation in the
NCore network. However FRM status for NOy
instruments currently is not available. The negative bias
direction is a standard caveat to any instrument relying on
internal air stream conversion of atmospheric species prior
to detection.
Ambient concentrations of
both
low
Low
A lack of adequate spatial coverage is the primary concern
F-56
-------
Source
Description
NOy through observations.
Potential influence of
uncertainty in element
Direction
(negative
implies
less
relative
protection)
Magnitude
Knowledge-
Base
uncertainty
Comments
for SO2 + SO4 observations. FRM status is not available
for SO4; although the long track record of accurate and
precise CASTNET FP measurements indicates that
achieving FPJV1 status is a low hurdle.
Ecosystem Components
*
BC0
Neco
Pre-industrial base cation
concentrations
negative
positive
Medium-
high
low
high
medium
Both the F-factor approach and process based MAGIC
modeling were used to generate BC0* Excellent
agreement between both approaches was established in the
Shenandoah streams. The more comprehensive data
requirements of MAGIC limit its widespread use to the
Adirondacks, although for consistency the F-factor
approach was applied nationwide. The analyses also
illustrated greater divergence at higher critical loads, or
areas with greater acid buffering capacity and high bas
cation levels. These conditions often are screened out of
our population distribution analyses, and when included
do not affect the location within the distribution of the
more sensitive water bodies. Since MAGIC (the
preferred approach) tends to overestimate BC0*relative to
the F factor approach, and the F-factor is more widely
applied nationally, the BC0* estimates are viewed as
conservative leading to a slight positive bias in estimating
critical loads. Although we have many modeled
estimates of BCO*, there is a lack of direct measurements
of BC weathering rates.
The term Neco, as defined, has a relatively medium
confidence level and is a direct function of the uncertainty
F-57
-------
Source
Description
Potential influence of
uncertainty in element
Direction
(negative
implies
less
relative
protection)
Magnitude
Knowledge-
Base
uncertainty
Comments
inherent in the deposition estimates from CMAQ and
surface measurements of NO3. However, this
"measurement" difference approach reflects the average of
all influencing processes (dinitrification, uptake,
immobilization) over the time period of measurements.
Consequently, there is an inherent assumption of a
relatively static system (Neco is applied in a steady state
model) that generally is not tested. In concept, a true
steady state vision of Neco would be based on a mature
forested ecosystem. The relative bias of Neco is related,
largely, to the relative productivity of the forest. The
challenge in determining any potential bias in Neco is to
determine the relative "maturation age" of an ecosystem
which requires knowledge of future land use activities. In
areas of high land use restrictions of a recovering forest,
Neco would be assumed to be overestimated. The
relative magnitude of Neco often is mitigated by the
dominance of SOx in controlling acidification processes in
many systems. Furthermore, it is unclear to what extent
any stored N will be released back into the system, which
is assumed to not occur in the linked system model.
Annual runoff rate
(distance/time) for a
catchement.
both
low
high
Data used to calculate Q was compiled in 1985.
Streamflow data were collected at over 12,000 gauging
stations during 1951-80; 5,951 stations were selected for
the analysis. See Gebert and others (1987) for a complete
description of how the runoff was determined from the
streamflow datF. Appropriate maps of the data can show
F-58
-------
Source
Description
Potential influence of
uncertainty in element
Direction
(negative
implies
less
relative
protection)
Magnitude
Knowledge-
Base
uncertainty
Comments
the geographical distribution of runoff in tributary streams
for the years 1951-80 and can describe the magnitudes and
variations of runoff nationwide. The data was prepared to
reflect the runoff of tributary streams rather than in major
rivers in order to represent more accurately the local or
small scale variation in runoff with precipitation and other
geographical characteristics.
W.F., Graczyk, D.J., and Krug, W.R., 1987, Average
annual runoff in the United States, 1951-80: U.S.
Geological Survey Hydrologic Investigations Atlas HA-
710, scale 1:7,500,000.
DOC
Surface water dissolved
organic carbon
negative
low
medium
Water bodies with high DOC levels (> 10mg/l) were
screened out of the critical load calculations in order to
avoid naturally acidic systems. However, the inherent
assumption of ANC = £strong CA - £strong AN does not
explicitly account for contributions of weak organic acids.
Consequently, a small positive bias pervades the critical
load calculations (i.e., the CL estimates are high). The
knowledge base value of M reflects a general shortage of
DOC data.
F-59
-------
REFERENCES:
Aiyyer, A, Cohan, D., Russell, F., Stockwell, W., Tanrikulu, S., Vizuete, W., and Wilczak, J.
2007. Final Report: Third Peer Review of the CMAQ Model.
Altieri, K.E., BJ. Turpin and S.B. Seitzinger, 2009, Composition of dissolved organic nitrogen
in continental precipitation investigated by bu ultra-high resolution FT_ICR mass
spectrometry, Environmental Science and Technology, 43, 18, 6950 - 6955.
Byun, D.W., and Schere, K.L. 2006. Review of the Governing Equations, Computational
Algorithms, and Other Components of the Models-3 Community Multiscale Air Quality
(CMAQ) Modeling System. J. Applied Mechanics Reviews, 59 (2), 51-77'.
Dennis, R. and K. Foley, 2009, Adapting CMAQ deposition fields for critical loads analyses,
presented at 2009 NADP Confernce; manuscript in preparation.
Dennis, R., R. Mathur, I.E. Pleim and J.T. Walker, 2010, Fate of Ammonia Emissions at the
Local to Regional Scale as Simulated by the Community Multiscale Air Quality Model,
Atmospheric Pollution Research.
Foley, K.M., Roselle, SJ, Appel, K.W., Phave, P.V., Pleim, I.E., Otte, T.L., Mathur, R, Sarwar,
G., Young, J.O., Gilliam, R.C., Nolte, C.G, Kelly, J.T., Gilliland, F.B., Bash, J.O. (2010)
Incremental testing of the Community Multiscale Air Quality (CMAQ) modeling system
version 4.7, Geosci. Model Dev., 3, 205-226.
Holdren, G.R., T.C. Strickland, P.W.Shaffer, P.F. Ryan, P.L. Ringold and R.S. Turner. 1992.
Sensitivity of Critical Load Estimates for Surface Waters to Model Selection and
Regionalization Schemes, J. Environ. Qual. 22:279-289.
Otte, T.L. and Pleim, I.E. (2010) The Meteorology-Chemistry Interface Processor (MCIP) for
the CMAQ modeling system, Geosci. Model Dev., 3, 243-256.
PRISM Climate Group, Oregon State University, http://www.prismclimate.org, created 4 Feb
2004U.S. EPA (Environmental Protection Agency) 2009.
Risk and Exposure Assessment for Review of the Secondary National Ambient Air Quality
Standards for Oxides of Nitrogen and Oxides of Sulfur-Main Content - Final Report. U.S.
Environmental Protection Agency, Research Triangle Park, NC, EPA-452/R-09-008a.
F-60
-------
Appendix G
Cumulative Uncertainty Analysis
This appendix provides analyses of the relative uncertainty in the AAI equation. The
analysis involves a Monte Carlo style approach that incorporates the relative uncertainties of the
input parameters. Cumulative results are illustrated in the form of the trade-off curve of the
allowable atmospheric concentrations of the oxides of N and S.
From the general form of the standard:
AAI = {ANClim + CL/QJ - NHx/Qr - TNOy [NOy]/Qr - TSOx[SOx]/Qr
There are six sources of uncertainty in calculating an AAI value for measured
atmospheric concentrations of the oxides of N and S; however these individual uncertainties are
not applied in the same manner. Four of these, Neco, NHx, Tnoy and Tsox rely on data from
CMAQ modeling. The uncertainty in CMAQ values can be high (see appendix F for detailed
discussion of CMAQ modeling). These values as they are used in the AAI equation are averaged
either across individual water bodies in the ecoregion or across CMAQ 12 km grids within the
ecoregion. In order to translate the uncertainty to the AAI equation, these averaged values were
bootstrapped using uncertainty in CMAQ modeling to obtain an estimated uncertainty in the
average values. Uncertainty in the parameters BCo and Q affect the selection of a representative
Critical Load (CD) value for the region. To analyze this, the CL value was bootstrapped to
generate a distribution of CL values reflecting the underlying uncertainty in BCo and Q. In all
cases the error was modeled as a normal Gaussian distribution and the bootstrapping was
repeated 20,000 times to generate the distributions.
Two ecoregions were chosen to assess the uncertainty, the Northern Highlands (region
5.3.1) and the Northern Great Plains (region 9.3.3). Region 5.3.1 was chosen because it had the
largest number of critical load data points to work with (819). Region 9.3.3 was chosen because
it is a relatively less acid sensitive region that had a large enough number of Critical Load data
points (164) to use for analyses. For all of these analyses an ANC level of 50 and a percentile of
75% were selected.
-------
Nitrogen Deposition (Ndep)
Modeled nitrogen deposition {Ndep) is used in two different components of the AAI, the Neco
value and in Tnoy. Figure G-l shows the results of bootstrapping analyses. These analyses
allowed for 100% uncertainty in the deposition at the grid level. Averaged over the entire
regions the bootstrapped distribution has uncertainty that is +15% in Region 5.3.1 and +10% in
Region 9.3.3.
Histogram of n_dep_bootstrap$means
Histogram of n dep bootstrapSmeans
54 56 58 60
Mean of Ndeposition
22 23 24
Mean of Ndeposition
A.
B.
Figure G-l. Bootstrap distribution of mean nitrogen deposition for Regions 5.3.1 (A) and 9.3.3
(B)
Nitrogen Uptake (Neco)
The Neco term directly uses the mean deposition. The Neco is calculated as the Average Neco =
Ndep - Nitrogen Leaching (Nleach), where Ndep is the regional average and Nleach is a lake
specific value. Neco is calculated at each lake individually, and the mean Neco for the region is
used in the AAI equation. Ndep was selected randomly from the distribution generated above
(Figure G-l). Nleach was allowed to have 30% uncertainty. Figure G-2 shows the results of the
Neco bootstrapping analysis of the mean values for the two regions. The results indicate that the
G-2
-------
mean Neco term varies widely with a range of <20 to >120 in Region 5.3.1 and <10 to 60 in
Region 9.3.3. This equates to a range of uncertainty that is from approximately -75% to 135% in
Region 5.3.1 and -65% to 218% in Region 9.3.3.
Histogram of neco_bootstrap
Histogram of neco_bootstrap
20 40 60 80 100 120
Mean of Neco
20 40 60 80 100 120
Mean of Neco
A.
B.
Figure G-2. Bootstrap distribution of mean Nitrogen uptake (Neco) for Regions 5.3.1 (A) and
9.3.3 (B)
Transference Ratios (Tnoy, Tsox)
The Nitrogen transference ratio (Tnoy) also uses CMAQ derived values for Nitrogen deposition,
but Tnoy, and the transference ratio for S, Tsox) are calculated at the grid level. The transference
ratio analyses included uncertainty at the grid level in N and S deposition and in the ambient
concentrations of NOy and Sox (Nconc and Scone), also modeled by CMAQ. The transference
ratios were calculated as Tnoy = Ndep/Nconc and Tsox = Sdep/Sconc. Uncertainty in the
CMAQ values was included randomly before the ratio was calculated, with uncertainty in
deposition (Ndep and Sdep) up to 75%, and uncertainty in concentration (Nconc and Scone) up to
25%. The uncertainty in the deposition terms is lower than used above because this value
incorporates the concentration terms in CMAQ, so a portion of the uncertainty in these two terms
will covary and would be directly offset in the calculation of the transference ratios. The mean
value for the region was then calculated. The results of these analyses are shown in figures G-3
G-3
-------
and G-4. In both analyses there were a small number of extreme outliers and the axes in the
figures have been truncated to allow better visualization of the results. These outliers were
included in the final analyses. The results were similar for Tnoy and Tsox. The range of values,
excluding outliers, was larger for both ratios in Region 5.3.1, but in both regions the uncertainty
ranged from -75% to > 100%, excluding the extreme outliers.
A.
Histogram of tnoy_bootstrap
15 20 25
Mean of Transfer Ratio for NOy
B.
Histogram of tnoy bootstrap
10 12 14 16 18
Mean of Transfer Rafio for NOy
Figure G-3. Bootstrap distribution of mean nitrogen transference ratios (Tnoy) for Regions 5.3.1
(A) and 9.3.3 (B)
G-4
-------
Histogram of tsoy_bootstrap
Histogram of tsoy bootstrap
20 30 40
Mean of Transfer Ratio for SOy
15 20 25
Mean of Transfer Ratio for SOv
A.
B.
Figure G-4. Bootstrap distribution of mean sulfur transference ratios (Tsox) for Regions 5.3.1
(A) and 9.3.3 (B)
Reduced Nitrogen
Reduced nitrogen deposition (NHx) was analyzed in the same way as Ndep with the same
uncertainty of 100%. The results are show in figure G-5. The average NHx deposition is higher
in Region 5.3.1, but in both regions the range or uncertainty is approximately +20%.
G-5
-------
Histogram of nhx bootstrapSmeans
Histogram of nhx_bootstrap$means
17 18 19
Mean of NHx Deposition
10.0 10.5
Mean of NHx Deposition
A.
B.
Figure G-5. Bootstrap distribution of mean reduced nitrogen deposition (NHx) for Regions 5.3.1
(A) and 9.3.3 (B)
Critical Load
The critical load (CL) that is selected to represent the ecoregion is dependent on the Target
Percentile and level set as part of the standard. This value is represented in the AAI equation as
the Base Cation weathering (BCo) and the hydrologic term, Q. Uncertainty in these two terms
will affect the value of the CL. As represented in the AAI equation, BCo and Q are constants
that, together with the selected ANC level, are equal to the representative CL. To model this
uncertainty the CL values were bootstrapped including 70% uncertainty in the BCo term and 5%
uncertainty in the Q term. These results are shown in Figure G-6. The CL has an uncertainty
range of approximately +30% in Region 5.3.1 and -25% to 35% in Region 9.3.3.
G-6
-------
Histogram of q25_cl
Histogram of q25_cl
60
q25_cl
q25_cl
A.
B.
Figure G-6. Bootstrap distribution of representative critical load values for Regions 5.3.1 (A)
and 9.3.3 (B)
Combined Uncertainty in the AAI
To illustrate the combined effect of the uncertainty developed above a trade-off curve
representing the allowable combined concentrations of S and N was developed. The trade-off
curve was derived using the raw data for these ecoregions (Figure F-7, dashed lines). The curve
was then derived again using values from the derived distributions ofNeco, NHx, Tnoy, Tsox,
and the CL. This was repeated 20,000 times with the resulting curves plotted in light gray in
Figure G-7. The solid black lines represent 95% confidence intervals, the inner quartiles (25th
and 75th percentiles) and the mean. For this analysis the Neco term was treated separately from
the CL selection to allow plotting of the tradeoff curves.
In Region 5.3.1 the mean value derived from the modeling is the same as the value
derived from the raw data and the curves are directly on top of each other. The 95% confidence
intervals represent a range of approximately +15% and the inner quartile approximately +5%
relative to the maximum allowable concentration of S oxides. The uncertainty associated with
the maximum allowable concentration of NOx is larger when looking at the 95% confidence
G-7
-------
interval, approximately ±45%, and at the inner quartiles, +20% uncertainty. The larger
uncertainty on the NOx axis is due to the Neco and NHx terms.
In Region 9.3.3 the mean value derived from the modeling is the same as the curve
derived from the raw data. The 95% confidence intervals represent a range of uncertainty that is
approximately +25% and +10% for the inner quartiles relative to the maximum allowable
concentration of S oxides. The uncertainty associated with the maximum allowable
concentration of NOx is in the same range for this region.
4
NOy
4
NOy
A.
B.
Figure G-7. NOy/SOx tradeoff curve with results of cumulative uncertainty analysis for Regions
5.3.1 (A) and 9.3.3 (B)
G-8
-------
-------
United States Office of Air Quality Planning and Standards Publication No. EPA-452/R-1 l-005b
Environmental Protection Health and Environmental Impacts Division February 2011
Agency Research Triangle Park, NC
------- |