United States
Environmental Protection
^^Bml M lkAgency
EP A/600/R-20/278
September 2020
www.epa. gov/ncea/isa
Integrated Science Assessment
for Oxides of Nitrogen, Oxides of
Sulfur, and Particulate Matter—
Ecological Criteria
(Final)
Center for Public Health and Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC

-------
Disclaimer
This document has been reviewed in accordance with the U.S. Environmental Protection
Agency policy and approved for publication. Mention of trade names or commercial
products does not constitute endorsement or recommendation for use.
2

-------
INTEGRATED SYNTHESIS
IS.1 Introduction to This Integrated Science Assessment (ISA)
IS.1.1 Purpose
The Integrated Science Assessment (ISA) is a comprehensive evaluation and synthesis of
the policy-relevant science. Policy-relevant science is that which is "useful in indicating
the kind and extent of all identifiable effects on public health or welfare which may be
expected from the presence of [a] pollutant in the ambient air," as described in
Section 108 of the Clean Air Act (C'AA. 1990).1 This ISA communicates critical science
judgments on the ecological criteria for oxides of nitrogen, oxides of sulfur, and
particulate matter (PM). Accordingly, this ISA is the scientific foundation for the review
of the ecological effects of the current secondary (welfare-based) National Ambient Air
Quality Standards (NAAQS) for oxides of nitrogen, oxides of sulfur, and particulate
matter. The Clean Air Act definition of welfare effects includes, but is not limited to,
effects on soils, water, wildlife, vegetation, visibility, weather, and climate, as well as
effects on man-made materials, economic values, and personal comfort and well-being.
The nonecological welfare effects associated with particulate matter, such as climate and
visibility, are considered part of a separate, ongoing review of PM that is outlined in the
Integrated Review Plan (IRP) for the National Ambient Air Quality Standards for
Particulate Matter (U.S. EPA. 2016a). The human health effects are evaluated in separate
assessments conducted as part of the review of the primary (human health-based)
NAAQS for oxides of nitrogen (U.S. EPA. 2016c). oxides of sulfur (U.S. EPA. 2016b).
and as noted above, particulate matter (U.S. EPA. 2019).
Oxides of nitrogen, oxides of sulfur, and particulate matter are reviewed here together
because they are interrelated through complex chemical and physical atmospheric
processes and because they all contribute to nitrogen (N) and sulfur (S) deposition, which
in turn contributes to well-documented ecological effects. In this document, the term
"oxides of nitrogen" refers to all forms of oxidized nitrogen (NOy) compounds, including
1 The general process for developing an ISA, including the framework for evaluating weight of evidence and
drawing scientific conclusions and causal judgments, is described in a companion document, Preamble to the
Integrated Science Assessments (U.S. EPA. 2015').
3

-------
NO, NO2, and all other oxidized N containing compounds formed from NO and NO2.1
Oxides of sulfur2 are defined here to include sulfur monoxide (SO), sulfur dioxide (SO2),
sulfur trioxide (SO3), disulfur monoxide (S2O), and sulfate (S042 ). However, SO, SO3,
and S2O are present at much lower ambient levels than SO2 and SO42 and are therefore
not discussed further. Particulate matter is composed of some or all of the following
components: nitrate (NO? ). SO42 . ammonium (NH4 ), metals, minerals (dust), and
organic and elemental carbon (C).
This ISA updates the 2008 Integrated Science Assessment for Oxides of Nitrogen and
Sulfur—Ecological Criteria [hereafter referred to as the 2008 ISA (U.S. EPA. 2008)1. as
well as the ecological portion of the Integrated Science Assessment for Particulate Matter
(U.S. EPA. 2009a). with studies and reports published from January 2008 through May
2017. Thus, this ISA updates the state of the science that was available for the 2008 ISA,
which informed decisions on the secondary oxides of nitrogen and oxides of sulfur
NAAQS in the review completed on March 20, 2012. In the final rulemaking, the
Administrator's decision was that, while the current secondary standards were inadequate
to protect against adverse effects from deposition of oxides of nitrogen and oxides of
sulfur, it was not appropriate under Section 109(b) to set any new secondary standards at
this time due to the limitations in the available data and uncertainty as to the amount of
protection the metric (Aquatic Acidification Index—see Section IS.2.2.6) developed in
the Policy Assessment (U.S. EPA. 2011) would provide against acidification effects
across the country (77 FR 20281). In addition, the Administrator decided that it was
appropriate to retain the current nitrogen dioxide (NO2) and sulfur dioxide (SO2)
secondary standards to address direct effects of gaseous NO2 and SO2 on vegetation.
Thus, taken together, the Administrator decided to retain and not revise the current NO2
and SO2 secondary standards: an NO2 standard set at a level of 0.053 ppm, as an annual
arithmetic average, and an SO2 standard set at a level of 0.5 ppm, as a 3-hour average, not
to be exceeded more than once per year (77 FR 20281). The current secondary standards
for PM are intended to address PM-related welfare effects, including visibility
impairment, ecological effects, and effects on materials and climate. These standards are
a 3-year annual mean PM2 5 concentration of 15 |ig/m3. with the 24-hour average PM2 5
and PM10 set at concentrations of 35 |ig/m3 and 150 |ig/m3. respectively.
1	This ISA reserves the abbreviation NOx strictly as the sum of NO and NO2—consistent with that used in the
atmospheric science community—and uses the term "oxides of nitrogen" to refer to the broader list of oxidized
nitrogen species. Oxides of nitrogen refers to NOy as the total oxidized nitrogen in both gaseous and particulate
forms. The major gaseous and particulate constituents of NOy include nitric oxide (NO), nitrogen dioxide (NO2),
nitric acid (HNO3), peroxyacetyl nitrate (PAN), nitrous acid (HONO), organic nitrates, and particulate nitrate (NO3).
This ISA uses the definitions adopted by the atmospheric sciences community.
2	Oxides of sulfur refers to the criteria pollutant category.
4

-------
This new review of the secondary oxides of nitrogen, oxides of sulfur, and particulate
matter NAAQS is guided by several policy-relevant questions that were identified in The
Integrated Review Plan for the Secondary National Ambient Air Quality Standard for
Nitrogen Oxides, Sulfur Oxides, and Particulate Matter [hereafter referred to as the 2017
IRP (U.S. EPA. 2017a)!.
To address these questions, this ISA aims to characterize the evidence available in the
peer-reviewed literature for ecological effects associated with:
•	the major gaseous and particulate constituents of total oxidized N (NOy), which
include NO, NO2, HNO3, PAN, HONO, organic nitrates, and NO.? :
•	the major gaseous and particulate constituents of SOx, which include SO2 and
SO42 ; and
•	PM composed of some or all of the following components: particulate NO3 ,
particulate SO42 . ammonium (NH4+), metals, minerals (dust), and organic and
elemental carbon (C).
The assessment activities include:
•	Identifying policy-relevant literature.
•	Evaluating strength, limitations, and consistency of findings.
•	Integrating findings across scientific disciplines and across related ecological
outcomes.
•	Considering important uncertainties identified in the interpretation of the scientific
evidence.
•	Assessing policy-relevant issues related to quantifying ecological risks, such as
ambient air concentrations, deposition, durations, and patterns associated with
ecological effects; the relationship between ambient air concentrations, deposition,
and ecological response and the existence of thresholds below which effects do
not occur; and species and populations potentially at increased risk of ecological
effects.
New analyses with the goal of quantifying risk, such as new model runs, Critical Loads
(CLs) exceedance maps, and quantified uncertainties regarding modeled scenarios are not
conducted in the ISA. These types of analyses, if pursued, require the selection of
chemical or biological limits that define CLs and represent adversity. These analyses
would also require choosing a time period over which to average deposition. Such
scope-of-analysis decisions are more appropriate for the Risk and Exposure Assessment,
as described in the 2017 IRP (U.S. EPA. 2017a). The information summarized in this ISA
will serve as the scientific foundation of the Risk and Exposure and Policy Assessments
during the current review of the secondary oxides of nitrogen, oxides of sulfur, and
particulate matter NAAQS.
5

-------
IS.1.2 Process and Development
The U.S. EPA uses a structured and transparent process to evaluate scientific information
and determine the causality of relationships between air pollution and ecological effects
[see Preamble (U.S. EPA. 2015)1. The ISA development includes approaches for
literature searches, criteria for selecting and evaluating relevant studies, and a framework
for evaluating the weight of evidence and forming causal determinations. As part of this
process, the ISA is reviewed by the public and by the Clean Air Scientific Advisory
Committee (CASAC), which is a formal independent panel of scientific experts. This ISA
informs the review of the secondary oxides of nitrogen, oxides of sulfur, and particulate
matter NAAQS and therefore integrates and synthesizes information characterizing NOy,
SOx, and PM air concentrations. It also examines deposition of these substances and their
ecological effects. Relevant studies include those examining atmospheric chemistry,
spatial and temporal trends, and deposition, as well as U.S. EPA analyses of air quality
and emissions data. Relevant ecological research includes geochemistry, microbiology,
physiology, toxicology, population biology, and community ecology. The research
includes experimental laboratory and field additions of the pollutants, as well as gradient
studies.
The U.S. EPA conducted literature searches to identify relevant peer-reviewed studies
published since the previous ISA (i.e., from January 2008 through May 2017;
Figure IS-1V
6

-------
Considered
Excluded
(not relevant)
Manual screening
for pertinence
(title)
Manual screening for
pertinence
(abstract/full text)
Topic-specific
keyword searches
by authors
Cited in second draft of ISA
(availableon HERO project page)
References from 2008 ISA and other sources
2,203 references
Relevant literature for
each topic
Sorted by automatedtopic
classification of abstracts,
OR
Ranked based on numbers of
citations of 2008 ISA references
Keyword searches
OR
Citation searches
of Web of Science
for peer-reviewed papers
published 01/01/2008-7/31/2017
231,805 references
HERO = Heaith and Environmental Research Online; ISA = Integrated Science Assessment.
Figure IS-1 Workflow for collecting relevant literature for the 2017 Integrated
Science Assessment for Oxides of Nitrogen, Oxides of Sulfur, and
Particulate Matter—Ecological Criteria.
Multiple search methods were used in the Web of Science database [Preamble (U.S.
EPA. 2015). Appendix 2], including searches by keyword and by citations of 2008 ISA
references. Subject-matter experts and the public were also permitted to recommend
studies and reports during kick-off workshops held by the U.S. EPA in March 2014 for
oxides of nitrogen and oxides of sulfur and in February 2015 for particulate matter. The
new references were sorted by automated methods into topic areas based on wording in
the publication's abstract or numbers of citations of 2008 ISA references, and the
resultant relevant literature was reviewed by the ISA authors. Studies were screened first
based on the title and then by the abstract; studies that did not address a relevant research
topic based on this screening were excluded. The U.S. EPA also identified studies from
previous assessments as definitive works on particular topics to include in this ISA. The
HERO project page for this ISA
(https://hcronct.cpa.LHJv/hcronct/iiidcx.cfm/proicct/pagc/proicct id/2965) contains the
references that are cited in the ISA and electronic links to bibliographic information and
abstracts.
7

-------
The Preamble to the Integrated Science Assessments (U.S. EPA. 2015) describes the
general framework for evaluating scientific information, including criteria for assessing
study quality and developing scientific conclusions. For ecological studies, emphasis is
placed on studies that characterize quantitative relationships between criteria pollutants
and ecological effects that occur at concentration and deposition levels relevant to current
ambient levels in the U.S. However, experimental studies with higher exposure
concentrations are included if they contribute to an understanding of mechanisms.
This ISA draws conclusions about relationships between NOy, SOx, and PM and
ecological effects by integrating information across scientific disciplines and related
ecological outcomes and synthesizing evidence from previous and recent studies.
Determinations are made about causation, not just association, and are based on
judgments of consistency, coherence, and scientific plausibility of observed effects, as
well as related uncertainties. The ISA uses a formal causal framework [Table II of the
Preamble (U.S. EPA. 2015)1. which is based largely on the aspects for causality proposed
by Sir Austin Bradford Hill to classify the weight of evidence according to the five-level
hierarchy summarized below.
•	Causal relationship
•	Likely to be a causal relationship
•	Suggestive of, but not sufficient to infer, a causal relationship
•	Inadequate to infer the presence or absence of a causal relationship
•	Not likely to be a causal relationship
APPENDIX 7This ISA includes the Preface (legislative requirements and history of the
secondary oxides of nitrogen, oxides of sulfur, and particulate matter NAAQS), an
Executive Summary, an Integrated Synthesis, and 16 appendices. The general process for
developing an ISA is described in a companion document, Preamble to the Integrated
Science Assessments (U.S. EPA. 2015). The Integrated Synthesis summarizes the
scientific evidence that best informs policy-relevant questions that frame this review.
Appendix 1 is an introduction to the appendices. Appendix 2 characterizes the sources,
atmospheric processes, and the trends in ambient concentrations and deposition of NOy,
SOx, and PM. Appendix 3 describes direct effects of NOy and SOx gases on plants and
lichens. Appendix 4-Appendix 6 describe N and S deposition effects on terrestrial
biogeochemistry and the terrestrial biological effects of terrestrial acidification and N
enrichment. Appendix 7 describes the effects of N and S deposition on aquatic
biogeochemistry. Appendix 8-Appendix 10 characterize the biological effects of
freshwater acidification, freshwater N enrichment, and N enrichment in estuaries and
near-coastal systems. Appendix 11 describes the effects of N deposition on wetlands, and
Appendix 12 characterizes the ecological effects of S as a nutrient. Appendix 13 presents
8

-------
information on climate modification of ecosystem response to N and S, and Appendix 14
discusses ecosystem services. Appendix 15 is a review of the ecological effects of forms
of PM that are not related to N or S deposition. Finally, Appendix 16 presents case
studies for six locations in the U.S. (southern/central California, northeastern U.S., Rocky
Mountain National Park, southeastern Appalachia, Tampa Bay, and the Adirondacks)
where data are sufficient to well characterize the ecological effects ofN and S deposition.
These sites would therefore make good candidates for further study to better understand
the linkages across various effects and ecosystems and to better assess risk and exposure.
IS.2 Connections, Concepts, and Changes
IS.2.1 Connections
Although scientific material in this ISA is divided into separate appendices for
atmospheric science and the multiple ecological effects, the strong links between the
atmosphere and terrestrial and aquatic ecosystems are acknowledged (Figure IS-2).
Emissions of NOy, SOx, and PM contribute to an accumulation of N and S in the
environment that creates a multitude of effects on terrestrial, wetland, and aquatic
ecosystems. Nitrogen is a vital component of all biological systems, serving as an
essential element to molecules such as amino acids and nucleic acids, which are among
the biochemical building blocks of life. As an organizing concept to understand the
effects of N within the environment, the sequence of transfers, transformations, and
environmental effects has been described as the "N cascade" (Galloway and Cowling.
2002). The concept of cascading effects also applies to S, which is also an essential
macronutrient. Specifics of biogeochemical cycling and biological effects of N are
discussed in Section IS.5 and for S are discussed in Section IS.9.
9

-------
Ecosystem Services
Appendix 14
Biological Effects
of S-riutrient
Appendix 12
Gas-phase
Ecological Effects
Appendix 3
Oxidation
so2—»-h2so4
NOx—>¦ HNOj
Dissolution
2H++ S042"
->-H++N0r
Other Ecological
Effects of PM
Appendix 15
Dry deposition
NO ,NH , SO
Biological Effects of
Terrestrial Acidification
Appendix 5
Biological Effects of
Terrestrial N Enrichment
Appendix 6
n2o no,
Wetlands
Appendix 11
Soil BGC
Appendix 4
Aquatic BGC
Appendix 7
Wet deposition
H+, NH4+, N03", SQ,2"
Biological Effects of
Freshwater Acidification
Appendix 8
Biological Effects
of Freshwater j
N Enrichment
Appendix 9
Biological Effects of
Estuarine N Enrichment
Appendix 10
Deposition
Ecological
Effect
Atmospheric Sciences
Appendix 2
Climate Modification
of Ecological Effects
Appendix 13
Ambient Air
Concentration
Ca2+ = calcium ion; GHG = greenhouse gas; H+ = hydrogen ion; H2S04 = sulfuric acid; HN03 = nitric acid; Mg2+ = magnesium ion;
N20 = nitrous oxide; N = nitrogen; NH3 = ammonia; NH4+ = ammonium; NHX = NH3 + NH4* + reduced organic nitrogen compounds;
NO = nitric oxide; N02 = nitrogen dioxide; N03" = nitrate; NOx = NO + N02; PAN = peroxyacetyl nitrate; PM = particulate matter;
S02 = sulfur dioxide; S042" = sulfate; SOx = S02 + S042"; VOC = volatile organic compounds.
The sum of reactive oxidized nitrogen species is referred to as NOY (NOY = NO + N02 + HN03 + 2N205 + HONO + N03" + N20
PAN + other organic nitrates).
Although not explicitly indicated, wet and dry deposition of PM components (e.g., metals, minerals, and secondary organic aerosols)
also occur and contribute to ecological effects.
Source: Modified from U.S. EPA (2008).
Figure IS-2 Overview of atmospheric chemistry, deposition, and ecological
effects of emissions of oxides of nitrogen, oxides of sulfur, and
reduced nitrogen.
IS.2.2 Concepts
This ISA draws on many methodological approaches and disciplines within the larger
scientific fields of ecology and atmospheric sciences. Hie studies discussed herein are
best understood in the context of some general concepts within these fields, such as
ecosystem scale, structure, and function (Section IS.2.2.1); deposition and source
apportionment to ecosystems (Section IS.2.2.2); critical loads (Section IS.2.2.3k
biodiversity (Section IS.2.2.4); the effects of reduced versus oxidized forms of N
10

-------
(Section IS.2.2.5); and the metric developed in the previous secondary NAAQS review,
the Aquatic Acidification Index (AAI; Section IS.2.2.6V The topics discussed in this
"Concepts" section do not have separate sections dedicated to them in the Integrated
Synthesis. The topics of ecosystem recovery, ecosystem services, and uncertainty, while
conceptual in nature, are not discussed here because they are the focus of more detailed
discussions in Section IS. 11. Section IS. 13. and Section IS. 14. respectively.
Ecosystem structure comprises both biodiversity and geography. Biodiversity
encompasses many quantitative measures of the abundance and distribution of organisms
within a defined geographical area (for a more explicit definition, see Section IS.2.2.1
and Section IS.2.2.4V Ecosystem function refers to processes that control fluxes and
pools of matter and energy in the ecosystem (Section IS.2.2.1). The loss of biodiversity is
a key consequence of the air pollutants discussed in this ISA. The importance of
preserving biodiversity and ecosystem function contributes to the sustainability of
ecosystem services that benefit human welfare and society in general (Section IS.2.2.4
and Appendix 14).
In human health assessments, dose-response relationships are used to identify
quantitative relationships between chemical exposure (dose) and health outcomes
(response), with emphasis on identifying thresholds, or the lowest doses at which
negative health outcomes are observed. In ecology, CLs provide a similar quantitative
relationship between chemical dose (e.g., deposition) and specific, quantitative changes
in ecological properties or processes (Section IS.2.2.3V For CLs to be used in evaluating
the effects of deposition upon ecosystems that receive N or S from multiple sources,
those other sources must be considered in comparison to deposition level
(Section IS .2.2.2). as well as the heterogeneous sensitivities of organisms and ecosystems
to different chemical forms of deposition (Section IS.2.2.5V
IS.2.2.1 Ecosystem Scale, Structure, and Function
For this assessment, an ecosystem is defined as the interactive system formed from all
living organisms (biota) and their abiotic (chemical and physical) environment within a
given area (IPCC. 2007). Ecosystem spatial boundaries are somewhat arbitrary,
depending on the focus of interest or study. Thus, the spatial extent of an ecosystem may
range from very small, well-circumscribed systems such as a small pond, to biomes at the
continental scale, or the entire globe (U.S. EPA. 2008). Ecosystem spatial scale does not
always correlate with complexity. A small pond may be a complex system with multiple
trophic levels ranging from phytoplankton to invertebrates to several feeding guilds of
fish. A large lake, on the other hand, may be a very simple ecosystem, such as the Great
11

-------
Salt Lake in Utah that covers approximately 1,700 square miles but contains only
bacteria, algae, diatoms, and two invertebrate species (U.S. EPA. 2013). All ecosystems,
regardless of size or complexity, have multiple interactions between biota and abiotic
factors. Ecosystems include both structural (geography and biodiversity [e.g., soil type
and food web trophic levels]) and functional (flow of energy and matter
[e.g., decomposition, nitrification]) attributes. Ecosystem changes are often considered
undesirable if important structural or functional components of the ecosystems are altered
following pollutant exposure (U.S. EPA. 2013. 1998).
Biotic or abiotic structure may define an ecosystem. Abiotic structure includes climatic
and edaphic components. Biotic structure includes species abundance, richness,
distribution, evenness, and composition, measured at the population, species, community,
ecosystem, or global scale. A species (for eukaryotic organisms) is defined by a common
morphology, genetic history, geographic range of origin, and ability to interbreed and
produce fertile offspring. A population consists of interbreeding groups of individuals of
the same species that occupy a defined geographic space. Interacting populations of
different species occupying a common spatial area form a community (Barnthousc ct al..
2008). Community composition may also define an ecosystem type, such as a pine forest
or a tall grass prairie. Pollutants can affect the ecosystem structure at any of these levels
of biological organization (Suteretal.. 2005).
Individual plants or animals may exhibit changes in metabolism, enzyme activities,
hormone function, or may suffer gross lesions, tumors, deformities, or other pathologies.
However, only some organism-level endpoints affected by pollution, such as growth,
survival, and reproductive output, have been definitively linked to effects at the
population level and above (U.S. EPA. 2013). Population-level effects of pollutants
include changes over time in abundance or density (number of individuals in a defined
area), age or sex structure, and production or sustainable rates of harvest (Barnthouse et
al.. 2008). Community-level attributes affected by pollutants include species richness,
species abundance, composition, evenness, dominance of one species over another, or
size (area) of the community (U.S. EPA. 2013). Pollutants may affect communities in
ways that are not observable in organisms or populations (Bartell. 2007). including
(1) effects resulting from interactions between species, such as altered predation rates or
competitive advantage; (2) indirect effects, such as reducing or removing one species
from the assemblage and allowing another to emerge (Petraitis and Latham. 1999); and
(3) alterations in trophic structure.
Alternatively, ecosystems may be defined on a functional basis. "Function" refers to the
suite of processes and interactions among the ecosystem components that involve energy
or matter. Examples include water dynamics and the flux of trace gases such as rates of
12

-------
photosynthesis, decomposition, nitrification, or carbon cycling. Pollutants may affect
biotic structure indirectly. For example, a pollutant may first alter abiotic conditions
(e.g., soil chemistry), which in turn influences biotic structure and function (Bartell.
2007).
Some ecosystems, and some aspects of particular ecosystems, are less vulnerable to
long-term consequences of pollutant exposure. Other ecosystems may be profoundly
altered if a single attribute is affected by pollution. Thus, spatial and temporal definitions
of ecosystem structure and function become essential factors in defining affected
ecosystem services and in determining CLs for certain pollutants, either as single
pollutants or in combination with other stressors.
The main causal determinations of this ISA (Section IS.2.3) are that N and S deposition
affect ecosystem structure, with effects ranging from biogeochemical alterations in soil
and water chemistry to multiple levels of biological organization, including species-level
alterations of physiological processes and shifts in biodiversity and ecological function.
IS.2.2.2 Deposition and Source Contribution of Nitrogen (N) and Sulfur (S) to
Ecosystems
Deposition of N and S results from a variety of human activities and atmospheric
processes. Emissions from stationary, mobile, and agricultural sources undergo
atmospheric transformation (Section IS.3.1) to form products that are eventually
deposited out of the air onto the land or waterscape (Section IS.3.3). The contribution of
atmospheric deposition to total loading for N and S varies within and among terrestrial,
wetland, freshwater, and estuarine ecosystems.
In the 2008 ISA, atmospheric deposition was identified as the main source of
anthropogenic N to unmanaged terrestrial ecosystems. This conclusion has been
confirmed by new studies on N sources to lands and waterways (Appendix 4.2). Across
all watersheds, atmospheric N deposition is the second largest overall human-mediated N
source; agriculture is the largest, and the largest N source to 33% of watersheds. Current
deposition levels in the U.S. are discussed in Appendix 2 and Section IS.3.3. No new
information has been published on nonatmospheric sources of S in terrestrial ecosystems
(Appendix 4.2); S inputs from the atmosphere are discussed in Appendix 2 and
Section IS.3.3.
In the 2008 ISA, atmospheric deposition was also identified as the main source of N to
some freshwater ecosystems, including headwater streams, high-elevation lakes, lower
order streams in undisturbed areas, and freshwater wetlands (e.g., bogs and fens).
Evidence for the influence of N deposition on water chemistry has been further supported
13

-------
by new studies that quantify the contribution of N deposition to total N loading in
freshwater lakes and streams, and which quantify atmospheric contributions during storm
events (Table 7-1). As shown in these studies, deposition can represent a substantial
portion of total N loading to surface waters. However, other nonpoint and point sources
of N dominate N inputs to high-order streams.
In fresh surface waters and wetlands, S that contributes to enrichment induces acidifying
effects. Sources of S include weathering of minerals in sediments and rocks, leaching
from terrestrial S cycling, internal cycling, and direct atmospheric deposition. The 2008
ISA showed that drought can release S stored in wetlands or lake sediments because
bound sulfide (S2 ) is exposed to atmospheric oxygen and oxidized to SO42 .Increases in
waterborne SO42 concentration through various concurrent processes has been observed
as a result of drought in whole-lake observational research (93% increase in Little Rock
Lake, WI, from 1.5 to 2.9 mg/L), and in response to variation in water levels from
climate change-induced droughts in modelling using Model of Acidification of
Groundwater in Catchments (MAGIC). New evidence confirms that fluctuating water
levels in wetlands increase SO42 concentration in pulses following water level recovery.
The importance of atmospheric deposition as a cause of estuarine eutrophication is
determined by the relative contribution of the atmospheric versus nonatmospheric sources
of N input. Sources of N in coastal areas may include direct deposition to the water
surface, coastal upwelling from oceanic waters, and transport from watersheds.
Freshwater inflows to estuaries often transport N from agriculture, urban, wastewater,
and atmospheric deposition sources. Atmospheric deposition constitutes less than half of
the total N supply in most, but not all, estuaries (Table 7-9). Both point sources and
nonpoint sources (including runoff, as well as atmospheric deposition) have been
identified as targets for mitigation of N loading in coastal areas. Seawater contains high
concentrations of SO42 , so atmospheric inputs of S are unlikely to contribute
substantially to biogeochemical or biological effects in coastal areas.
IS.2.2.3 Critical Loads Concept and General Approaches
The following section provides a discussion of important concepts regarding Critical
Loads (CLs). The definition of a CL is, "a quantitative estimate of an 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" (Nilsson and Grennfelt.
1988). This definition is intended as background material to support a better
understanding of the CL calculations presented throughout the ISA. The main concepts
presented here include CLs as an organizing principle, CL heterogeneity across the
14

-------
landscape, more than one CL for a given location, the pros and cons of methods used to
calculate CLs (e.g., empirical, steady state, and dynamic), and a comparison of CLs
versus target loads. Uncertainty in calculating CLs is discussed in Section IS. 13.
Throughout this ISA, the CL concept is used as an organizing principle to relate
atmospheric deposition to ecological endpoints that indicate impairment. The
development of a quantitative CL estimate requires a number of steps. An illustrative
example of the eight general steps is shown in Figure IS-3.
1) Disturbance
Acidification
Eutrophication
2) Receptor
Forest
Lake
Grassland
Lake
3) Biological
indicator
Sugar
Maple
Norway
Spruce
Brook trout
Fish species
richness
Species
diversity
Primary
productivity
4) Critical
biological
response
Failure to
reproduce
Seedling
death
Presence
absence
Species
loss
Species
loss
Excess
productivity
5) Chemical
indicator
Soil % Base
Saturation
Soil Ca/AI
ratio
Lakewater
ANC
Lakewater
ANC
Soil C/N
ratio
Lakewater
no3
6) Critical
chemical
limit
10%
1.0
0 peq/L
50 peq/L
20
10 peq/L
7) Atmospheric
pollutant
O
U>
S04, no3,
nh4
so4, no3,
nh4
so4, no3,
nh4
no3, nh4
no3, nh4
8) Critical
pollutant load
???
???
???
???
???
???
Al = aluminum; ANC = acid-neutralizing capacity; C = carbon; Ca = calcium; L = liter; peq = microequivalents; N = nitrogen;
NH4 = ammonium; N03 = nitrate; S04 = sulfate.
Source: U.S. EPA (2008).
Figure IS-3 An example of the matrix of information considered in defining
and calculating critical loads (see discussion in text). Note that
multiple alternative biological indicators, critical biological
responses, chemical indicators, and critical chemical limits could
be used.
It is important to recognize that there is no single "definitive" CL for an ecological effect.
CL estimates reflect the current state of knowledge and the selected limits, indicators, and
responses. Changes in scientific understanding may include, for example, new
dose-response relationships, better resource maps and inventories, larger survey data sets,
continuing time-series monitoring, and improved numerical models.
15

-------
Calculating multiple CLs for a given pollutant at a single location is not uncommon
because of the nested sequence of disturbances, receptors, and biological indicators
considered for a given pollutant. Multiple CL values may also arise from an inability to
agree on a single definition of "harm." Calculation of CLs for multiple definitions of
"harm" may be deemed useful in subsequent discussions of the analysis and in the
decision-making steps that may follow CL calculation.
The heterogeneity of natural environments can affect responsiveness of ecosystems to
deposition load. For example, the high spatial variability of soils almost guarantees that
for any reasonably sized soil-based "receptor" that might be defined in a CL analysis,
there will be a continuum of CL values for any indicator chosen. Although the range of
this continuum of values might be narrow, there is nevertheless an a priori expectation in
any CL analysis that multiple values (or a range of values) will result from the analysis.
Given the heterogeneity of ecosystems affected by N and S deposition, published CL
values for locations in the U.S. vary depending on both biological and physical factors.
The three approaches to developing CLs (i.e., empirical observation, steady-state
modeling, and dynamic modeling) each have strengths and limitations. It is suggested
that the combined approach of calculating CLs from biogeochemical simulation models
in conjunction with empirical analyses is the most effective way to characterize the
effects of deposition to a given environment (Fcnn et al.. 2015V For all three types of
models, spatial boundaries of where to apply a CL are important. For example, a CL may
apply to a watershed, ecoregion, or species range, depending on how the CL is defined.
An important advantage of empirical CLs is that they are based on measured
(vs. modeled) changes in ecological variables in response to inputs. Consequently, the
links between deposition and the measured response variable are direct; full process-level
knowledge is not required. Empirical CLs are important for validating CL values
determined with models (Tenn et al.. 2015).
Fenn et al. (2015) discussed that the advantages of models, "are that ecosystem responses
to alternative scenarios can be tested. These might include changes in atmospheric
deposition, disturbance or climatic conditions, and responses to silvicultural treatments,
grazing, fire, and other disturbances. Simulation modeling allows temporal aspects of
ecosystem response in relation to CLs and CL exceedances to be evaluated, including
evaluation of historical and future conditions."
Two key ways that steady-state and dynamic models differ in their modeling of CLs is by
how they assume ecosystem equilibrium and by the amount of input data they need for
parameterization.
16

-------
Steady-state models assume that the ecosystem is in equilibrium with the CL of
deposition; therefore, the long-term sustainable deposition, that is the CL, is indicated.
This is the relevant information needed to provide protection from deposition in
perpetuity as the system comes into equilibrium with the pollutant CL. In the U.S., few
(if any) ecosystems qualify as steady-state systems. Therefore, the assumption of
equilibrium in the steady-state model is often false. The steady-state models give no
information concerning the time to achieve the equilibrium or what may happen to the
receptor along the path to equilibrium. The recovery of an ecosystem based on a CL from
a steady-state model may take several hundred years. In other words, the assumption that
attainment of a deposition value below the steady-state CL will result in biological
recovery within a specified time period may not be valid. Dynamic models calculate
time-dependent CLs and, therefore, do not assume an ecosystem that is in equilibrium.
The time-dependent calculation is relevant information to provide protection from
damage by the pollutant within a specific time frame. Generally, the shorter the time
frame selected, the lower the CL.
Data requirements for steady-state models tend to be much lower than for dynamic
models. Therefore, the data required to conduct dynamic modeling are not available for
as many places as the data required to conduct steady-state modeling. The few
national-scale modeling efforts for both terrestrial and aquatic acidification are both done
with steady-state models for this reason.
The results of all three CL approaches are difficult to extrapolate across geographic
space. Spatially, variation in biological and biogeochemical processes imposed by
climate, geology, biota, and other environmental factors may alter the
deposition-response relationship. Empirical CLs may only be applied with confidence to
sites with highly similar biotic and environmental conditions (Pardo et al.. 201 la). This is
particularly problematic in areas where deposition has received sparse research
attention—as is sometimes the case for CLs of N deposition related to N driven
eutrophication (Appendix 6.4). Models may be run at different locations, but the data
needed to parameterize them is not always available.
CLs are different from target loads. Fenn etal. (2011) defined the "target load" as
follows: "The acceptable pollution load that is agreed upon by policy makers or land
managers. The target load is set below the CL to provide a reasonable margin of safety,
but could be set higher than the CL at least temporarily." Target loads are selected based
on the level of ecosystem protection desired, economic considerations, and stakeholder
input at a given location.
17

-------
IS.2.2.4
The Importance of Biodiversity
There are causal relationships between additions of N and/or S to an ecosystem and
biodiversity loss in terrestrial, freshwater, wetland, and estuarine ecosystems in the U.S.
(Table IS-1). What does it mean to lose biodiversity? Biodiversity loss not only means
the extirpation of unique living species, it represents the potential loss of ecosystem
function and ecosystem services, as shown by several decades of research in a wide
variety of natural svstcms(Hoopcr et al.. 2012; Balvanera et al.. 2006; Tilman. 2000).
Numerous studies demonstrate that the number and diversity of organisms in a system
control the abundance of habitat for other species, the biogeochemical cycling of
nutrients and carbon, and the efficiency at which biotic systems are able to transform
limited resources into biomass (Cardinale et al.. 201IV Among plant communities, higher
biodiversity leads to higher overall plant productivity and greater retention of soil
nutrients (Reich et al.. 2012; Tilman. 2000). In multitrophic systems, higher prey
diversity leads to both higher predator growth rates and a smaller impact of predation on
prey abundance (Duffy et al.. 2007). Positive impacts of biodiversity on ecosystem
services have been documented in forests (Gamfeldt et al.. 2013; Zhang et al.. 2012).
grasslands (Tilman et al.. 2012). arid and semiarid ecosystems (Maestre et al.. 2012). and
marine systems (Gamfeldt et al.. 2015; Worm et al.. 2006) and include effects such as
greater carbon storage, fruit production, wood production, and nutrient cycling. In marine
ecosystems, biodiversity loss has been linked to increased rates of exponential decreases
in water quality through metrics such as higher numbers of beach closures and harmful
algal blooms [HABs; Worm et al. (2006)1. Notably, HABs are linked to increased disease
prevalence among humans, domestic animals/pets, and aquatic organisms (Johnson et al..
2010). In addition to the relationship between HABs and disease, there is now empirical
evidence from many ecosystems of a broader link between declines in biodiversity and
increased transmission and severity of disease (Johnson et al.. 2015) caused by plant,
wildlife, and human pathogens. As a whole, these decades of research have produced an
overwhelming body of evidence indicating that the loss of biodiversity risks a
deterioration of the ecosystem goods and services on which humanity depends on
(Gamfeldt et al.. 2015; Cardinale et al.. 2012).
One of the most important consensus observations in biodiversity research is that
ecosystem processes are more stable (have less temporal variability) at higher levels of
diversity (Cardinale et al.. 2012; McCann. 2000; Naeem and Li. 1997; Tilman and
Downing. 1994). This stability occurs because species respond differently to
environmental variation. In diverse communities, it is more likely that declines in the
growth of one species caused by an environmental change will provide more resources
for competing species (Cardinale et al.. 2012; Tilman. 2000). This property was predicted
by economists and is similar to how more diversified investment portfolios provide
18

-------
enhanced stability under fluctuating market conditions (Doak et al.. 1998; Tilman ct al..
1998). Notably, there is also consensus that the impact of biodiversity on ecosystem
processes is nonlinear, wherein declines in ecosystem processes accelerate as the number
of species in a system declines (Card in ale et al.. 2012V Accelerating ecosystem service
declines in response to species loss may be because different ecosystem functions require
the presence of different sets of species (Isbell et al.. 2015; Reich et al.. 2012; Zavaleta et
al.. 2010). The increased stability of diverse ecosystems makes these systems less
vulnerable to environmental change or collapse caused by external forces such as drought
or human disturbance (Isbell et al.. 2015; Tilman et al.. 2012; Isbell etal.. 2011; Worm et
al.. 2006). For example, coastal systems with higher species diversity had lower rates of
fishery collapse and extinction for commercially important fish and invertebrate species,
and large marine ecosystems with higher fish diversity recovered more quickly from
collapse (Worm et al.. 2006). Thus, there is strong evidence that high biodiversity helps
sustain ecosystem services and makes these ecosystem services more resilient to
environmental change.
IS.2.2.5 Reduced versus Oxidized Nitrogen Effects across Ecosystems
Individual biochemical and geochemical processes involve specific chemical forms of N,
suggesting that there may be consequences in many ecosystems from the ongoing trend
of decreasing NOy deposition and increasing NHx deposition in many parts of the U.S.
(Section IS.3). The largest body of evidence that the effects of reduced versus oxidized N
may have different consequences for ecological structure and function is for estuaries
where the form of N delivered to some coastal areas of the U.S. is shifting from primarily
N03 to an increase in reduced forms of N. Although unlikely to be attributed solely to
atmospheric sources due to the large contribution of N from wastewater, agriculture, and
other sources, inputs of ammonia (NH3) and NH4 selectively favor specific
phytoplankton functional groups (e.g., cyanobacteria, dinoflagellates) including harmful
species (Figure 10-7). Shifts in phytoplankton community composition to species that
respond strongly to reduced N have been observed in some coastal regions
(Appendix 10.3.2). Growth of some species of phytoplankton (Appendix 10.2.2) and
macroalgae (seaweed; Appendix 10.2.3) appear to be related to the form of N. There is
also increasing evidence in freshwater systems for the importance ofN in harmful algal
blooms (HABs), and several studies have shown that the form of N influences freshwater
algal species composition (Appendix 9.2.6.1). In terrestrial systems, oxidation-reduction
status of inorganic N seems to have little influence on the biological responses to N
deposition (Appendix 4.3.12).
19

-------
Because some soil biogeochemical processes involve specific chemical forms of N
(e.g., denitrification, ammonium toxicity), there is the potential that biological responses
to N deposition (or N addition) could depend on whether the dominant form of deposited
N is oxidized (NOy) or reduced (NHx). Different responses to individual forms of N have
been observed for some soil biogeochemical processes (Table 4-13) and terrestrial
biological responses (Table 6-1). Moreover, a number of individual studies have
observed differential effects of NH4+ versus NO3 additions on plant community diversity
[e.g., Kleiin et al. (2008); Dias et al. (2014)1. In general, however, meta-analyses in the
literature have tended to find no difference in the effects of individual forms of N on
terrestrial biological endpoints like plant productivity or microbial biomass (Table 6-1).
This result suggests that terrestrial community diversity is also generally not
differentially affected by the form of N, possibly because plant uptake of N is mediated
by soil biogeochemical cycles that often rapidly transform N between oxidized and
reduced forms.
Evidence of wetland responses to different chemical forms of N come primarily from N
addition experiments conducted outside of the U.S. In European bogs and fens, both
forms of N addition decreased ecosystem N retention, but oxidized N addition caused
dissolved organic nitrogen (DON) leaching, while reduced N caused dissolved inorganic
nitrogen (DIN) leaching as well as cation leaching (Appendix 11.3.1.6). Reduced N
caused greater physiological stress or injury than equivalent loads of oxidized N in moss
species (Appendix 11.4.5 and Appendix 11.5.5).
IS.2.2.6 Aquatic Acidification Index (AAI)
The 2017 IRP (U.S. EPA. 2017a) described the Aquatic Acidification Index (AAI) to be
a novel approach for a multipollutant standard intended to address deposition-related
effects. Scientifically, the AAI represented an advancement in ecological methodology to
(1) calculate CLs for aquatic acidification on a national scale, when previously CLs had
been calculated on the spatial scale of a watershed and (2) provide a uniform level of
ecological protection at the national scale. These advancements were accomplished by
first aggregating CLs calculated for the same chemical limit within a defined spatial
region. Next, the distribution of the "population" of CL values was evaluated, and the
percentage of water bodies to protect was selected as a potential method to evaluate
different conservation targets. The AAI also presented novel advancements in
atmospheric sciences, including (1) using transference ratios to relate atmospheric
concentrations of criteria pollutants to deposition levels and (2) allowing quantification of
criteria pollutants (NOy and SOx) and noncriteria pollutant (e.g., NHX) contributions to
total acidifying deposition. As a scientific publication, the AAI is documented in Scheffe
20

-------
et al. (2014). The AAI was originally developed in the 2011 NOxSOx Policy Assessment
(U.S. EPA. 2011). and the equation is described with terms that traditionally define a
NAAQS [the indicator,1 averaging time,2 form,3 and level4—further described in the
2017 IRP (U.S. EPA. 2017aYI.
Key scientific aspects of the AAI equation, as the form of a potential standard, are
described in the following excerpt from 2017 IRP (U.S. EPA. 2017a):
"The AAI, as described in the PA (U.S. EPA. 2011). was constructed
from steady-state ecosystem modeling, and included atmospheric
transference ratios and deposition of reduced forms of nitrogen
(ammonia gas and ammonium ion, expressed as NHx). These
nonoxidized forms of nitrogen were included since ecosystems respond
to total nitrogen deposition, whether from oxidized or reduced forms.
More specifically, the AAI equation was defined in terms of four
ecological and atmospheric factors and the ambient air indicators NOy
and SOx:
AAI = F1 - F2 - /3[NOy] - /^[SOx]
Equation IS-1
where Fl5 represents the ecosystems natural ability to provide
acid-neutralizing capacity (e.g., geology, plant uptake of nitrogen
deposition) and other processes; F26 represents acidifying deposition
associated with reduced forms of nitrogen, NHx; and F31 and l'F are the
1	The "indicator" of a standard defines the chemical species or mixture that is measured in determining whether an
area attains the standard.
2	The "averaging time" defines the time period over which ambient measurements are averaged (e.g., 1-hour, 8-hour,
24-hour, annual).
3	The "form" of a standard defines the air quality statistic that is compared to the level of the standard in determining
whether an area attains the standard.
4	The "level" defines the allowable concentration of the criteria pollutant in the ambient air.
5	Fl is defined as: /I.Y<"illn +	with .lAT'i,,,, representing a target ANC level. With regard to the PA
developed distributions of calculated critical loads for a specific ecoregion; in setting an AAI-based standard, a
percentile would need to be specified to reference the value of CLr to be used in the AAI equation [U.S. EPA (2011).
p. 7-37], The PA described the percentile as an aspect of the form for the standard [U.S. EPA (2011). Section 7.7],
6	F2 is defined as: NHwhere NHX is the deposition divided by O, [U.S. EPA (2011). p. 7-37],
7	F3 is defined as: 7NO. where 7NOy is the transference ratio that converts deposition of NOy to ambient air
concentrations of NOy [U.S. EPA (2011). p. 7-37],
8	F4 is defined as: 7:SOv/
-------
transference ratios that convert concentrations of NOy and SOx to related
deposition of nitrogen and sulfur ITJ.S. EPA (201 1). Section 7.7]."
Several other key scientific considerations are included in the AAI that were discussed in
the 2011 NOxSOx Policy Assessment (U.S. EPA. 2011).
•	Spatial heterogeneity of factors in the AAI equation: The value of factors in the
AAI equation vary across the U.S. Factors could be calculated for a spatial
boundary based on an ecologically similar landscape (e.g., Omernick ecoregion).
•	Temporal heterogeneity: There is a relatively high degree of interannual
variability expected in the AAI because it is so strongly influenced by the amount
and pattern of precipitation that occurs within a region from year to year;
therefore, averaging calculated annual AAI values over 3 to 5 years would provide
reasonable stability.
•	Level: With regard to a level for the AAI, the 2011 NOxSOx Policy Assessment
(U.S. EPA. 2011) concluded that consideration should be given to a level within
the range of 20 to 75 (j,eq/L, noting that a target Acid Neutralizing Capacity
(ANC) value of 20 |icq/L would be a reasonable lower end of this range, so as to
protect against chronic acidification-related adverse impacts on fish populations
which have been characterized as severe at ANC values below this level.
IS.2.3 Changes: New Evidence and Causal Determinations
Since the 2008 ISA, several conceptual changes have occurred in our understanding of
the atmospheric sciences and ecological effects of NOx, SOx and PM. They include our
understanding of the sources of N deposition and in the relationship between atmospheric
concentration and deposition (Section IS.3 and Appendix 2). Models of N deposition rely
on accurate emissions data. Since the 2008 ISA, deposition of oxidized nitrogen has been
decreasing but deposition of reduced nitrogen has been increasing. As a result, the
uncertainty in total reactive N emissions (NOx + NHx) has increased because emissions
estimates that have the lowest levels of uncertainty are from stationary and mobile
sources, which contribute more to NOx than NHx emissions, and higher levels of
uncertainty are associated with agricultural emissions, which contribute more to NHx
than NOx emissions.
A better understanding of the relationship between atmospheric concentration and
deposition has resulted from advances in understanding bidirectional exchange of NH3
and NOy chemistry within canopies. These advances have led to the first efforts to
provide a detailed characterization of N and S deposition on a national scale, by using
both measured and modeled values to provide estimates of total sulfur and nitrogen
deposition across the U.S.
22
-------
New evidence since the 2008 ISA increases the weight of evidence for ecological effects,
confirming concepts previously identified and improving quantification of dose
(deposition)-response relationships, particularly for N deposition. The ecological effects
are described by the causality determinations. There are 18 causality statements in this
ISA (Table IS-1). Fourteen are causal relationships repeated from the 2008 ISA or
modified from the 2008 ISA to include specific endpoints. One is a likely causal
relationship repeated from the 2009 PM ISA. Three are new endpoint categories not
evaluated in the 2008 ISA. Table IS-3 shows that N and S deposition cause alteration of
(1) biogeochemical components of soil and water chemistry and (2) multiple levels of
biological organization ranging from physiological processes to shifts in biodiversity and
ecological function (Figure IS-4).
The current NO2 and SO2 secondary NAAQS are set to protect against direct damage to
vegetation by exposure to gas-phase oxides of nitrogen and oxides of sulfur. Research
continues to support causal relationships between SO2, NO2, NO, peroxyacetyl nitrate
(PAN), HNO3, and injury to vegetation (Table IS-1). but research that tests plant response
to the lower exposure levels that represent current atmospheric NOy and SOx
concentrations is limited. Therefore, little evidence is available to help determine whether
current monitored concentrations of gas-phase NOy and SOx are high enough to injure
vegetation.
It is clear that the criteria pollutants NOy, SOx, and PM, in addition to the noncriteria
pollutant NH3, contribute to total N and S deposition, which alters the biogeochemistry
and the physiology of organisms, resulting in harmful declines in biodiversity. Decreases
in biodiversity mean that some species become relatively less abundant and may be
locally extirpated. The current period in Earth's history is the Anthropocene. In addition
to a spike in soil radiocarbon from nuclear bomb testing (Turnev et al.. 2018). a defining
attribute of the Anthropocene is global, human-driven mass extinctions of many species.
The biodiversity loss reported in this assessment contributes to the Anthropocene loss of
biodiversity (Rockstrom et al.. 2009). In addition to the loss of unique living species, the
decline in total biodiversity is harmful because biodiversity is an important determinant
of the stability of ecosystems and the ability of ecosystems to provide services to
humanity (see more on biodiversity in Section IS.2.2.4V
23
-------
Table IS-1 Causal determinations for relationships between criteria pollutants
and ecological effects from the 2008 NOx/SOx Integrated Science
Assessment (ISA) or the 2009 Particulate Matter (PM) ISA, for other
effects of PM, and the current draft ISA.

Causal Determination
Effect Category
2008 NOx/SOx ISA
Current ISA
Gas-phase direct phytotoxic effects
Gas-phase SO2 and injury to vegetation
Causal relationship
Causal relationship
Section IS.3 and ADDendix 3.6.1


Gas-phase NO, NO2, and PAN and injury to vegetation
Causal relationship
Causal relationship
Section IS.3 and ADDendix 3.6.2


Gas-phase HNO3 and injury to vegetation3
Causal relationship
Causal relationship
Section IS.3 and ADDendix 3.6.3


N and acidifying deposition to terrestrial ecosystems
N and S deposition and alteration of soil biogeochemistry
in terrestrial ecosystems'5
Causal relationship
Causal relationship
Section IS.5.1 and ADDendix 4.1


N deposition and the alteration of the physiology and
growth of terrestrial organisms and the productivity of
terrestrial ecosystems0
Not included
Causal relationship
Section IS.5.2 and ADDendix 6.6.1


N deposition and the alteration of species richness,
community composition, and biodiversity in terrestrial
ecosystems0
Causal relationship
Causal relationship
Section IS.5.2 and ADDendix 6.6.2


Acidifying N and S deposition and the alteration of the
physiology and growth of terrestrial organisms and the
productivity of terrestrial ecosystemsd
Not included
Causal relationship
Section IS.5.3 and ADDendix 5.7.1


Acidifying N and S deposition and the alteration of
species richness, community composition, and
biodiversity in terrestrial ecosystemsd
Causal relationship
Causal relationship
Section IS.5.3 and ADDendix 5.7.2


N and acidifying deposition to freshwater ecosystems
N and S deposition and alteration of freshwater
biogeochemistrye
Causal relationship
Causal relationship
Section IS.6.1 and ADDendix 7.1.7


24
-------
Table IS-1 (Continued): Causal determinations for relationships between criteria
pollutants and ecological effects from the 2008 NOx/SOx
Integrated Science Assessment (ISA) or the 2009
Particulate Matter (PM) ISA, for other effects of PM, and
the current draft ISA.

Causal Determination
Effect Category
2008 NOx/SOx ISA
Current ISA
Acidifying N and S deposition and changes in biota,
including physiological impairment and alteration of
species richness, community composition, and
biodiversity in freshwater ecosystems'
Causal relationship
Causal relationship
Section IS.6.3 and ADDendix 8.6


N deposition and changes in biota, including altered
growth and productivity, species richness, community
composition, and biodiversity due to N enrichment in
freshwater ecosystems9
Causal relationship
Causal relationship
Section IS.6.2 and Appendix 9.6


N deposition to estuarine ecosystems
N deposition and alteration of biogeochemistry in
estuarine and near-coastal marine systems
Causal relationship
Causal relationship
Section IS.7.1 and Appendix 7.2.10


N deposition and changes in biota, including altered
growth, total primary production, total algal community
biomass, species richness, community composition, and
biodiversity due to N enrichment in estuarine
environments11
Causal relationship
Causal relationship
Section IS.7.2 and Appendix 10.7


N deposition to wetland ecosystems
N deposition and the alteration of biogeochemical cycling
in wetlands
Causal relationship
Causal relationship
Section IS.8.1 and Appendix 11.10


N deposition and the alteration of growth and productivity,
species physiology, species richness, community
composition, and biodiversity in wetlands
Causal relationship
Causal relationship
Section IS.8.2 and Appendix 11.10


S deposition to wetland and freshwater ecosystems
S deposition and the alteration of mercury methylation in
surface water, sediment, and soils in wetland and
freshwater ecosystems'
Causal relationship
Causal relationship
Section IS.9.1 and Appendix 12.7


25
-------
Table IS-1 (Continued): Causal determinations for relationships between criteria
pollutants and ecological effects from the 2008 NOx/SOx
Integrated Science Assessment (ISA) or the 2009
Particulate Matter (PM) ISA, for other effects of PM, and
the current draft ISA.
Causal Determination
Effect Category	2008 NOx/SOx ISA	Current ISA
S deposition and changes in biota due to sulfide	Not included	Causal relationship
phytotoxicity, including alteration of growth and
productivity, species physiology, species richness,
community composition, and biodiversity in wetland and
freshwater ecosystems
Section IS.9.2 and Appendix 12.7
2009 PM ISA	Current Draft ISA
Other ecological effects of PM (course and fine particles, without regard to chemical speciation)
PM and a variety of effects on individual organisms and Likely to be a causal	Likely to be a causal
ecosystems	relationship	relationship
Section IS. 10 and Appendix 15.7
C = carbon; Hg = mercury; HN03 = nitric acid; ISA = Integrated Science Assessment; N = nitrogen; NO = nitric oxide;
N02 = nitrogen dioxide; PAN = peroxyacetyl nitrate; S = sulfur; S02 = sulfur dioxide.
aThe 2008 ISA causality statements for gas-phase HN03 was phrased as "changes in vegetation."
bThe 2008 ISA included two causality statements for terrestrial biogeochemistry which were phrased as "relationship between
acidifying deposition and changes in biogeochemistry" and "relationship between N deposition and the alteration of
biogeochemical cycling of N."
The 2008 ISA causality statement for biological effects of N enrichment in terrestrial ecosystems was phrased as "relationship
between N deposition and the alteration of species richness, species composition, and biodiversity."
dThe 2008 ISA causality statement for biological effects of acidifying deposition in terrestrial ecosystems was phrased as
"relationship between acidifying deposition and changes in terrestrial biota."
eThe 2008 ISA included three causality statements for freshwater biogeochemistry phrased as "relationship between acidifying
deposition and changes in biogeochemistry related to aquatic ecosystems," "relationship between N deposition and the alteration
of biogeochemical cycling of N," and "relationship between N deposition and the alteration of biogeochemical cycling of C."
'The 2008 ISA causality statement for biological effects of acidifying deposition in freshwater ecosystems was phrased as,
"relationship between acidifying deposition and changes in aquatic biota."
9The 2008 ISA causality statement for biological effects of N deposition in freshwater ecosystems was phrased as "relationship
between N deposition and the alteration of species richness, species composition, and biodiversity in freshwater aquatic
ecosystems."
hThe 2008 ISA causality statement for biological effects of N deposition to estuaries was phrased as "relationship between N
deposition and the alteration of species richness, species composition, and biodiversity in estuarine ecosystems."
'The 2008 ISA causality statement for biological effects of S deposition effects on ecosystems was phrased as "relationship
between S deposition and increased methylation of Hg, in aquatic environments where the value of other factors is within
adequate range for methylation."
26
-------
NOx SOx PM Integrated Science Assessment for Ecological Effects*
Indicator
Gases * Nitrogen Deposition Sulfur Deposition ^'^Depositon11^
Class of Pollutant Effect
Direct
Phytotoxic N-enrichment/Eutrophication Sulfide Toxicity Mercury Methylation Acidification
Scale of Ecological Response
Population
Geochemistry Individual 	 Community Ecosystem
Individual
Ecosystem
Productivity
Biodiversity
Growth rate
Physiological
alteration, stress
or injury
Soil or sediment
chemistry
Surface water
chemistry
Terrestrial Terrestrial Wetland Fresh Water Estuary Wetland Fresh Water Wetland Fresh Water Terrestrial Fresh Water
Causal
Not likely
Inadequate
Suggestive
Likely causal
Not evaluated in causal framework
Causality framework
*	A causal relationship is likely to exist between deposition of PM and a variety of effects on individual organisms and ecosystems, based
on information from the previous review and limited new findings in this review
*	Includes: NO, N02, HN03, S02, and PAN
Figure IS-4 Causal relationships between the criteria pollutants and ecological effects.
27
-------
Since the 2008 ISA, there is more evidence to support the direct effects of gaseous SOx
and NOy on vegetation. This causality determination is uniquely modified by the
observation that there is little or no evidence that such effects are continuing at current,
lower levels of exposure now occurring in the U.S.
Since the 2008 ISA, the largest increase in ecological evidence is for terrestrial N driven
eutrophication effects (Section IS.5.1. Section IS.5.2. Appendix 4, and Appendix 6). This
new research confirms the causal relationship between N deposition and ecological
effects documented in the 2008 ISA. Further, this new research improves our
understanding of the mechanistic links that inform causal determinations between N
additions via atmospheric deposition, biogeochemistry, and biota in terrestrial ecosystems
(Table IS-1). There is now stronger empirical evidence from across most regions of the
U.S. to quantify the levels of N deposition (empirical CLs) that cause biodiversity
declines of lichens and grasses/forbs. There is new evidence to quantify empirical CLs
across much of the U.S. for nitrate leaching, tree survivorship, and mycorrhizal
biodiversity. Many of the N deposition effects are due to historical and continuing N
deposition.
New research confirms that N + S deposition causes terrestrial ecosystem acidification, as
documented in the 2008 ISA (Table IS-1). New evidence to characterize terrestrial
acidification (soil biogeochemistry changes and biological effects) across large regions of
the U.S. is available; in particular, new modeling work has improved calculation of CLs
for soil acidification (Section IS. 5.3; Appendix 4 and Appendix 5). Many of the
acidification effects are due to historical and continuing N and S deposition
(Section IS. 11V
New evidence for freshwater acidification CLs builds on several decades of research
documenting freshwater acidification effects on aquatic biota in the U.S. and confirms the
causal relationships determined in the 2008 ISA (Table IS-1). Many of the acidification
effects are due to historical and continuing N and S deposition (Section IS. 11).
The sources of N driven eutrophication of fresh waters, estuaries, and wetlands include
atmospheric N deposition and N from agricultural and other wastewaters. New research
has helped show how these respective sources contribute to total loading. In freshwater
ecosystems where atmospheric deposition is the primary source of N, such as in high
alpine watersheds, new CLs since the 2008 ISA support previous observations of
increased algal productivity, species changes, and reductions in diversity. New evidence
also supports clear links between aqueous S concentrations in aquatic systems and both
mercury methylation and sulfide toxicity; however, quantitatively linking these outcomes
to atmospheric deposition remains a challenge.
28
-------
IS.3
Emissions and Atmospheric Chemistry
The atmospheric chemistry from emission to deposition discussed in this ISA is for the
criteria pollutants NOy, SOx, and PM. In addition to gas-phase indicators like SO2 and
NO2 used to monitor criteria pollutant trends, deposition of total N, total S, and total N +
S that accounts for a wider range of species is also a main focus.
A wide variety of N containing compounds (oxidized + reduced, and organic + inorganic)
contribute to wet and dry N deposition (Appendix 2.1). NHX (NHX = NH3 + NH/)
includes both the PM component NH44" and gas-phase NH3. The contribution of NH3 to
total observed inorganic N deposition may range from 19% in northwestern U.S.
locations to 63% in locations in the southwestern U.S. and is generally greater in the
summer than in the winter. Therefore, NH3 is discussed in the ISA along with NOy and
relevant PM components to better understand and compare their contributions to both wet
and dry N deposition. In addition, PM impacts discussed in this document are also mainly
focused on N and S containing species, which together usually make up a large fraction
of PM25 mass in most areas of the U.S. and have greater and better understood ecological
impacts than other PM components.
Gaseous, particulate, and dissolved forms of NOy, SOx, and NHx all contribute to
atmospheric wet and dry deposition. The major components of particulate matter in the
U.S. are NO3 , SO42 , NH4+, particulate organic matter, elemental carbon, crustal
material, and sea salt. While organic matter usually accounts for a large fraction of PM2 5,
only a small portion can be identified at a molecular level. As a result, there is little
information on organic PM impacts, except for individual compounds that make minor
contributions to mass. Assessment of ecological impacts of major PM species is largely
limited to NO3 , SO42 . and NH4+. Of these, SO42 and NO3 are also components of total
oxides of sulfur and nitrogen, respectively. NO3 , SO42 . and NH4+ usually have a strong
influence on acid deposition. NO3 and NH4+, and in some cases organic nitrogen
(organic nitrates and reduced organic N), make a substantial contribution to N deposition.
Since the 2008 ISA, there have been several new developments including:
•	Expansion of ambient monitoring networks to include NH3 and NOy at selected
sites, and comparisons of monitoring methods with research grade instruments
(Appendix 2.4);
•	Adoption of new methods, such as data-model fusion, to integrate deposition
information across the U.S. (Appendix 2.5);
•	Incorporation of bidirectional exchange into models of dry deposition
(Appendix 2.5.2); and
29
-------
• Improvements in techniques using satellite-based measurements and chemical
transport model simulations to estimate emissions, concentrations, and dry
deposition of NO2, SO2, and NH3 (Appendix 2.6).
IS.3.1 Sources and Atmospheric Transformations
Both gaseous and particulate forms of N and S contribute to atmospheric deposition. The
main contributors to acidifying precipitation are H2SO4, HNO3, and NH/, which are
formed from precursor emissions of SO2, NOx (NO + NO2), and NH3 (Appendix 2.2).
Gaseous emissions of NH3 are dominated by agricultural fertilizer application and animal
waste from intensive animal feeding operations, with important local contributions from
motor vehicles and episodic contributions from wildland and agricultural fires. Roughly
half of SO2 emissions are from by electricity-generating units (EGUs), mainly coal-fired
power plants. Notably, SO2 emissions from EGUs have been decreasing. NOx emissions
have a wider distribution of sources, with substantial contributions from highway and
off-highway vehicles, lightning, and EGUs. Primary PM2 5 and PM10 emissions are
dominated by dust and fires, but much of the PM25 mass in the U.S. is produced by
reactions that form secondary PM2 5 from gas phase precursor N and S species. Because
of these processes, a sharp decrease in SO2 emissions and smaller, but substantial
decreases in NOx emissions have occurred since the passage of the Clean Air Act
Amendments in 1990. Emissions of NOx in the U.S. declined 61% between 1990 and
2017 (U.S. EPA. 2020). while nationwide annual average 98th percentile NO2
concentrations decreased by 53% from 1990 to 2017 (U.S. EPA. 2016c). Total emissions
of SO2 decreased by 89% from 1990 to 2017 (U.S. EPA. 2020). resulting in a decrease in
SO2 concentrations of 89% in the eastern U.S. and 45% in the western U.S.
(Appendix 2.6.5). National annual NH3 emissions have fluctuated as a result of changes
in both emissions and methods of estimating emissions. However, no clear trend is
evident for national NH3 emissions, with estimates for 1990 and 2017 differing by less
than 1% (U.S. EPA. 2020). National NH3 monitoring is too recent for evaluating
long-term concentration trends, although more limited studies of NH3 emissions,
concentrations and deposition each suggest slight increases may have occurred
(Appendix 2.6.4).
Major components of particulate N and S include NH/, NO3 . and SO42 . which are
primarily derived from gaseous precursors NH3, NOx, and SO2 (Appendix 2.3). Together,
NO3 , SO42 . and NH4+ make up a large fraction of PM25 mass in most areas of the U.S.
Formation of particulate N and S is described in the 2019 ISA for Particulate Matter
(U.S. EPA. 2019). An understanding of the sources, chemistry, and atmospheric
processes for these gas-phase and PM species provides a background for understanding
acidifying and N deposition.
30
-------
IS.3.2
Measurement and Modeling Techniques
Monitoring networks across the U.S. measure NOy, SOx, and NHx species involved in
deposition (Appendix 2.4.1). The National Atmospheric Deposition Program/National
Trends Network (NADP/NTN) has monitored precipitation chemistry for several decades
at many U.S. sites. The Clean Air Status and Trends Network (CASTNET) has monitored
concentrations of inorganic gas and particulate-phase N and S species since 1990.
Monitoring of NH3 (Appendix 2.5.3) in the Ammonia Monitoring Network (AMoN), part
of the NADP network, was initiated at a subset of CASTNET sites in 2007. NH3 was also
measured as a part of the Southern Aerosol Characterization (SEARCH) network from
2004 until its termination in 2016. The Interagency Monitoring of Protected Visual
Environments (IMPROVE) network and the Chemical Speciation Network (CSN)
measure PM and PM components, including NO3 and SO42 , although these data are not
routinely used to estimate deposition rates (Appendix 2.4.1).
Atmospheric N deposition rates are calculated from measurements and models. Direct
measurement of NO2 concentration has limited utility for quantifying NOy deposition
rates in areas with less urban influence. Because NOy is composed of diverse chemical
species with a wide range of deposition velocities and physical properties, concentrations
of unmeasured component species of NOy in general and of all NOy species in
data-sparse regions must be provided by regional models. For NO2 and NH3 this can be
done in conjunction with satellite-based remote sensing data (Appendix 2.4.2).
Estimates of dry deposition (Appendix 2.5.2) over the contiguous U.S. are inferred by
atmospheric models, used with monitoring network data. When combined with accurate
estimates of historical trends in emissions and meteorology, these models are able to
capture the historical long-term changes in PM2 5 SO42 . NO3 . and NH4+, but are subject
to uncertainties in their treatment of turbulence, surface interactions, and in particular,
seasonal variability in NO3 deposition, mainly because of uncertainties in NH3
emissions. Consequently, dry deposition rates (and ratios of wet-to-dry deposition)
continue to be uncertain.
IS.3.3 Spatial and Temporal Variability in Deposition
Overall deposition of total N (oxidized + reduced N) has decreased slightly over the past
since 2000 (Appendix 2.6.2). This is because although NOy deposition has declined
considerably in the contiguous U.S., deposition of NHx has increased. The large spatial
variability in N deposition and changes in geographic distribution of 3-year average N
deposition between 2000-2002 and 2016-2018 are evident in the maps (Figure IS-5) of
31
-------
3-year average annual dry + wet deposition of NOy and NHX over the contiguous U.S.
estimated using the TDEP (Total Deposition) modeling approach (Appendix 2.6), which
combines output from the Community Multiscale Air Quality (CMAQ) system with wet
deposition from the NADP/NTN (Schwcdc and Lear. 2014) and air concentrations from
CASTNET.
According to TDEP estimates for 2016-2018 (Appendix 2.6), much of the eastern
contiguous U.S. is estimated to receive at least 10 kg N/ha/yr dry + wet deposition, with
some areas receiving more than 15 kg N/ha/yr. Estimates for the spatial extent of the
areas receiving at least 10 kg N/ha/yr of deposition and the overall amount of N deposited
could be low because reduced organic N species are not routinely monitored.
In general, wet deposition of reduced N exceeds that of oxidized N across the contiguous
U.S. According to estimates based on CASTNET and NADP data and CMAQ modeling
results (Figure 2-16), deposition of N nationwide occurs mainly by dry deposition of
HNO3 and NH3 (with NH3 dominant) and wet deposition of NH4+ and NO3 (with NH4+
dominant). Hybrid satellite/modeling and CMAQ results indicate that dry deposition of
NO2 is also a nontrivial source of deposited N in many areas (Appendix 2.6.6). Over the
past 30 years, NADP/NTN data show that wet deposition of inorganic N
(oxidized + reduced) decreased in areas such as the Northeast but remained constant or
increased in areas such as the central U.S. (see Figure 2-18 in Appendix 2.6). Wet
deposition of total inorganic N has remained fairly constant over the past 30 years,
despite declines in NOx emissions, indicating that most of the increases in N wet
deposition seen today is of reduced inorganic N. Data for total (wet + dry) deposition are
available for a shorter time series than wet deposition, but show a similar increase in the
share of reduced N relative to oxidized N. Figure IS-6 shows reductions in TDEP 3-year
average oxidized N deposition over the contiguous U.S. between 2000-2002 and
2016-2018, while Figure IS-8 shows the decrease in reduced N deposition compared
between the same periods.
32
-------
Source: C ASTNTTT/CM AQ/N A DP
Tola] deposition of nitrogen 0002
USHPA02/19/19
Total N
(kg-N/ha)
1
-0

-2

-4

-6

-B

-10

-12

-14

-16

-18
I
->20
Source: CASTTnET/CMAQ/NADP
Total deposition of nitrogen 1618
USEPA 10/21/19
Total N
(kg-N/ha)
Ha = hectare; kg = kilogram; N = nitrogen.
Source; CASTNET/NTN/AMON/SEARCH. We acknowledge the Total Deposition (TDep) Science Committee of the National
Atmospheric Deposition Program (NADP) for their role in making the TDep data and maps available.
Figure IS-5 Wet plus dry deposition of total nitrogen over 3-year periods.
Top: 2000-2002; Bottom: 2016-2018.
33
-------
Total deposition of oxidized N 0002
(iSFPA oo/i 2/1 s
Source: CASTNET/CMAQ/NADP
Total oxN
(kg-N/ha)
Total oxN
(kg-N/ha)
Source: CASIWEIVCMAQ/NADP
Total deposition of oxidized N 1618
lisHi'A 10/21/19
OxN = oxidized nitrogen.
Source: CASTNET/CMAQ/NTN/AMON/SEARCH. We acknowledge the Total Deposition (TDep) Science Committee of the National
Atmospheric Deposition (NADP) Program for their role in making the TDep data and maps available.
Figure IS-6 Wet plus dry deposition of oxidized nitrogen over 3-year periods.
Top: 2000-2002; Bottom: 2016-2018.
34
-------
Source: CASTNET/CMAQ/NADP
Total reN
(kg-N/ha)

-0

-1

-2

-3

-4

-5
c

— D
-7

-8

-9

->10
Total deposition of reduced N 0002
USEPA 09/12/18
Total deposition of reduced N 1618
USEPA 10/21/19
Source: CASTNET/CMAQ/NADP
Total reN
(kg-N/ha)
-0

"1

-2

-3

-4

-5

-6

-7

-8
i
-9
I
->10
reN = reduced nitrogen.
Source: CASTNET/CMAQ/NTN/AMON/SEARCH. We acknowledge the Total Deposition (TDep) Science Committee of the National
Atmospheric Deposition Program (NADP) for their role in making the TDep data and maps available.
Figure IS-7 Wet plus dry deposition of reduced (inorganic) nitrogen over
3-year periods. Top: 2000-2002; Bottom: 2016-2018.
35
-------
For S deposition, wet deposition tends to dominate over dry deposition in large areas of
the contiguous U.S. However, in some regions mainly in the West, dry deposition of
mainly SO2 makes a greater contribution than wet deposition. Anthropogenic emissions
of S and subsequent deposition have declined markedly since the 1990s, with the most
pronounced declines in the eastern U.S. Currently, some of the highest values of total
(wet + dry) SOx deposition in the U.S. are in parts of the Ohio Valley region
(Figure 2-41). However, Figure IS-8 shows that TDEP 3-year average total S deposition
has decreased substantially between 2000-2002 and 2016-2018, especially in this
region.
Both N and S deposition contribute to acidification of ecosystems. The pH of rainwater
has increased markedly across the U.S. since 1990, coincident with decreases in the wet
deposition of nitrate and SO42 . However, there are still widespread areas affected by
acidifying precipitation, mainly in the eastern U.S. (see Appendix 2.6). Total acidifying
deposition (wet + dry N + S, expressed as H+ equivalents) fluxes for 2016 to 2018 ranged
from a few tenths of H+ keq/ha/yr overmuch of the western U.S. to over 1.5 H+ keq/ha/yr
in parts of the Midwest and the Mid-Atlantic regions, and in other isolated hotspots
surrounding areas of concentrated industrial or agricultural activity (Figure IS-9).
36
-------
Source: CASTNET/CMAQ/NADP
Total deposition of sulfur 0002
USEPA 09/12/18
Total S
(kg-S/ha)
i
-0

-2

-4

-6

-8

-10

-12

-14

-16

-18
1
->20
Total S
(kg-S/ha)
[i
-8
-10
r12
¦->20
of sulfur 1618
USEPA 10/21/19
Total deposition
Source: CASTNET/CMAQ/NADP
S = sulfur.
Source; CASTNET/CMAQ/NTN/AMON/SEARCH. We acknowledge the Total Deposition (TDep) Science Committee of the National
Atmospheric Deposition Program (NADP) for their role in making the TDep data and maps available.
Figure IS-8 Wet plus dry deposition of total sulfur over 3-year periods.
Top: 2000-2002; Bottom: 2016-2018.
37
-------
Total N+S
(keq/ha)

-0.0

-0.2

-0.4

-0.6

-0.8

-1.0

-1.2

-1.4

-1.6

-1.8
1
->2.0
Total N+S deposition 1618
USEPA 10/21/19
Source: CASTNET/CMAQ/NADP
eq. = equivalents; H+ = hydrogen ion; ha = hectare; N = nitrogen; S = sulfur; yr = year.
Source: NADP. Note: We acknowledge the Total Deposition (TDep) Science Committee of the National Atmospheric Deposition
Program (NADP) for their role in making the TDep data and maps available.
Figure IS-9 Total acidifying deposition of total oxidized nitrogen, reduced
nitrogen, and oxidized sulfur expressed as H+ equivalents per
hectare per year over the contiguous U.S. 2016-2018.
Dry deposition rates are a strong function of surface characteristics, which modify the
structure of surface layer turbulence and the resistance to uptake by vegetation
(Appendix 2.5.2). As a result, spatially aggregated estimates of dry deposition fluxes are
subject to uncertainty, in addition to uncertainties that are inherent in the measurement of
species concentrations and in the inference of dry fluxes (see Section IS. 13). Wet fluxes
are not directly influenced by surface characteristics (although orography affects
transport and precipitation) but are subject to smaller uncertainties in the measurement of
rainfall and chemistry.
IS.4 Gas-Phase Direct Phytotoxic Effects
New evidence supports the causal determinations made in the 2008 ISA regarding
gas-phase effects on vegetation, and there are no new causal statements for gas-phase
38
-------
effects. As in the 2008 ISA, the current ISA concludes that there are causal relationships
between SO2, NO2, NO, PAN, HNO3, and injury to vegetation. This determination is
based on consistent, coherent, and biologically plausible evidence (Appendix 3.2,
Appendix 3.3, and Appendix 3.4; Table IS-1V The clearest evidence for these
conclusions comes from studies available at the time of the 2008 ISA, but there have
been some additional studies since then. Most evidence on the direct effects of gaseous
NOy and SOx comes from controlled exposure studies across many species of vegetation.
Most controlled exposure studies over the past several decades have used concentrations
of gas-phase NOy and SOx above current ambient conditions observed in the U.S.
Relevant information is lacking on exposures and effects reflecting the more recent lower
pollutant conditions. Therefore, there is little evidence available to inform whether
current monitored concentrations of gas-phase NOy and SOx are high enough to injure
vegetation.
NH3 can also have direct phytotoxic effects if the uptake exceeds the ability of a plant to
detoxify and assimilate it. However, reduced N gases such as NH3 are not criteria air
pollutants or oxides of N and, therefore, are not the focus of this review of the gas-phase
effects. Direct damage from NH3 to foliage can occur on higher plants and effect
bryophytes and lichens. Declines in shrubs and lichens and changes in peat bogs have
been reported with NH3 exposure. Besides being potentially phytotoxic to vegetation,
NH3 exposure can lead to more N inputs into plants and ecosystems through foliage
uptake. Ammonia deposition that leads to N enrichment is an important consideration
when evaluating total N deposition. These N nutrient effects to vegetation are discussed
in Appendix 6.
IS.4.1 Sulfur Dioxide
In the 2008 ISA, evidence was sufficient to infer a causal relationship between exposure
to SO2 and injury to vegetation. The current secondary standard for SO2 is a 3-hour
average of 0.50 ppm, which is designed to protect against acute foliar injury in
vegetation. There has been limited research on acute foliar injury since the 1982 PM-SOx
Air Quality Criteria Document (AQCD), and there is no clear evidence of acute foliar
injury below the level of the current standard. The limited new research since 2008 adds
more evidence that SO2 can have acute negative effects on vegetation but does not
change conclusions from the 2008 ISA regarding the causal relationship between SO2
exposure and vegetation damage or the SO2 levels producing these effects (see
Appendix 3.1). Consistent with the 2008 ISA, the body of evidence is sufficient to infer
a causal relationship between gas-phase SO2 and injury to vegetation.
39
-------
Increased SO2 exposure concentrations and longer exposure times are associated with
decreases in plant growth and yield. The 1982 PM-SOx AQCD concluded that more
definitive concentration-response studies were needed before useable exposure metrics
could be identified. However, very few studies of the effects of SO2 on the growth of
vegetation in the U.S. have been conducted since 1982. Recent studies from eastern
Europe indicate recovery of tree growth in response to decreases in SO2 concentrations
since the 1980s and that annual SO2 concentrations of 4 ppb decreased silver fir (Abies
alba) growth. In West Virginia, the growth of eastern red cedar (Juniperus virginiana)
trees increased with declines in SO2 emissions since the 1980s.
IS.4.2 Nitrogen Oxide, Nitrogen Dioxide, and Peroxyacetyl Nitrate
In the 2008 ISA, evidence was sufficient to infer a causal relationship between exposure
to NO, NO2, and PAN and injury to vegetation. It is well known that in sufficient
concentrations, NO, NO2, and PAN can have phytotoxic effects on plants by decreasing
photosynthesis and inducing visible foliar injury. However, the 1993 Oxides of Nitrogen
AQCD concluded that concentrations of NO, NO2, and PAN in the atmosphere are rarely
high enough to have phytotoxic effects on vegetation (U.S. EPA. 1993). and very little
new research has been performed at concentrations currently observed in the U.S. (see
Appendix 3.3). It is also known that these gases alter the N cycle in some ecosystems,
and nutrient effects of N are discussed in Section IS.5. Thus, consistent with the previous
2008 ISA, the body of evidence is sufficient to infer a causal relationship between
gas-phase NO, NO2, and PAN and injury to vegetation.
IS.4.3 Nitric Acid
In the 2008 ISA, evidence was sufficient to infer a causal relationship between exposure
to HNO3 and changes to vegetation. The 2008 ISA reported experimental exposure to
HNO3 resulted in damage to the leaf cuticle of pine and oak seedlings, which may
predispose those plants to other stressors such as drought, pathogens, and other air
pollutants. Since the 2008 ISA, Padgett et al. (2009) investigated dry deposition of HNO3
on the foliage in a fumigation study and confirmed the earlier research. Nitric acid can
also add to N nutrient enrichment of ecosystems and is discussed in Section IS.5. The
2008 ISA also reported several lines of evidence that past and current HNO3
concentrations may be contributing to the decline in lichen species in the Los Angeles
basin. Subsequent studies conducted in the Los Angeles basin since the 2008 ISA provide
further evidence of the impacts (see Appendix 3.4). These new studies continue to
40
-------
support the causal findings of the 2008 ISA, such that the body of evidence is sufficient
to infer a causal relationship between gas-phase HNO3 and changes to vegetation.
IS.5 Terrestrial Ecosystem Nitrogen Enrichment and Acidification
For terrestrial ecosystems, new evidence reinforces causal findings from the 2008 ISA
and provides the basis for two new causal statements that reflect a more comprehensive
understanding of how N and acidifying deposition alter terrestrial ecosystem biota
(Table IS-1). In general, N deposition may cause soil N enrichment and stimulate the
growth of opportunistic species. However, in sensitive soils, deposition of N and/or S can
cause soil acidification, which may decrease growth and cause mortality among sensitive
plant species. Atmospheric deposition ofN and S alter the species composition of
terrestrial systems by one of four mechanisms: (1) nutrient enrichment (eutrophication;
Appendix 4 and Appendix 6), (2) acidification (Appendix 4 and Appendix 5), (3) direct
damage (Appendix 3), and (4) secondary effects (e.g., wildfire; Appendix 6). Ecosystems
and communities may be simultaneously affected by one or more mechanisms depending
on the sensitivity of environmental and biological properties to each mechanism.
Despite the abundance of N in the environment, plants are unable to directly access the
large pools of N contained in the atmosphere as N2 gas and in the soil as large organic
molecules. Consequently, the limited availability of reactive N often constrains biological
activity in terrestrial ecosystems. N deposition is therefore considered nutrient
enrichment because N additions generally stimulate plant growth and productivity
(cumulative growth of all vegetation within a community), which has been recognized
since the second half of the 19th century. In comparison, the biological effects of
acidifying deposition are less common and largely constrained to ecosystems with
historically high rates of deposition and that are vulnerable because of factors such as
geology and climate. While S is also an essential macronutrient, less S is required for
growth than N, and areas affected by acidifying deposition typically receive S at rates
that greatly exceed biotic demand. Instead, the impact of acidifying deposition stems
from the disruptions to biochemical processes caused by decreased pH and shifts in soil
physiochemical processes that decrease the supply of other essential nutrients (e.g., Ca,
Mg) and from increased mobilization of toxic forms of Al.
Current knowledge of soil biogeochemistry indicates soil N enrichment and soil
acidification occur in sensitive ecosystems across the U.S. at present levels of deposition.
Newly published work indicates decreasing SO2 emissions and S deposition have led to
early signs of recovery from acidification in some northeastern watersheds, but areas in
the Southeast do not show recovery (Appendix 4). There are many well-defined soil
41
-------
indicators related to the biological effects of acidifying (N + S) deposition. New evidence
uses these indicators to describe the status of ecosystems, either by empirical observation
or models. Soil indicators for acidification are more typically modeled than those for
eutrophication effects. There is an abundance of new information on biogeochemical
pools and processes, including a new conceptual framework for the N saturation of
terrestrial ecosystems.
The enrichment of terrestrial ecosystems by N deposition often increases plant
productivity and causes changes in physiology and growth rates that vary among species.
This has been observed for herbaceous plants and trees across ecoregions. The changing
growth rates transform competitive interactions between species, and consequently, lower
species diversity is often observed with increasing N deposition within terrestrial
communities. The level of N deposition negatively affecting community composition is
often expressed as a Critical Load (CL). There are many new CLs available since the
2008 ISA, including those for lichens, herbaceous plants, and mycorrhizae.
The process of terrestrial acidification has been well understood and documented for
decades. Recent research, since the 2008 ISA, has confirmed and strengthened this
understanding and provided more quantitative information, especially across
regional-scale landscapes. Several studies have evaluated the relationships between soil
chemical indicators of acidification and ecosystem biological endpoints (see Table 5-6),
and some biogeochemical models are well established. There have been new advances in
the parameterization of acidification models to U.S. soils since the 2008 ISA
(Appendix 4.5) resulting in better certainty of CLs. Biological endpoints included in the
evaluations include physiological and community responses of trees and other vegetation
(such as lichens), soil biota, and fauna.
The following section summarizes the main effects of N and S deposition on terrestrial
ecosystems.
IS.5.1 Soil Biogeochemistry
In the 2008 ISA, evidence was sufficient to infer causal relationships between
(1) acidifying deposition and changes in terrestrial biogeochemistry and (2) between N
deposition and terrestrial biogeochemical cycling of N. There is new evidence of how
deposition contributes to total loading in ecosystems, as well as new information from
addition, gradient, and time-series studies characterizing how deposition affects soil pools
and processes. Much of the new work focuses on the effects of N deposition, with
relatively little work focusing on S deposition. Soil N enrichment and soil acidification
occur in sensitive ecosystems across the U.S. at present levels of deposition. Decreasing
42
-------
S emissions have led to early signs of recovery from acidification in some northeastern
watersheds, but areas in the Southeast do not show recovery (for additional discussion on
recovery see Section IS. 11). Deposition rates of total N (NOy + NHX) are relatively
unchanged across much of the contiguous US (Appendix 2.7). Accordingly, there are no
signs of recovery from N enrichment effects. CL determinations have been made at the
ecoregion scale for NOs leaching. CLs for biological effects are summarized below
(Section IS.5.1.2. Section IS.5.2.2. and Section IS.5.3.3). The body of evidence is
sufficient to infer a causal relationship between N and S deposition and alteration of
soil biogeochemistry in terrestrial ecosystems, which is consistent with the conclusions
of the 2008 ISA.
IS.5.1.1 Soil Processes and Indicators
Deposition ofN or N + S alters soil chemistry, which can have cascading effects on
aquatic ecosystems (for effects on aquatic biology and chemistry see
Appendix 7-Appendix 10). Soil acidification is a natural process that can be accelerated
by N or S deposition. Deposition in the forms of HNO3 and H2SO4 can directly acidify
soils; however, deposition of reduced forms of N (e.g., NHx) can also cause soil
acidification by releasing hydrogen ions (H+) during the microbial oxidation of NH44" to
NC>3~. There are a number of soil biogeochemical processes associated with acidification
(Table IS-2). Base cations can counterbalance acid anions. Base cations are added to the
soil by weathering and atmospheric deposition and are removed by leaching and
biological uptake. Where acidifying deposition rates are high relative to base cation input,
deposition can deplete exchangeable base cation pools in soils. There are several useful
indicators of soil acidification (Table IS-2) that have quantitative relationships to
biological responses (Appendix 5).
Table IS-2 Summary of key soil geochemical processes and indicators
associated with eutrophication and acidification.
N Driven
Nutrient
Endpoint	Enrichment Acidification	The Effect of Deposition
Process
N saturation	X	X	New empirical evidence suggests revising the N saturation
concept; specifically, it is now observed that NO3" leaching
can occur even if the ecosystem N capacity to retain N has
not yet been saturated.
43
-------
Table IS-2 (Continued): Summary of key soil geochemical processes and
indicators associated with eutrophication and
acidification.
N Driven
Nutrient
Endpoint	Enrichment Acidification	The Effect of Deposition
Soil N accumulation	X	X	New meta-analyses across ecosystem types confirm
inorganic soil NO3" concentration increases with N addition.
A new gradient study confirms that N concentration
increases with N deposition. A new addition study confirms
increased soil N accumulation. New studies on Soil N
accumulation are summarized in Table 4-3.
NO3" leaching	X	X	New meta-analyses confirm leaching increases with N
additions. Regional-scale gradient analyses: <8 kg N/ha/yr
onset of leaching; <1 kg N/ha/yr in European forests; in the
NE U.S., 90% retention for sites receiving 7 kg N/ha/yr to
60% retention for sites receiving 11 kg N/ha/yr.
New USFS CLs for the onset of leaching: 8-10 kg N/ha/yr in
eastern and western U.S., 17 kg N/ha/yr in the Sierra
Nevada and San Bernardino Mountains. New studies on
Soil N accumulation are summarized in Table 4-3.
S accumulation and	X	Some soils (notably in many watersheds in the SE U.S.)
adsorption	have the capacity to adsorb substantial quantities of S, with
essentially no acidification of drainage water. Nevertheless,
S adsorption capacity is finite, and under continual high S
deposition loading, the adsorptive capacity of soil will
eventually be exceeded.
New studies of 27 watersheds in the SE indicate most will
begin releasing SO42" in the next two decades; NE
watersheds show a net loss of S from soils now in response
to decreased levels of atmospheric S deposition. New
studies on soil S accumulation are summarized in Table 4-4.
SO42" leaching	X	Atmospheric S deposition generally increases leaching of
SO42" to surface waters. The amount of deposition that
causes the onset of leaching varies across the landscape.
New studies on soil SO42" leaching are summarized in
Table 4-4.
Base cation leaching	X	Base cation (Ca, Mg, K, Na) release from soil particles to the
and exchange	soil solution occurs in response to the input of acid anions
(SO42" and NO3") from deposition.
New studies confirm base cation depletion continues to
occur in the Rocky Mountains (threshold 28 kg N/ha/yr) and
in U.K. grasslands, while in a NE forest, 17 yr of N addition
did not cause further depletion. A meta-analysis suggests
cation depletion soon after increased deposition of acid
anions, but this depletion tapers off with time. New studies
on base cation leaching and exchange are summarized in
Table 4-5.
44
-------
Table IS-2 (Continued): Summary of key soil geochemical processes and
indicators associated with eutrophication and
acidification.
Endpoint
N Driven
Nutrient
Enrichment
Acidification
The Effect of Deposition
Al mobilization

X
The threshold for inorganic Al mobilization from soil is
<15-20% soil base saturation. This is an extremely
important effect of acidifying deposition because inorganic
monomeric Al is toxic to biota (Appendix 5 and Appendix 8).
Inorganic Al is minimally soluble at pH 6.0, but solubility
increases steeply at pH below 5.5.
New studies on Al in soils are summarized in Table 4-6.
Nitrification
X
X
Nitrification releases 2 mol hydrogen ion (H+) per mol NHV
converted to NO3", acidifying soils. As soil inorganic N
accumulates, net nitrification rates often increase, and NO3"
can leach from the ecosystem.
New N gradient and meta-analysis studies confirm N
addition increases nitrification. New studies on nitrification
are summarized in Table 4-6.
Denitrification
X

Denitrification is the microbial reduction of NO3" to NO2",
NO, the greenhouse gas N2O, and N2, which occurs under
anaerobic conditions. In Europe, soil switched from a source
to a sink after two decades of N deposition exclusion. New
meta-analysis confirms N addition increases denitrification
rates. New studies on denitrification are summarized in
Table 4-6.
DOC leaching
X
X
In recent years, the DOC of many lakes and streams has
risen, with the source likely from the soils in the adjacent
terrestrial watershed. However, the mechanism causing the
observed increase is unclear and may be due to a
combination of soil recovery from acidification, changes in
climate (e.g., temperature and precipitation), and N
deposition among other mechanisms. New studies are
summarized in Table 4-10.
Decomposition
X
X
The addition of N can stimulate the breakdown of labile
compounds that degrade during the initial stages of
decomposition, but added N can suppress the
decomposition of more recalcitrant material. There are new
addition studies and meta-analyses on mechanisms and
response trends.
New studies are summarized in Table 4-8.
Indicator
Soil [N]
X
X
Increases in soil [N] indicate soil N accumulation and the
size of the soil N pool that may be assimilated by organisms
or mobilized via leaching.
Soil C:N ratio
X
X
Decreasing soil C:N linked to changes in decomposition and
increases in nitrification and NO3" leaching.
<20-25 causes increased nitrification and elevated risk of
NO3" leaching in the U.S. and <25-30 for increased NO3"
leaching in Europe.
45
-------
Table IS-2 (Continued): Summary of key soil geochemical processes and
indicators associated with eutrophication and
acidification.
Endpoint
N Driven
Nutrient
Enrichment
Acidification
The Effect of Deposition
Soil base saturation

X
Increasing N + S deposition decreases the soil pool of
exchangeable base cations.
<15-20% exchange ion chemistry is dominated by inorganic
Al and may cause injury to vegetation (see Appendix 5).
Soil Bc:AI ratio

X
Increasing N + S deposition decreases the soil pool of
exchangeable base cations, often decreasing the Ca:AI
ratio.
Ca:AI <1.0 causes physiological stress, decreased growth,
and mortality of sensitive plant species (see Appendix 5).
Fungi-to-bacteria ratio
X

New indicator: increasing N deposition decreases the
fungi-to-bacteria ratio and causes a transition from N to C
limitation among soil food webs.
Al = aluminum; Al3+ = aluminum(lll); Be = base cations; C = carbon; Ca = calcium; DOC = dissolved organic carbon; H+ = hydrogen
ion; ha = hectare; K= potassium; kg = kilogram; Mg = magnesium; N = nitrogen; N2 = molecular (atmospheric) nitrogen;
N20 = nitrous oxide; Na = sodium; NE = northeastern; NH4+ = ammonium; NO = nitric oxide; N02" = nitrite; N03" = nitrate;
S = sulfur; SE = southeastern; S042" = sulfate; U.K. = United Kingdom; U.S. = United States; USFS = U.S. Forest Service;
yr = year.
Some of the same processes and indicators associated with acidification are also
associated with the N enrichment of soils in response to the input of exogenous N
(Appendix 4.3). The 2008 ISA documented that the increase in global reactive N (defined
as NOy + NHX + organic N) that occurred over the previous century was largely due to
three main causes: (1) widespread cultivation of crops that promote conversion of N2 gas
to organic N through biological N fixation, (2) fossil fuel combustion converting
atmospheric N2 and fossil N to NOx, and (3) the Haber-Bosch process, which converts
nonreactive N2 to reactive N to sustain food production and some industrial activities
(Galloway et al.. 2003; Galloway and Cowling. 2002).
The 2008 ISA documented that atmospheric deposition of N can increase soil N, the
accumulation of which is linked to increased N leaching and decreased retention of N.
CLs for the onset of elevated NO3 leaching are given in Appendix 4.6.2.2.
The 2008 ISA described the conceptual model of N saturation, which occurs when N
input rates to terrestrial ecosystems exceed the uptake capacity of the soils and biota and
is indicated by the onset of increased soil N leaching. However, more recent work has
revised the N saturation model in response to observations in which N leaching resulted
from N input rates that are faster than vegetation and soil uptake rates, thus distinguishing
capacity N saturation from kinetic N saturation. Budgets from 83 forested watersheds in
the northeastern U.S. show that N retention averages 76% of the incoming atmospheric N
46
-------
deposition and decreases from 90% retention at 7 kg N/ha/yr of deposition to 60%
retention at 11 kg N/ha/yr of deposition.
The 2008 ISA documented that N enrichment is associated with changes in microbially
mediated biogeochemical processes, including nitrification, denitrification, and
decomposition (Appendix 4.3). The addition of N can increase nitrification (the microbial
conversion of NH4+ to NO, ). which contributes to soil acidification. N deposition to soils
can decrease surface soil C:N ratio, which can stimulate nitrification when C:N ratios fall
below 20 to 25. The NOs created by nitrification may be leached, biologically
immobilized, or denitrified. Denitrification is the microbial reduction of NO3 to NO2 ,
NO, the greenhouse gas N2O, and N2, which occurs under anaerobic conditions. Several
syntheses have been published since 2008 evaluating N addition effects on denitrification
and nitrification in terrestrial ecosystems. A new meta-analysis shows N addition
substantially increases denitrification from many types of ecosystems (e.g., coniferous
forest, deciduous forest, tropical forest, wetland, grassland), but not heathlands. Among
five chemical forms of N studied, NO.? addition showed the strongest stimulation of N2O
emission. Using data extracted from 206 peer-reviewed papers, a second meta-analysis
observed that the largest changes in the ecosystem N cycle caused by N addition were
increased nitrification (+154%), N2O emissions (+134%), and denitrification (+84%).
IS.5.1.2 National-Scale Sensitivity and Critical Loads
As of the 2008 ISA, the regions of the U.S. with abundant acid-sensitive soils had been
well delineated. These acid-sensitive ecosystems are generally located in mountainous
terrain in the eastern U.S. and are underlain by bedrock resistant to weathering. However,
a similar delineation of the areas sensitive to the eutrophication effects of N had not yet
been completed. There is strong evidence demonstrating that biogeochemical sensitivity
to deposition-driven eutrophication and acidification is the result of historical loading,
geologic/soil conditions (e.g., mineral weathering and S adsorption), and/or natural
sources of N and S loading to the system.
Since the 2008 ISA, several new publications have advanced our understanding of soil
recovery from acidification and CLs. New publications report the results of field
observations and modeling studies on soil recovery from acidification, specifically in the
northeastern U.S., and the lack of recovery in the southern Appalachian Mountains
(Table 4-18). New ecoregion-scale terrestrial CLs for NO3 leaching were published in
2011 and have been updated by more recent published work. Finally, Clark et al. (2018)
estimated areas exceeding CLs for terrestrial acidification and NO3 leaching for the
contiguous U.S. for 1800 to 2025. For terrestrial acidification, area exceeding the
47
-------
minimum CL peaked at almost 2.8 million km2 by 1975 before declining; whereas, for
N03 leaching, the area exceeding the minimum CL peaked at roughly 3.4 million km2
around 1995.
IS.5.2 Biological Effects of Terrestrial Nitrogen Enrichment
The enrichment of terrestrial ecosystems by N deposition often increases plant
productivity and causes changes in physiology and growth rates that vary among species.
This combination of effects can alter the composition and decrease diversity of terrestrial
communities by transforming competitive interactions between species and changing the
availability of other essential resources, including light, water, and nutrients. Because N
deposition can cause both eutrophication and acidification and these processes can occur
simultaneously, the relationship between N deposition and community composition has
often been derived empirically. Many of the effects ofN deposition are similar across
ecosystems and life forms because N is an essential macronutrient, but the composition
and magnitude of how these effects are expressed within an ecosystem can differ as a
result of biotic and abiotic influences. Consequently, as with the 2008 ISA, we have
grouped the effects of N deposition on physiology and biodiversity by biome (e.g., forest,
tundra, grassland, and arid lands), with further framing by life form (e.g., plants,
microorganisms) and functional groups (e.g., trees, herbaceous plants). In comparison,
the broadest CLs created by the scientific community are at the ecoregion level, in which
spatial boundaries across the landscape are typically defined based on ecological,
climatological, and geological differences.
The 2008 ISA documented consistent evidence that N additions increased plant
productivity broadly across a wide range of terrestrial ecosystems. Since 2008, a large
body of new research on the biological effects of added N in terrestrial ecosystems has
been published from investigations of plant and microbial physiology, long-term
ecosystem-scale N addition experiments, regional and continental-scale monitoring
studies, and syntheses. These studies have been conducted in ecosystems representing
biomes in the U.S., including tundra, grasslands, arid and semiarid lands, and tropical,
temperate, and boreal forests. Because of the breadth of this research, there is a strong
mechanistic and empirical understanding for many of the biological effects of added N.
This body of evidence is sufficient to infer a causal relationship between N
deposition and the alteration of the physiology and growth of terrestrial organisms
and the productivity of terrestrial ecosystems.
The varying effects of N deposition on the growth and physiology of individual species
have consequence(s) for biodiversity. In the 2008 ISA, evidence was sufficient to infer a
48
-------
causal relationship between N deposition and the alteration of species richness, species
composition, and biodiversity in terrestrial ecosystems. The 2008 ISA documented
consistent evidence of reduced species richness and altered community composition from
N addition studies in the U.S. and N deposition gradient studies in Europe for grassland
plant diversity, forest understory plants, and forest mycorrhizal fungi. There was also
consistent evidence of altered plant and mycorrhizal community composition from N
addition studies in arid and semiarid ecosystems, particularly in southern/central
California. There was little evidence of changes in forest overstory tree composition.
Since the 2008 ISA, new research techniques have been developed to understand
community composition, a larger number of communities have been surveyed, and new
regional and continental-scale studies have made it possible to isolate the influence of N
deposition from other environmental factors. This new research has provided more
extensive and mechanistic evidence, and combined with the findings of the 2008 ISA,
this body of evidence is sufficient to infer a causal relationship between N deposition
and the alteration of species richness, community composition, and biodiversity in
terrestrial ecosystems.
IS.5.2.1 Physiology and Biodiversity
At the time of the 2008 ISA, terrestrial ecologists had used meta-analyses to broadly
quantify the effects that N deposition can have on the growth of terrestrial plants,
concluding that N additions stimulate plant productivity by 20-30% in grasslands,
forests, tundra, and wetlands, increase aboveground productivity in herbaceous plant
communities, alter plant tissue chemistry, decrease biomass of mycorrhizal fungi, and
alter litter decomposition (Appendix 6.6.1). Recent research has provided further
coherent and consistent evidence that N additions stimulate plant growth and
productivity, but this research is still dominated by studies of temperate ecosystems and
aboveground plant responses (Figure 6-1 and Figure 6-2).
In the 2008 ISA, the positive plant growth response to N deposition was attributed to
higher rates of photosynthesis. However, evidence for this is mixed: increases in
photosynthesis following N additions have been observed across a variety of plant
functional types, but higher rates of photosynthesis have not been consistently observed
in response to chronic N additions meant to simulate atmospheric deposition. There is
new support for another mechanism that would increase aboveground growth: decreases
in the quantity of C allocated by plants to roots and mycorrhizae. There was evidence in
the 2008 ISA that N additions increase aboveground biomass more than belowground
biomass, raising the shoot-to-root ratio among plants, but evidence is now more
consistent and widespread. Plants also invest substantial amounts of C to support
49
-------
mycorrhizal fungi, but there is evidence this investment declines when N is added to
terrestrial ecosystems.
Evidence that biodiversity change can be a consequence of N deposition has accumulated
since 2008 and includes new information for major taxonomic groups, including
herbaceous plants, overstory trees, and two groups of symbionts (lichens and
mycorrhizae). Evidence is now more widespread for decreases in lichen species richness
as the result of N deposition in the U.S. There are direct observations that N deposition in
the U.S. is changing mycorrhizal community composition and altering herbaceous plant
species richness across a broad range of ecosystems, including forests, grasslands, arid
and semiarid ecosystems, and alpine tundra. In addition, based on variation in mortality
and growth rates of co-occurring tree species, there is also indirect evidence that N
deposition is altering overstory tree community composition.
A substantial body of research linking changes in biodiversity to shifts in N availability
has been developed. Two hypotheses for species loss are (1) the random-loss hypothesis
and (2) the functional trait hypothesis. The random-loss hypothesis suggests rare species
are most likely to disappear as increased competition for resources, such as light,
eliminates less successful individuals; whereas the functional trait hypothesis predicts
that organisms with disadvantageous traits (e.g., shorter plants) will be outcompeted
when N is added. Both hypothesized mechanisms can operate simultaneously, and both
tie the changes in physiology, growth, and productivity caused by increased N
availability to declines in biodiversity.
As noted in Appendix 4, soil microorganisms have important roles in regulating N and C
cycling. There are several mechanisms to alter soil microbial biomass and physiology,
including changes in soil pH, increases in inorganic N availability, shifts in soil food
webs, and changes in the quantity and quality of available C. There were some
observations in the 2008 ISA that added N decreases microbial biomass, but there is now
more evidence that added N generally negatively or neutrally affects microbial biomass C
and microbial biomass N (Table 6-4).
IS.5.2.1.1 Forests
Forests occur within every U.S. state, but are most abundant in the eastern U.S., montane
and coastal portions of the western U.S., and Alaska. The distribution of forests is bound
by water availability, cold temperatures, and land management. In the 2008 ISA, there
was consistent evidence that N additions stimulated forest productivity, but these
responses varied widely and included both neutral and negative effects of N additions on
tree growth. However, there had been no empirical analyses of how atmospheric N
50
-------
deposition altered forest productivity in the U.S. at broad scales. The 2008 ISA lacked
information on whether N deposition had any impact on the diversity and composition of
forest overstory trees, but it did present evidence for changes in the composition of
herbaceous vegetation, epiphytic lichens, and microbial communities. The addition of
new research since the 2008 ISA provides coherent evidence that N deposition alters the
physiology and growth of overstory trees and provides indirect evidence that N
deposition changes the community composition of overstory trees. Further, new research
supports N deposition altering the physiology, growth, and community composition of
understory plants, lichens, mycorrhizal fungi, soil microorganisms, and arthropods
(Appendix 6.2.3 and Appendix 6.3.3).
As of the 2008 ISA, most long-term N addition experiments were conducted in temperate
forests in the northeastern U.S. or in temperate or boreal forests in Europe. In these
studies, conifer species were less likely than broadleaf species to exhibit positive growth
responses to added N and more frequently exhibited increased mortality and decreased
growth. Since the 2008 ISA, new observations from experiments, forest inventory
studies, model simulations, and data synthesis efforts have been published, quantifying
increases in forest net primary productivity (NPP), net ecosystem productivity (NEP),
and ecosystem C storage (Figure 6-3). Overall, evidence is consistent that N deposition
broadly increases tree growth and forest productivity, including specific evidence
indicating that current rates of N deposition in the contiguous U.S. broadly stimulate
aboveground forest productivity (Appendix 6.2.3.1).
Despite these broad effects, it is also clear that with N addition growth and mortality
responses vary by tree species. Many of the observations in the 2008 ISA have been
reinforced by more recent research, including long-term forest inventory data collected
from across the U.S. and Europe. Recent analyses of U.S. forest inventory data by Horn
et al. (2018) found that tree species vary in their growth and mortality responses to N
deposition (Appendix 6.2.3.1). Responses of individual tree species ranged from
consistently increasing growth with greater N deposition; to increasing growth at lower N
deposition but decreasing growth at higher levels; to consistently decreasing growth with
greater N deposition. Mortality responses showed a similar pattern between species.
Notably, species with varying responses in growth and mortality co-occurred in places in
the U.S. Thus, this indirect evidence suggests that changes in tree community
composition are occurring due to N deposition (Appendix 6.3.3.1). These analyses
represent an advancement in our understanding from the time of the 2008 ISA.
In comparison, there is direct evidence that N deposition is altering the composition of
forest understory plant communities (Appendix 6.3.3.2). The evidence for altered forest
understory plant communities (also known as herbaceous layer or groundcover
51
-------
vegetation) comes from both the 2008 ISA and from the more recent literature. Changes
in understory plant communities have been observed in monitoring plots along
atmospheric N deposition gradients in the U.S. and in Europe. In Europe, forest
understory plant communities have shifted with increasing N toward more
nutrient-demanding and shade-tolerant plant species.
Higher rates of aboveground tree growth in response to N deposition might be due to
shifts in C allocation away from belowground processes. Changes in C allocation in
response to additional N have been accompanied by decreases in the abundance of
mycorrhizal fungi and changes in mycorrhizal community composition (Table 6-2,
Table 6-14). Evidence for composition change is particularly abundant in
ectomycorrhizal fungal communities (Table 6-14); there are fewer observations of how
arbuscular mycorrhizal fungal communities change in response to N additions (Table 6-3;
Table 6-16). There are also numerous observations of altered total microbial (including
bacterial) biomass and community composition. For microbial biomass, most studies
identified since 2008 showed either negative or neutral effects of N additions, consistent
with the results of syntheses published before the 2008 ISA (Table 6-4). Changes in soil
microbial community composition were identified along an N deposition gradient in
Europe and in all three N addition studies (Table 6-14). The effects of N additions on
individual microbial taxonomic groups (bacteria, archaea, fungi, etc.) have been less
consistent (Table 6-15). Overall, there is evidence that N additions can decrease total
microbial biomass and alter microbial communities in forest soils.
Within soil food webs, soil microorganisms have both direct and indirect links to
arthropods. Because arthropods feed upon both microorganisms and litter, they can be
important regulators of decomposition, nutrient cycling, and forest productivity. Several
studies have examined the response of forest arthropod communities to added N,
including a group of studies on insect herbivores conducted in southern California
(Table 6-17). There is coherent evidence thatN additions can alter forest arthropod
communities.
Epiphytic lichens have long been recognized as sensitive to air pollution. Although these
organisms often make up a small portion of forest biomass, they have important roles in
hydrologic cycling, nutrient cycling, and as sources of food and habitat for other species.
New research on lichen community composition identified since the 2008 ISA has further
added to the consistent and coherent evidence that lichen communities in the U.S. and
Europe are sensitive to current levels of atmospheric N deposition (Appendix 6.2.6;
Table 6-23). In particular, the U.S. Forest Service's Forest Inventory and Analysis
Program has ample data on the abundance of lichens throughout the U.S., and shifts in
lichen community composition clearly attributable to atmospheric N pollution have been
52
-------
observed in forests throughout the West Coast, in the Rocky Mountains, and in
southeastern Alaska. Shifts in epiphytic lichen growth or physiology have been observed
along atmospheric N deposition gradients in the highly impacted area of southern
California, but also in more remote locations such as Wyoming and southeastern Alaska
(Table 6-5). Experimental N studies have also created more detailed insight into changes
in lichen physiology processes.
Overall, there is widespread evidence from forests that N deposition alters the growth and
physiology of trees, and indirectly suggests N deposition affects tree community
composition. Nitrogen deposition in forests also alter the growth, physiology, and
biodiversity of herbaceous plants, lichens, soil microorganisms, and arthropods.
IS.5.2.1.2 Tundra
Within the U.S., tundra ecosystems are limited to Arctic ecosystems in Alaska and to
relatively isolated, high elevation sites. Although these ecosystems tend to be remote, the
influence of atmospheric N deposition is distinct and there was evidence in the 2008 ISA
indicating that alpine tundra plant communities were sensitive to atmospheric N
deposition. Alpine organisms may be more sensitive to N deposition because of the
unique nature of N cycling in these ecosystems, which tend to have limited inorganic N
availability. Since the 2008 ISA, numerous studies of tundra physiological, productivity,
and community composition responses to added N have been published, providing further
evidence that N deposition alters the growth and physiology of alpine plant communities,
including vascular plants (herbaceous and woody), bryophytes, and lichens
(Appendix 6.2.4), as well as evidence of altered soil microbial communities (Table 6-8;
Table 6-19).
As in forests, increases in N content in response to additional N are widespread in tundra
plant communities (Table 6-6). Higher tissue N concentrations in response to added N
have been observed in multiple studies for vascular plants, bryophytes, and lichens. The
2008 ISA noted that plant growth and biomass responses tended to be species specific.
Subsequent studies have confirmed this result (Table 6-6), showing varying responses to
added N among ecosystem types, plant functional groups, and species. Whereas vascular
plants tend to show a positive response to added N, both bryophytes and lichens tend to
decrease in biomass and cover (Table 6-5; Table 6-6).
Given the varying effects of N addition on species physiology and growth, the numerous
observations of N addition impacts on species richness, species diversity, and community
composition among vascular plants, bryophytes, and lichens in alpine and Arctic tundra
ecosystems are unsurprising (Appendix 6.3.4; Table 6-18). Within the U.S., these
53
-------
observations have included effects of N additions on plant community composition in
Colorado and Washington. In northern Europe, decreases in plant species richness along
atmospheric N deposition gradients have been documented. Overall, this new research
has provided further evidence that experimental N additions can alter plant biodiversity in
alpine and Arctic tundra ecosystems and has provided new evidence that current rates of
atmospheric N deposition in Europe are associated with a loss of plant species richness in
these ecosystems.
There are relatively few observations regarding the effect of N additions on total
microbial biomass or the biomass response of individual microbial taxonomic groups in
tundra ecosystems, and these results have also been largely inconsistent. However, new
research has provided evidence that N additions can alter microbial community
composition in alpine tundra ecosystems (Table 6-8; Table 6-19).
IS.5.2.1.3 Grasslands
Grasslands are most prevalent in the central U.S., yet also are widely distributed across
the U.S. in areas where woody vegetation is excluded by environmental factors or
management. There was widespread evidence at the time of the 2008 ISA that the
growth, physiology, and productivity of grassland plants could be altered by N
deposition. In addition, there were multiple lines of evidence in the 2008 ISA that
grassland plant, mycorrhizal, and microbial communities were sensitive to N inputs.
Combined with subsequent research, the evidence is clear that physiology, growth, and
community composition of plants, mycorrhizae, soil microorganisms, and arthropods are
sensitive to N inputs in grasslands.
Although NPP can be limited by multiple factors (e.g., water, herbivores, other nutrients)
in all ecosystems, limitations other than N tend to be more marked in grasslands than
forests, making it harder to understand and predict the effects of increased N availability.
However, the general response is similar (Appendix 6.2.5): N additions stimulate NPP,
increase foliar N, and increase allocation to aboveground biomass (increased ratio of
shoot:root mass).
Evidence from the U.S. of grassland plant community composition change in the 2008
ISA was based on N addition studies in Mediterranean grasslands in California and
northern prairie ecosystems. However, large-scale assessments of biodiversity across
observed atmospheric N deposition gradients were restricted to Europe. Recent research
provides further evidence that N deposition reduces grassland biodiversity in the U.S. and
Europe (Appendix 6.3.5). Since 2008, there have been direct observations of reduced
species richness along atmospheric N deposition gradients for grasslands in the U.S. and
54
-------
Europe. These gradient studies have documented an interaction with soil pH, noting that
N deposition causes a greater loss of species richness and a shift in community
composition at sites with lower pH. Together, these findings from deposition gradients in
the U.S. and Europe provide coherent evidence that N deposition causes shifts in plant
community composition and the loss of plant species richness through mechanisms of
both acidification and eutrophication. Experimental studies published since 2008 have
provided more insight into the mechanisms linking changes in plant and microbial
community composition to increased N availability, providing evidence that declines in
species richness increase with time and that competition for resources such as water may
exacerbate the effects of N addition on diversity.
Overall, the additional studies in grassland ecosystems have confirmed that many of the
responses observed in the older N addition studies also occur at present rates of
atmospheric N deposition. These changes include losses in forb species richness (which
make up the majority of grassland biodiversity), greater growth of grass species (which
make up the majority of grassland biomass), changes in reproductive rates, as well as
shifts in mycorrhizal (Table 6-16), soil microbial (Table 6-20), and arthropod
populations. In total, because of the prevalence of N limitation in grasslands and the
dominance by fast-growing species that can shift in abundance rapidly (in contrast to
forest trees), grasslands appear especially sensitive to N input rates comparable to N
deposition across much of the contiguous U.S.
IS.5.2.1.4 Arid and Semiarid
Arid and semiarid ecosystems are abundant in areas of the western U.S. where climate or
orography create annually or seasonally dry conditions. At the time of the 2008 ISA, a
large amount of information was available on how N deposition affected the growth and
physiology of plants and microorganisms in arid and semiarid ecosystems, and there was
coherent evidence that plant communities in these ecosystems could be altered by the
added N. Evidence for these effects was particularly strong in coastal sage scrub (CSS),
chaparral, and Mojave Desert ecosystems in southern California. Within the CSS
ecosystems, N deposition has been linked to increased mortality in native shrubs,
decreased abundance of arbuscular mycorrhizal fungi, higher cover of invasive annual
plants, and increased wildfire activity. Similar increases in invasive annual plant cover
and fire frequency have also been attributed to N deposition in areas of the Mojave
Desert downwind of urban centers in southern California. Research since 2008 has
further documented these effects, with consistent evidence that N deposition can affect
the physiology, growth, and community composition of plants and soil microorganisms
in arid and semiarid systems.
55
-------
The effects of N deposition on physiological and biogeochemical processes in arid and
semiarid ecosystems are even more clearly dependent on moisture availability than in
grasslands (Appendix 6.2.6). In these ecosystems, inorganic N often accumulates in the
soil during dry periods, and growth and physiological responses to additional N are only
observed when and where sufficient moisture is available. Two additional important
effects of aridity include (1) higher soil base saturation and pH that buffer these systems
from the acidification effects of N deposition and (2) spatially patchy nutrient availability
that develops beneath isolated shrub canopies. One important effect of N deposition on
arid and semiarid ecosystems is to decrease the patchiness of nutrient availability, which
promotes the growth of invasive annual plants in the spaces between the isolated shrubs.
The growth of these annual plants creates a more continuous fuel bed for wildfires,
increasing the prevalence of fire, and shifting plant community composition toward more
fire-adapted plant species.
Since 2008, increases in aboveground plant biomass or plant cover have been observed in
the U.S. in the Mojave and Sonoran Deserts, and in southern California chaparral, and
internationally in China and Spain (Appendix 6.2.6). Given the linkage to fire, it is
notable that there have been multiple observations of increased annual plant growth in the
Mojave Desert in response to added N.
New research has also provided further evidence that N deposition alters plant
communities in arid and semiarid ecosystems, particularly in southern California, but also
in other locations (Appendix 6.3.6). Many of these studies documented changes in plant
community composition, with fewer observations of plant species loss or changes in plant
diversity. Overall, this body of research has provided consistent and coherent evidence
that N deposition is altering the growth, physiology, and community composition of
plants in arid and semiarid ecosystems. Relative to plants, there are fewer studies of
microbial communities (Table 6-12; Table 6-22), but these studies provided evidence that
N additions can alter the abundance, physiology, and community composition of soil
microorganisms in arid and semiarid ecosystems.
IS.5.2.2 National-Scale Sensitivity and Critical Loads
At the time of the 2008 ISA, there had been little quantification of the extent and
distribution of N sensitivity in terrestrial ecosystems in the U.S. In the 2008 ISA, there
was no published U.S. national CL assessment. Since then, substantial work has been
done on quantifying N CLs for U.S. ecoregions. The most notable new work is the U.S.
Department of Agriculture—Forest Service (USDA-FS)^ ssessmeni of Nitrogen
Deposition Effects and Empirical Critical Loads (Pardo et al.. 201 la). That assessment
56
-------
was organized by Level 1 ecoregions, and where data were available, CL calculations
were made for individual ecosystem types (e.g., forests within the Mediterranean
California ecoregion) and life forms (i.e., lichens, mycorrhizal fungi). This ISA largely
follows that structure, reporting terrestrial N CLs for life forms (e.g., herbaceous plants)
within each ecoregion, which is a geographically defined area within a broader biome
(e.g., forests) based on distinct physical and biological features (e.g., Northwest Forested
Mountains, Eastern Temperate Forests, etc.).
Newer CL studies are presented in tandem with the CLs reported by Pardo etal. (2011a)
in Table 6-28 and Figure IS-10. The majority of values for new CLs are within the range
of CLs identified by Pardo et al. (201 la). Notably, however, Simkin et al. (2016)
identified a new average CL for herbaceous plants in open canopy (7.9 kg N/ha/yr)
forests in the Eastern Temperate Forest ecoregion, and new lower CLs were derived for
alpine ecosystems in the Northwest Forested Mountains ecoregion. There are also new
CLs for herbaceous species in two ecoregions previously lacking a CL for herbaceous
plants [Table 6-28, Simkin et al. (2016)1.
Recently, Clark et al. (2018) estimated CL exceedance areas for the contiguous U.S. over
a more than 200 year period. Overall, this analysis showed that terrestrial N CLs have
been exceeded for many decades in areas across the U.S. Exceedance areas peaked in
1995 for changes in lichen communities and plant community composition at 3.47 and
2.87 million km2, respectively, before declining marginally by 2006. The minimum forest
tree health CL was exceeded in 2.41 million km2 by 1855 and did not change much over
time, primarily because the relatively low CL compared to deposition values in the
Eastern Temperate and Northern forest ecoregions.
57
-------
Southern Semi-Arid Highlands

No Data
1






Temperate Sierras
4-7
O
o






Mediterranean California



5-33
17-39



3-6
7.8-9.2
o





North American Deserts
OQwu O
3
O







O
5-25 O





Great Plains









Northwest Forested Mountains
M •
4-10 O
•






2.5- 7.1Q
5-10


Marine West Coast Forests
5 O
CO 0-7- 9.2
5
o






Eastern Temperate Forests
<3
•
o
4-BO O
5-12
o
o
wO
5-7
• 1
17.5





Northern Forests
<3
Or-2i






0	5	10	15	20	25	30	35	40	45
kg N/ha/yr
CL = critical load; ha = hectare; kg = kilogram; N = nitrogen; yr = year.
The rectangles indicate the range of CLs designated by Pardo et al. (2011a1: the circles indicate new papers that have specified
CLs; data from Table 6-28.
Figure IS-10 Summary of critical loads for nitrogen in the U.S. for shrubs and
herbaceous plants (yellow), trees (blue), lichens (green), and
mycorrhizae (gray).
IS.5.3 Biological Effects of Acidification
Since publication of the 2008 ISA, the overarching understanding of terrestrial
acidification has not appreciably changed. Recent research has confirmed and
strengthened this understanding that acidification can be caused by acidifying deposition
(N + S) and provided more quantitative information, especially across regional-scale
landscapes. Several studies have evaluated the relationships between soil chemistry
indicators of acidification and ecosystem biological endpoints (see Table 5-6). Soil
chemistry indicators examined in recent literature include exchangeable base cations
(Be), soil pH, exchangeable acidity (H+ and Al), exchangeable Bc:Al ratio, base
58
-------
saturation, and A1 concentrations. The most common indicator used in determining CLs
is the soil solution Bc:Al ratio. Appendix 5.2.1 discusses the uncertainty considerations
when using this indicator. Biological endpoints included in the evaluations consisted of
physiological and community responses of trees and other vegetation, lichens, soil biota,
and fauna.
IS.5.3.1 Physiology and Growth
In the 2008 ISA, evidence was sufficient to infer a causal relationship between acidifying
deposition and changes in terrestrial biota; the evidence included changes in plant
physiology, plant growth, and terrestrial biodiversity. The physiological effects of
acidification on terrestrial ecosystems in the U.S. were well characterized at the time of
the 2008 ISA and included slower growth and increased mortality among sensitive plant
species. Consistent and coherent evidence from multiple species and studies in 2008
showed that the biological effects of acidification on terrestrial ecosystems were
generally attributable to physiological impairment caused by A1 toxicity and decreased
ability of plant roots to take up base cations (Appendix 3.2.2.3 of the 2008 ISA). Much of
the new evidence for the negative effects of acidifying deposition comes from Ca
addition experiments, in which the addition of Ca has alleviated many of the negative
plant physiological and growth effects. Consistent with the findings of the 2008 ISA, the
body of evidence is sufficient to infer a causal relationship between acidifying N and
S deposition and the alteration of the physiology and growth of terrestrial organisms
and the productivity of terrestrial ecosystems.
In the 2008 ISA, acidifying deposition, in combination with other stressors, was found to
be a likely contributor to physiological effects that led to the decline of sugar maple (Acer
saccharum) trees occurring in portions of the eastern U.S. with base-poor soils. Studies
since the 2008 ISA support these findings (see Appendix 5.2.1.1). For example, recent
field studies have shown relationships between soil chemical indicator threshold values
and tree responses. Substantial declines in sugar maple regeneration have been found at
soil base saturation levels <20%, which is consistent with the range reported in the 2008
ISA.
In new studies, sugar maple grew more rapidly and showed increased regeneration
responses with increasing exchangeable base cations, base saturation, and soil pH,
however, growth was stunted and regeneration reduced with increasing exchangeable Al.
In other studies, the growth, regeneration, and physiological responses of sugar maple to
the soil conditions created by acidifying deposition were reversed or ameliorated by Ca
additions. Similarly, the 2008 ISA reported that processes associated with soil
59
-------
acidification contributed to physiological stress, high mortality rates, and decreasing
growth trends of red spruce (Picea rubens) trees. New evidence from Ca addition studies
provides further support for these mechanisms (see Appendix 5.2.1.2). Added Ca
reversed or ameliorated many of the physiological responses to acidification.
In the 2008 ISA, there was limited information on the effects of acidification on other
tree species. Since the 2008 ISA, research has observed varying physiological sensitivity
to soil acidification among eight eastern U.S. tree species. New studies since the 2008
ISA have also added new information about the effects of acidifying deposition on forest
understory vegetation, grasslands, lichens, and higher trophic level organisms (snails and
salamanders) that support the terrestrial acidification conclusions of the 2008 ISA.
IS.5.3.2 Biodiversity
The 2008 ISA noted strong evidence that acidifying deposition could alter terrestrial
community composition and cause a loss of terrestrial biodiversity. The physiological and
growth effects of acidifying deposition are not uniform across species, resulting in altered
species composition and decreased biodiversity whereby sensitive species are replaced by
more tolerant species. For example, increasing soil base cation availability was tied to
greater sugar maple growth and seedling colonization, whereas American beech (Fagus
grandifolia) was relatively more dominant on soils with lower base cation availability
(see Appendix 5.2.1.3.1). Measurements of soil acid-base chemistry have been used as a
predictor of understory species composition, with 50 understory species associated with
high soil base cation status. In another set of studies, soil acid-base chemistry was
correlated with soil biodiversity and community composition. For example, addition of
Ca resulted in changes in soil bacterial community composition and bacterial community
structure that were correlated with soil exchangeable Ca, pH, and P (see Appendix 5.2.4).
Based on research included in the 2008 ISA and these new studies, the body of evidence
is sufficient to infer a causal relationship between acidifying N and S deposition and
the alteration of species richness, community composition, and biodiversity in
terrestrial ecosystems.
IS.5.3.3 National-Scale Sensitivity and Critical Loads
The sensitivity of soils to acidifying deposition is discussed in detail in Appendix 4. In
general, surficial geology is the principal factor governing the sensitivity of terrestrial
ecosystems soil to acidification from S and N deposition. Other factors that contribute to
the sensitivity of soils to acidifying deposition include topography, soil chemistry, and
60
-------
land use. Several widely accepted models are currently used in the U.S. to assess the
terrestrial effects of S and N deposition (Appendix 4.5). These models are typically used
to evaluate acidification effects on biota by assigning a value of a soil parameter that
relates to the onset of a harmful biological effect. Since the 2008 ISA, estimates of base
cation weathering (BCw), which are input to soil acidification models have improved and
are being applied for deriving new CLs in the U.S. Forests of the Adirondack Mountains
of New York, Green Mountains of Vermont, White Mountains of New Hampshire, the
Allegheny Plateau of Pennsylvania, and mountain tops and ridges forest ecosystems in
the southern Appalachians are the regions that are most sensitive to terrestrial
acidification from atmospheric deposition (Appendix 3.2.4.2 of the 2008 ISA).
Models used to determine CLs of acidifying deposition included SMB, STA, MAGIC,
ForSAFE-VEG, and empirical models. Several models and extrapolation methods to
estimate BCw rates were also investigated. The PROFILE model was evaluated as a
model to estimate soil BCw rates to support estimates of SMB CLs in the U.S. (see
Appendix 4.5). In general, recently published models used soil solution Bc:Al ranging
from 1.0 to 10.0 as an indicator to estimate CLs in North America.
Ecosystem sensitivities to ambient N and S deposition were also characterized by
developing CLs and exceedances (see Appendix 4.6; Figure IS-11. and Appendix 5.5).
Calculated CLs for forest plots based on the soil solution Bc:Al of 10.0 in the
northeastern U.S. ranged from 11 to 6,540 eq/ha/yr (eq quantifies the supply of available
H+ ions, combining the acidifying effects of N and S deposition), and 15-98% (calculated
using maximum and minimum weathering rates) of these plot-level CLs were exceeded
by N and S deposition. In this region, correlation analyses showed that the growth of
17 tree species were negatively correlated with CL exceedance. In Pennsylvania, CLs
based on the soil solution Bc:Al of 10.0 for hardwood forests ranged from 4 to
10,503 eq/ha/yr and were exceeded by estimated N and S deposition in the year 2002 in
53% of the plots. Several studies found that CL and exceedance determinations could be
influenced by BCw rates, soil chemical indicators, N retention, tree species-specific base
cation uptake, and the type and accuracy of deposition estimates (i.e., wet, bulk, total,
measured or modeled).
61
-------
Forest Ecosystems Critical Loads for Acidity
S + N eq/ha/yr
¦¦ 170- 1,000
H 1,001-2,000
2,001-4,000
| 4,001-6,000
6,001-8,800
States
No Data
eq = equivalent; ha = hectare; yr = year.
(A) McNultv et al. (20071; CLs are mapped at 1 -km2 grids (center map). For uncertainty, see Li and McNultv (2007). (B) Duarte et al
(2013); CLs are mapped at 4-km2 grids. (C and D) Phelan et al. (2014); CLs are mapped for each sampling site (Pennsylvania),
McDonnell et al. (2014); Sullivan et al. (2011b); Sullivan etal. (2011a"'; CLs are mapped as a single point at the center point of the
watershed (New York and North Carolina).
Source: http://nadp.slh.wisc.edu/committees/clad.
Figure IS-11 Forest ecosystem critical loads for soil acidity related to base
cation soil indicators.
IS.6 Freshwater Ecosystem Nitrogen Enrichment and
Acidification
For freshwater systems, new evidence reinforces causal findings from the 2008 ISA
(Table IS-1). It also expands the scope of existing causal findings to include additional
biota affected by N enrichment and acidifying deposition and supports quantification of
these effects with new CLs (see Section IS.6.3.2). Freshwater systems include lakes
(lentic systems) and rivers and streams (lotic systems). In freshwater ecosystems, N may
cause N enrichment/eutrophication. Aquatic eutrophication results in increased
62
-------
productivity of algae and aquatic plants, altered nutrient ratios, and sometimes decreased
oxygen levels. Deposition of N, S, or N + S can cause acidification, which affects
watershed biogeochemical processes and surface water chemistry. Freshwater N
enrichment and acidification take place in sensitive ecosystems across the U.S. at present
levels of deposition and may occur simultaneously in some water bodies.
New studies have added to the body of evidence in the 2008 ISA that N nutrient
enrichment and acidifying deposition alter freshwater biogeochemistry with subsequent
biological effects. There is new information on biogeochemical processes including
cycling of N and S. Both N enrichment/eutrophication and acidification can impact
physiology, survival, and biodiversity of sensitive aquatic biota. The 2008 ISA and new
studies provide examples of lakes and streams that show signs of eutrophication,
especially increased algal growth and shifts in algal biodiversity, in response to N
addition. The current causal statement for nutrient enrichment effects of N deposition
now includes altered algal growth and productivity as well as the endpoints of species
richness, community composition, and biodiversity reported in the 2008 ISA
(Table IS-1). For biological effects of aquatic acidification, the current causal statement
has been expanded from the 2008 ISA to include the specific endpoints of physiological
impairment, alteration of species richness, community composition, and biodiversity
(Table IS-1). New studies also show that despite reductions in acidifying deposition,
many aquatic ecosystems across the U.S. are still experiencing changes in ecological
structure and functioning at multiple trophic levels. Although there is evidence for
chemical recovery in many previously acidified ecosystems, biological recovery has been
limited (Appendix 8.4).
A number of freshwater monitoring efforts have facilitated the analysis of long-term
trends in surface water chemistry and ecological response in areas affected by acidifying
(N + S) deposition (Appendix 7.1.3). Many of these studies have been conducted in the
U.S., especially in the Northeast and the Appalachian Mountains. Although many of
these monitoring programs were in existence at the time of the 2008 ISA and were
considered in that analysis, more recent publications reflect the longer period of
monitoring and strengthen previous conclusions. Surface water chemistry data from
long-term monitoring by federal, state, and local agencies, as well as university research
groups and nonprofits has been combined into several publicly available metadatabases
(Appendix 7.1.3.2) enabling further regional trend analysis. Since the early 2000s,
U.S. EPA, together with the states, tribes, and other entities and individuals, have
collaborated on a series of statistically representative surveys (National Aquatic Resource
Surveys [NARS]) of the nation's waters, including surveys of lakes (U.S. EPA. 2016c.
2009b). streams (U.S. EPA. 2016f). wetlands (U.S. EPA. 2016g). and coastal waters
(U.S. EPA. 2016d). These periodic surveys, which are based on standard sampling and
63
-------
analysis protocols and consistent quality assurance, include chemical and biological
indicators of nutrient enrichment and acidification (Appendix 7.1.3).
IS.6.1 Freshwater Biogeochemistry
In the 2008 ISA, evidence was sufficient to infer a causal relationship between N and S
deposition and the alteration of biogeochemical cycling of N and C in freshwater
ecosystems, and between acidifying deposition and changes in biogeochemistry of fresh
waters. As documented in the 2008 ISA and by newer studies, biogeochemical processes
and surface water chemistry are influenced by characteristics of the catchment and the
receiving waters. A number of studies since 2008 have focused on improving
understanding of aquatic acidification and eutrophication processes mediated by N. Many
of these studies have focused on pathways of pollutant and other constituent movement
within ecosystems, including monitoring studies of various kinds. Chemical indicators of
N deposition identified by the 2008 ISA were NOs and DIN concentrations in surface
waters. Increased N deposition to freshwater systems via runoff or direct atmospheric
deposition, especially to N limited and N and phosphorus (P) colimited systems, can alter
N cycling (Appendix 7) and stimulate primary production (Appendix 9). Data from
long-term monitoring, experimental manipulations, and modeling studies provide
consistent and coherent evidence for biogeochemical changes associated with acidifying
N and S deposition. The strongest evidence for a causal relationship between acidifying
deposition and aquatic biogeochemistry comes from studies of changes in surface water
chemistry. Surface water chemistry indicators of acidic conditions and acidification
effects include concentrations of SO42 . NO3 . inorganic aluminum (Al), calcium (Ca),
sum and surplus of base cations, acid-neutralizing capacity (ANC), and surface water pH.
New information on biogeochemical cycling of N and S, acidifying deposition effects on
biogeochemical processes and changes to chemical indicators of surface water chemistry
associated with acidification and N nutrient enrichment is consistent with the conclusions
of the 2008 ISA, and the body of evidence is sufficient to infer a causal relationship
between N and S deposition and the alteration of freshwater biogeochemistry.
IS.6.1.1 Freshwater Processes and Indicators
Key processes and geochemical indicators of freshwater acidification and N enrichment
(Table IS-3) link to biological effects (Appendix 8 and Appendix 9). Surface water
chemistry integrates the sum of soil and water processes that occur upstream within a
watershed. Several key biogeochemical processes cause or contribute to surface water
eutrophication and acidification, and these processes have been the focus of substantial
64
-------
research over the last three decades. Since the 2008 ISA, experimental studies, isotopic
analyses, and monitoring and observational studies have further investigated the cycling
of S, N, C, and base cations; these studies substantiate and further quantify earlier
findings.
Spatial and temporal patterns of NO, in lakes and streams have typically been used as
indicators that a freshwater system is receiving excess N which will cause acidification or
eutrophication. Qualitatively, northeastern U.S. spatial patterns in surface water NO,
concentrations suggest an influence by atmospheric N deposition. However, considerable
variation in the relationship between stream chemistry and deposition was associated
with land use and watershed attributes. It was well known at the time of the 2008 ISA
that key processes such as nitrification and denitrification are quantitatively important
portions of the N cycle and that they can be influenced by atmospheric inputs. More
recent research has further substantiated these earlier findings and provided additional
quantitative context (Appendix 7.1.2.3).
Deposition is a source of S to watersheds that, along with geologic sources of S such as
sulfide minerals, contribute SO42 to surface waters (Appendix 4). The 2008 ISA found
that S deposition alters soil and drainage water chemistry through sustained leaching of
SO42 , associated changes in soil chemistry, and accumulation of S in the soil through
adsorption and biological assimilation. Declines in lake SO42 concentrations have been
observed in locations where S deposition has decreased significantly such as in the
Adirondack Mountains (Appendix 7.1.5.1). In addition, internal watershed sources of S,
which were earlier believed to be relatively minor in the northeastern U.S., have and will
likely continue to become proportionately more important as S deposition continues to
decline. Reductions in SOx deposition have not consistently resulted in increases of ANC
in surface water.
Table IS-3 Summary of key aquatic geochemical processes and indicators
associated with eutrophication and acidification.

N Driven


Nutrient

Endpoint
Enrichment
Acidification The Effect of Deposition
Process
NO3-
X
X Leaching from terrestrial ecosystems is an important source of NO3 in
leaching into

freshwater ecosystems. See NO3" leachinq in Table IS-2.
water bodies


65
-------
Table IS-3 (Continued): Summary of key aquatic geochemical processes and
indicators associated with eutrophication and
acidification.
Endpoint
N Driven
Nutrient
Enrichment
Acidification
The Effect of Deposition
so42~
leaching into
water bodies

X
Leaching from terrestrial ecosystems is an important source of SC>42~
in freshwater ecosvstems. See SO42" leachina in Table IS-2.
Nitrification
X
X
Nitrification is an acidifying process, releasing 2 mol hydrogen ion (H+)
per mol NhV converted to NO3". As the N cycle becomes enriched
through cumulative N addition, net nitrification rates often increase,
and NO3" concentrations increase.
Denitrification
X

Denitrification is the microbial process that transforms NO3" by
anaerobically reducing it to NO2", NO, N2O, and N2.
DOC
leaching into
water bodies
X
X
DOC contributes to acidity of freshwater ecosystems. See DOC
leachina in Table IS-2.
Indicator
Surface
water [NO3"]
X
X
Increased N deposition (to surface waters or to terrestrial watershed;
see Table IS-2) increases the water NOs" concentration.
High concentrations of NO3" in lakes and streams, indicative of
terrestrial ecosystem N saturation, have been found at a variety of
locations throuahout the U.S. (U.S. EPA. 2006b: Stoddard. 1994).
Comparison of preindustrial estimates to modern measurements
suggested elevated concentrations in water bodies as a result of N
deposition (Fenn et al.. 2011).
Surface
water DIN
X

Increased N deposition increases DIN in most freshwater aquatic
environments, largely as NO3".
Surface
water N:P
ratios
X

Increased N deposition can alter the ratio of N to P in freshwater
systems. Freshwater biota have different nutrient requirements and
changes in nutrient ratios may alter species richness, community
structure, and biodiversity, especially primary producers.
Surface
water [SO42"]

X
Increased S deposition (to surface waters or to terrestrial watershed,
see Table IS-2) increases the water SO42" concentration.
Comparison of preindustrial estimates to modern measurements
suggested elevated concentrations in water bodies are a result of S
deposition.
Surface
water (base
cation)

X
Several studies in the eastern U.S. suggested that base cation
concentrations in surface waters increased during the initial phases of
acidification into the 1970s. This trend reversed, and base cations
have decreased primarily in response to decreasing SO42" and NO3"
concentrations. Many base cations (especially Ca2+) are important
nutrients for aquatic biota.
66
-------
Table IS-3 (Continued): Summary of key aquatic geochemical processes and
indicators associated with eutrophication and
acidification.
N Driven
Nutrient
Endpoint Enrichment Acidification	The Effect of Deposition
Surface	X	Increased N and S deposition decrease ANC. Surface water ANC
water ANC	correlates with other biologically influential chemical metrics, including
pH, inorganic Al concentration, Ca concentration, and organic acidity.
ANC <50-100 [jeq/L typically poses a risk for species survival, species
richness, and biodiversity.
X	Surface water pH is a common alternative to ANC as an indicator of
acidification, but ANC is a better indicator at pH >6.0 and is less
sensitive to dissolved CO2. N and S deposition are associated with
decreasing pH in surface waters.
Acidifying N and S deposition increase mobilization of inorganic Al
from terrestrial ecosystems into surface water, increasing surface
water concentrations. Inorganic Al in surface waters is (1) widely toxic
and (2) leaches from terrestrial ecosystems only in response to acidic
conditions. Earlier studies demonstrated reduced growth and survival
of various species offish at inorganic Al concentrations between
approximately 2 and 7.5 pmol/L. Most recently, 20% mortality of
young-of-the year brook trout was documented in situ during a 30-day
period with a median inorganic Al concentration of 2 pmol/L.
Al = aluminum; ANC = acid-neutralizing capacity; Ca = calcium; C02 = carbon dioxide; DIN = dissolved inorganic nitrogen;
DOC = dissolved organic carbon; H+ = hydrogen ion; ha = hectare; kg = kilogram; L = liter; |jeq = microequivalents;
|jmol = micromole; N = nitrogen; N2 = molecular (atmospheric) nitrogen; N20 = nitrous oxide; NE = northeast; NH4+ = ammonium;
NO = nitric oxide; N02 = nitrogen oxide; N03" = nitrate; P = phosphorus; S = sulfur; S042" = sulfate; U.S. = United States;
USFS = U.S. Forest Service; yr = year.
IS.6.1.1.1 Acidification
The acidifying effects of N and S deposition in U.S. waters have been well characterized
for several decades. Traditionally, acidification involves both chronic and episodic
processes. Driscoll et al. (2001) characterized chronically acidic lakes and streams as
having ANC of <0 |icq/L throughout the year, while episodic acidification occurs when
ANC falls below 0 (j,eq/L only for hours to weeks. Chronic acidification refers to average
conditions and is often measured as summer and fall chemistry for lakes and as spring
baseflow chemistry for streams. Chronic acidification is no longer prevalent in regions of
the U.S. affected by acidic deposition (Fakhraci et al.. 2016; Fakhraei et al.. 2014).
Episodic acidification is associated with precipitation or snowmelt events when high
volumes of water flow through watersheds. Episodes generally cause changes in the
following chemical parameters: ANC, pH, base cations, SO42 concentration, NO.?
concentration, inorganic Al concentration, organic acid anions, or DOC. New studies
show that both N and S contributed to episodic acidification over a 20-year period at Bear
Surface
water pH
Surface	X
water
Inorganic Al
67
-------
Brook, ME (see Appendix 7.1.5.1.2). It is known that the biota in many streams/lakes are
impacted when the ANC is consistently below 50 j^icq/L. For example, the U.S. EPA
National Lakes Assessment used an ANC threshold of >50 j^icq/L as indicative of
nonacidified water bodies (U.S. EPA. 2009b).
The most widely used measure of surface-water acidification is ANC. As reported in the
2008 ISA and newer studies, ANC is the primary chemical indicator of historic
acidification and for predicting the recovery expected from decreasing atmospheric
deposition. ANC correlates with the surface water constituents (pH, Ca2+, and inorganic
A1 concentration) that contribute to or ameliorate acidity effects in biota. As reported in
the 2008 ISA, lake and stream ANC values decreased throughout much of the 20th
century in a large number of acid-sensitive lakes and streams throughout the eastern U.S.
This effect has been well documented in monitoring programs, paleolimnological studies,
and model simulations (Appendix 7.1.5.1). Biological indicators of acidification, such as
decreased fish species richness, are discussed in Appendix 8.3.
Surface water pH is another indicator of acidification. It also correlates with surface
water chemical constituents that have biotic effects (inorganic Al, Ca2+, and organic
acids). The 2008 ISA included the scientific consensus that low pH can have direct toxic
effects on aquatic species (U.S. EPA. 2008; Driscoll et al.. 2001). A pH value of 6.0 is
the level below which biota are at increased risk from acidification (Appendix 8.3). The
2008 ISA noted that increasing trends in pH (decreasing acidification) were common in
surface waters in the northeastern U.S. through the 1990s and up to 2004. This trend has
continued in more recent times at many locations (Appendix 7.1.2.5). Rates of change
have generally been relatively small.
As stated in the 2008 ISA, the concentration of dissolved inorganic monomeric Al in
surface waters is an especially useful indicator of acidifying deposition because (1) it is
toxic to many aquatic species and (2) it leaches from soils only under acidic conditions
including acidifying deposition, acid mine drainage, or from rare geologic deposits.
Inorganic Al has well-documented effects on aquatic biota at specific thresholds
(Appendix 8.3) and is often the greatest threat to aquatic biota below pH 5.5. The 2008
ISA noted that concentrations of inorganic Al decreased slightly in some surface waters
in the northeastern U.S. in response to decreased levels of acidifying deposition,
suggesting chemical recovery in some of these surface waters (U.S. EPA. 2008). and this
trend has generally continued (Appendix 7.1.5; see discussion on recovery
Section IS. 11V
Assessments of acidifying deposition effects dating from the 1980s and reported in the
2008 ISA showed S042 to be the primary acidifying ion in most acid-sensitive waters in
the U.S. The 2008 ISA presented temporal data that showed a trend of increasing
68
-------
concentrations of S042 in surface waters before the period of peak S emissions in the
early 1970s. After the peak, SO42 surface water concentrations decreased in a
widespread trend. The rate of recovery varied by ecosystem, and new studies indicate that
as atmospheric S deposition has declined, soils with large stores of historically deposited
S (e.g., the Blue Ridge Mountain region) have begun releasing this adsorbed S to
drainage water (Appendix 4), preventing or slowing aquatic recovery.
As stated in the 2008 ISA, the quantitatively most important component of the overall
surface water acidification and chemical recovery responses has been the change in base
cation supply. Decreases in base cation concentrations in surface waters in the eastern
U.S. have been ubiquitous over the past two to three decades and closely tied to trends in
S042 concentrations in surface waters. Change in base cation supply with surface water
acidification was highlighted in Charles and Christie (1991) and in the 2008 ISA. Base
cations are added to watershed soils by weathering of minerals and atmospheric
deposition, and are removed by uptake into growing vegetation or by leaching. Acidic
deposition increased leaching of base cations, because SO42 anions percolating through
the soil tend to carry base cations along with them to maintain the charge balance. In
watersheds that received high levels of historical acidic deposition, current exchangeable
concentrations of Ca2+ and other base cations are substantially reduced from likely
preindustrial levels, having been depleted by many years of acidic deposition. This base
cation depletion in watersheds constrains ANC and pH recovery of surface waters, as
described in the 2008 ISA. New studies of base cations, which include experiments,
modeling, and gradient studies, have further corroborated these earlier findings.
Changes in DOC concentration or properties can affect the acid-base chemistry of surface
waters and perhaps the composition of aquatic biota. In soils and water, DOC constitutes
only a portion of dissolved organic matter (DOM), which also includes other constituents
such as organic nitrogen, phosphorus, and sulfur. However, the very large majority of
studies that include DOC do not explicitly include all of DOM. It has been recognized
that surface water DOC concentrations decreased to some extent as a result of
acidification, and that these concentrations would likely increase with recovery.
However, the strength of this response and the magnitude of DOC changes have
exceeded scientific predictions. Recent research on this topic has been diverse and has
included experiments, observation, isotope studies, and synthesis and integration work.
Overall, these studies illustrate large increases in DOC with acidification recovery in
some aquatic systems. Increases in DOC constrain the extent of ANC and pH recovery,
but decrease the toxicity of dissolved A1 by converting some of it from inorganic to
organic forms (Lawrence et al.. 2013). However, DOC is not an indicator of recovery
everywhere; some recovering sites have not shown increasing trends in DOC.
69
-------
IS.6.1.1.2 Nitrogen Enrichment/Eutrophication
In aquatic systems, N is a nutrient that stimulates growth of primary producers (algae
and/or aquatic plants). Atmospheric deposition of N to freshwater systems can increase
the absolute supply of nutrients and alter N and P ratios. The freshwater ecosystems in
the U.S. most likely to be sensitive to nutrient enrichment from N deposition are
headwater streams, lower order streams, and alpine lakes, which have very low nutrients
and productivity and are far from local pollution sources ITJ.S. EPA (2008);
Appendix 9.1.1.4], These nutrient shifts alter stoichiometric composition of water
chemistry, thereby shifting the nutrient status of lakes. Even small inputs of N in low
nutrient water bodies can affect biogeochemical processing of N and increase the
productivity of photosynthesizing organisms, resulting in a larger pool of fixed carbon
(C). Nutrient enrichment leads to changes in aquatic assemblages and biodiversity in
freshwater (Appendix 9) and coastal regions (Appendix 10).
Indicators of altered N cycling include changes in the concentrations of NOs in surface
waters. The concentration of NOs in water provides an index of the balance between
removal and addition of N to terrestrial ecosystems. Studies of several types have been
conducted in recent years to elucidate these processes and include experimental studies,
isotopic analyses, and monitoring and observational studies. Both water column and
sediment N transformations have been further characterized (Appendix 7.1.2.3). New
research suggests that denitrification may, in some situations, play a larger role than was
previously recognized in the 2008 ISA in removing oxidized N from the watershed.
As reported in the 2008 ISA and in newer studies, atmospheric N has been positively
correlated to total N in lakes along gradients of atmospheric deposition. N deposition in
some high-deposition lakes has changed the nutrient status of these lakes from a
more-or-less balanced (mainly N deficient) state to more consistently P limited
conditions (Appendix 9.2.4). Since the 2008 ISA, several studies have reported increases
in P deposition to water bodies in the U.S., possibly affecting shifts in lake trophic status
from P to N limitation or colimitation, as well as prolonging N limitation
(Appendix 9.1.1.2). In higher order streams, N deposition typically mixes with N derived
from other nonatmospheric sources, including urban/suburban point and nonpoint
sources, industrial sources, and agricultural sources, with atmospheric sources typically
being most pronounced during high flow conditions (Table 7-2).
IS.6.1.2 Models
Models used to assess the effects of N and S deposition on U.S. ecosystems were
reviewed in the 2008 ISA (Annex A). Several of the models used for terrestrial
70
-------
ecosystems (see Section IS.5.3.3) such as MAGIC and PnET/BGC are also applicable to
aquatic systems. Both models have been widely applied, mainly to relatively small,
upland watersheds. Three other models, Spatially Referenced Regressions on Watershed
Attributes (SPARROW), Watershed Assessment Tool for Evaluating Reduction
Scenarios for Nitrogen (WATERS-N), and Surface Water Assessment Tool (SWAT)
have been used to evaluate N loading to mixed-use watersheds in larger river systems.
Another model that has been applied to the analysis of nutrient enrichment in aquatic
systems is AQUATOX, which simulates nutrient dynamics and effects on aquatic biota.
Few new freshwater acidification or eutrophication models have been developed and
published since 2008. A new national water quality modeling system (Hydrologic and
Water Quality System, HAWQS) is under development by Texas A&M University and
the USDA for the U.S. EPA's Office of Water (https: //epahawq s .tamu.edu/). The model
is intended to assist resource managers and policy makers in evaluating the effectiveness
of water pollution control efforts. Freshwater eutrophication and acidification models are
described in greater detail in Appendix 7.1.4.2.
IS.6.1.3 National-Scale Sensitivity
Sensitivity of lakes, streams, and rivers to biogeochemical changes associated with N and
S deposition varies across the U.S. The biogeochemical sensitivity to acidifying
deposition will be discussed together with biological sensitivity in Section IS.6.2.2.
Sensitivity to N enrichment will be discussed with biological sensitivity in
Section IS.6.3.2.
IS.6.2 Biological Effects of Freshwater Nitrogen Enrichment
In the 2008 ISA, evidence was sufficient to infer a causal relationship between N
deposition and the alteration of species richness, community composition, and
biodiversity in freshwater ecosystems. The freshwater systems most affected by nutrient
enrichment due to atmospheric deposition of N were remote oligotrophic high-elevation
lakes with low N retention capacity. In these ecosystems, N changes the biota, especially
by increasing algal growth and shifting algal communities. Freshwater organism
responses to N enrichment can be assessed through biological indicators, including
chlorophyll a, phytoplankton and periphyton (algae attached to a substrate) biomass,
diatoms, and trophic status. The current causal statement has been expanded to include
effects on algal growth and productivity (Table IS-1). New evidence since 2008 of N
enrichment includes paleolimnology, phytoplankton community dynamics,
macroinvertebrate response, and indices of biodiversity. This new evidence is consistent
71
-------
with the conclusions and strengthens the evidence base of the 2008 ISA, and together, the
body of evidence is sufficient to infer a causal relationship between N deposition and
changes in biota, including altered growth and productivity, species richness,
community composition, and biodiversity due to N enrichment in freshwater
ecosystems.
IS.6.2.1 Physiology and Biodiversity Effects
Inputs of N to freshwater systems stimulate algal growth, which leads to a cascade of
effects on algal community composition and biodiversity. Algal species have differential
responses to N loading and shifts in nutrient ratios, so dominant species may change in
response to N enrichment. As reported in the 2008 ISA and in newer studies, shifts in
nutrient limitation from N limitation to colimitation by N and P, or to P limitation, have
been observed in some alpine lakes. New biodiversity studies are summarized in
Table 9-3. Since the 2008 ISA, several meta-analyses have reported an increase in P
atmospheric deposition to water bodies, highlighting the need to account for how
sustained P deposition can modify the effects of anthropogenically emitted N deposition
on productivity (Appendix 9.1.1.4). P addition delays the shift to P limitation (prolonged
N limitation) for phytoplankton.
IS.6.2.1.1 Primary Producers
The body of evidence for biological effects of N enrichment in remote freshwater
systems (where atmospheric deposition is the predominant source of N) is greatest for
phytoplankton, the base of the freshwater food web. Most studies focused on
phytoplankton, although several new studies indicate that both benthic and pelagic
primary producers respond to N inputs, and at least some studies have shown that
periphyton outcompeted phytoplankton for limiting nutrients (Appendix 9.3.3). The 2008
ISA and new studies include lake surveys, fertilization experiments, and nutrient
bioassays that show a relationship between increased N concentrations in the water
column and increased pelagic and benthic algal productivity (measured by chlorophyll a
concentration). An increase in lake phytoplankton biomass with increasing N deposition
was reported in the Snowy Range in Wyoming and in Europe. New studies in the
Colorado Rocky Mountains, where atmospheric deposition ranged from 2 to 7 kg
N/ha/yr, found correlations between higher chlorophyll a and higher rates of deposition
(Appendix 9.2.1).
72
-------
The 2008 ISA and newer studies (Table 9-1 and Appendix 9.3.2) show a general shift in
algal dominance from chrysophytes that dominate low N lakes to cyanophytes and
chlorophytes in higher N lakes. Two nitrophilous species of diatom, Asterionella formosa
and Fragilaria crotonensis, serve as indicators of N enrichment in lakes; however,
increased relative abundance of A. formosa has also been attributed to lake warming in
some regions where N deposition is decreasing (Appendix 9.3.2). New studies show that
glacial meltwater has higher NOs relative to snow meltwater with different influences on
algal community composition in some regions of the U.S. (Appendix 9.3.2). In a
comparison of lakes in the Rockies with different meltwater sources, fossil diatom
richness in snowpack-fed lakes was at least double the richness of lakes with both glacial
and snow meltwater inputs; however, alterations in phytoplankton community structure
were not observed in lakes in the northern Cascade Mountains, WA. Some shifts in algal
biodiversity observed in high-elevation waters are attributed to climate change or nutrient
effects and climate as costressors (Appendix 13).
The role of N in freshwater harmful algal bloom formation has been further researched
since the 2008 ISA. Additional evidence continues to show that availability and form of
N influences algal bloom composition and toxicity, and inputs of inorganic N selectively
favor some HAB species, including those that produce microcystin. Microcystin is
prevalent in U.S. waters as reported in recent regional and national surveys. The risk of
HAB formation is low in high-elevation oligotrophic water bodies where N deposition is
the dominant source of N, but transport of atmospheric inputs can exacerbate eutrophic
conditions in downstream water bodies. Increased understanding of the role of N as a
limiting nutrient in many freshwater systems has led to recommendations to consider
both N and P in nutrient-reduction strategies.
Few studies in the U.S. have considered the effects of atmospheric deposition on aquatic
macrophytes, although declines in macrophyte occurrence were noted in a new survey of
Lake Tahoe that compared the lake's biota with that from a survey conducted in the
1960s (Caires et al.. 2013). Atmospheric N contributions are a substantial portion
(approximately 57%) of the total N loading to Lake Tahoe.
IS.6.2.1.2 Zooplankton
Compared to changes in primary producers, biological responses to N deposition at
higher trophic levels are not well characterized, but atmospheric N can alter food web
interactions (see Appendix 9.3.4). A few studies in the 2008 ISA and newer studies
showed zooplankton responses to N related shifts in phytoplankton biomass potentially
altering food web interactions.
73
-------
IS.6.2.1.3 Macroinvertebrates
Few studies published since the 2008 ISA have linked atmospheric N deposition to
taxonomic shifts and declines in invertebrates (Appendix 9.3.5). These studies do not
attribute shifts in the abundance of higher invertebrates to N deposition alone, because
their abundance is also determined by additional factors such as climate and the presence
of invasive species. New studies provide additional evidence that trophic interactions
may moderate algal growth following nutrient loading. In Lake Tahoe, which receives
57% of N inputs from atmospheric sources, endemic invertebrate taxa have declined 80
to 100% since the 1960s due to nutrient inputs and invasive species.
IS.6.2.2 National-Scale Sensitivity and Critical Loads
New data have not appreciably changed the identification of sensitive lakes and streams
in the U.S. since the 2008 ISA. Nutrient enrichment effects from N most likely occur in
undisturbed, low-nutrient headwater and lower order streams and lakes at higher
elevations in the western U.S. (Appendix 9.1), including the Snowy Range in Wyoming,
the Sierra Nevada, and the Colorado Front Range. A portion of these lakes and streams
where effects are observed are in Class I wilderness areas which are afforded special
Clean Air Act protections. The responses of high-elevation lakes vary with catchment
characteristics (Appendix 9.1) and N deposition estimates at these high elevation sites are
associated with considerable uncertainty, especially dry deposition (Appendix 2). In these
systems, even low inputs of atmospheric N can shift N limitation to colimitation by N and
P, or to P limitation (Appendix 9.2.4), altering algal species composition and
productivity.
In the 2008 ISA, diatom assemblage shifts were reported at N deposition rates as low as
1.5 kg/N/ha/yr. Additionally, a hindcasting exercise in remote alpine Rocky Mountain
National Park lakes associated algal changes between 1850 and 1964 with an increase in
wetN deposition of 1.5 kg N/ha/yr. Since the 2008 ISA, empirical and modeled CLs for
the U.S. have been estimated based on surface water NO, concentration, diatom
community shifts, and phytoplankton biomass nutrient limitation shifts indicative of a
shift from N limitation to P limitation. A CL ranging from 3.5 to 6.0 kg N/ha/yr was
identified for high-elevation lakes in the eastern U.S. based on the nutrient enrichment
inflection point [where NOs concentrations increase in response to increasing N
deposition; Baron etal. (2011)1. Another CL of 8.0 kg N/ha/yr for eastern lakes based on
the value of N deposition at which significant increases in surface water NO3
concentrations occur was estimated by Pardo etal. (2011b). In both Grand Teton and
Yellowstone National Parks, CLs for total N deposition ranged from <1.5 ± 1.0 kg
74
-------
N/ha/yr to >4.0 ± 1.0 kg N/ha/yr (Nanus et al.. 2017). Exceedance estimates were as high
as 48% of the Greater Yellowstone area study region, depending on the threshold value
ofNCV concentration in lake water selected as indicative of biological harm. An
empirical CL of 4.1 kg N/ha/yr above which phytoplankton biomass P limitation is more
likely than N limitation was identified by Williams et al. (2017b) for the western U.S.
Modeled CLs ranged from 2.8 to 5.2 kg/N/ha/yr.
IS.6.3 Biological Effects of Freshwater Acidification
The 2008 ISA found evidence sufficient to infer a causal relationship between acidifying
deposition and changes in aquatic biota, including strong evidence that acidified aquatic
habitats had lower species richness of fishes, macroinvertebrates, and phytoplankton. The
effects of acidifying deposition on aquatic ecosystems also include physiological
impairment or mortality of sensitive species and shifts in biodiversity of both flora and
fauna. Organisms at all trophic levels are affected by acidification, with clear linkages to
chemical indicators for effects on algae, benthic invertebrates, and fish (Table 8-9).
Biological effects are primarily attributable to low pH and high inorganic Al
concentration. ANC integrates chemical components of acidification (Table IS-2) but
does not directly alter the health of biota.
Effects of acidification on fish species are especially well characterized and many species
are harmed. Both in situ and lifestage experiments in fish support thresholds of chemical
indicators for biological effects. Most of these effects were documented in a rigorous
review of acidification effects on aquatic biota that was included in the 2008 ISA.
Overall, the updated research synthesized in this ISA reflects incremental improvements
in scientific knowledge of aquatic biological effects and indicators of acidification as
compared with knowledge summarized in the 2008 ISA. The fundamental understanding
of mechanisms has not changed, and the causal relationships between acidifying
deposition and biological effects on aquatic ecosystems are now, and were in 2008, well
supported. New studies also show that despite reductions in acidifying deposition,
alterations in aquatic biodiversity and ecosystem functioning caused by acidification
persist. Although there is evidence for chemical recovery in many ecosystems, biological
recovery has been limited (Section IS.6.2.2). New research is consistent with the causal
determination in the 2008 ISA and has strengthened the evidence base for these effects.
The current causal statement has been expanded to include specific endpoints of
physiological impairment, as well as effects at higher levels of biological organization
(Table IS-1). The body of evidence is sufficient to infer a causal relationship between
acidifying deposition and changes in biota, including physiological impairment and
75
-------
alteration of species richness, community composition, and biodiversity in
freshwater ecosystems.
IS.6.3.1 Physiology and Biodiversity Effects
The deterioration in water quality caused by acidification affects the physiology,
survivorship, and biodiversity of many species from several taxonomic groups and at
multiple trophic levels. As stated in the 2008 ISA, biological effects are primarily
attributable to low pH (or ANC) and high inorganic A1 concentrations under chronic or
episodic acidic conditions. During acidification episodes, water chemistry may exceed
the acid tolerance of resident aquatic biota, with effects that include fish mortalities,
changes in species composition, and declines in species richness across multiple taxa.
Studies reviewed in the 2008 ISA showed that the earlier aquatic lifestages were
particularly sensitive to acidification. New effects thresholds have been identified for
aquatic organisms consistent with observations in the 2008 ISA (Table 8-10). New
evidence is congruent with findings in the 2008 ISA that high levels of acidification (to
pH values below 5 and ANC lower than the range of 50 to 100 (j,eq/L) eliminate sensitive
species from freshwater streams. This information is reviewed below.
15.6.3.1.1	Primary Producers
Phytoplankton are primary producers at the base of the aquatic food web. These
photosynthetic organisms vary in tolerance of acidic conditions and include diatoms,
cyanobacteria, dinoflagellates, and other algal groups. The 2008 ISA reported reduced
species richness of freshwater plankton in response to acidification-related decreases in
pH and increases in inorganic Al. Effects were most prevalent when pH decreased to the
5 to 6 range. Effects on productivity are uncertain. Since the 2008 ISA, several
paleolimnological and field studies have further linked phytoplankton community shifts
to chemical indicators of acidification (Appendix 8.3). For example, Lacoul etal. (2011)
reviewed information on the effects of acidification and observed that the largest declines
in phytoplankton species richness occur over a pH range of 4.7 to 5.6 in Atlantic Canada.
15.6.3.1.2	Zooplankton
Zooplankton comprise many groups of freshwater unicellular and multicellular organisms
including protozoans, rotifers, cladocerans, and copepods. Zooplankton feed on
phytoplankton or other zooplankton. Decreases in ANC and pH and increases in
76
-------
inorganic A1 concentration have been shown to contribute to the loss of zooplankton
species or abundance in lakes. In the 2008 ISA, thresholds for zooplankton community
alteration were between pH 5 and 6. In the Adirondacks, a decrease in pH from 6 to 5
decreased zooplankton richness in lakes, and at ANC <0, zooplankton richness was only
45% of the richness in unacidified lakes. Newer studies support effects in a similar pH
range (see Appendix 8.3.1.2). Zooplankton have also been used as indicators of
biological recovery (Appendix 8.4.2).
15.6.3.1.3	Benthic Invertebrates
Acidification has strong impacts on aquatic invertebrates because H+ and A1 are directly
toxic to sediment-associated invertebrates like bivalves, worms, gastropods, and insect
larvae. In the 2008 ISA and in new studies in Appendix 8.3.3, decreases in ANC and pH
and increases in inorganic A1 concentration contribute to declines in abundance or
extirpation of benthic invertebrate species in streams. Acidification to pH values below
5 eliminates mayflies (Ephemeroptera), a taxa indicative of stream water quality, along
with other aquatic organisms. Since the 2008 ISA, a survey of benthic macroinvertebrates
by Baldigo et al. (2009) in the Adirondack Mountains indicated that macroinvertebrate
communities were intact at a pH above 6.4, with moderate acidification effects at pH 5.1
to 5.7, and severe acidification effects at a pH less than 5.1. Similarly, thresholds of
pH 5.2 to 6.1 were identified for sensitive invertebrates from Atlantic Canada
(Appendix 8.3.3).
15.6.3.1.4	Fish
The effects of low pH and ANC and of high inorganic Al concentrations have been well
characterized in fish for many decades (Appendix 8.3.6). The 2008 ISA reported that
acidification impairs gill function and can cause respiratory and circulatory failure in fish.
Sensitivity to pH and inorganic Al varies among fish species, and among lifestages within
species, with early lifestages more sensitive to acidification. The most commonly studied
species were brown trout (Salmo trutta), brook trout (Salvelinus fontinalis), and Atlantic
salmon (Salmo salar). Studies published since the 2008 ISA, especially in Atlantic
salmon, add to the existing information on sublethal effects and confirm variation in
sensitivity among lifestages (Appendix 8.3.6.1). Since 2008, new studies include
acidification effects on migratory activities and behavior. New studies on fish show
behavioral effects at pH <6.6 (Appendix 8.3.6.5).
As summarized in Baker et al. (1990) and the 2008 ISA, fish populations in acidified
streams and lakes of Europe and North America have declined, and some have been
77
-------
eliminated as a result of atmospheric deposition of N and S and the resulting changes in
pH, ANC, and inorganic A1 concentrations in surface waters. There is often a positive
relationship between pH and the number of fish species, particularly between pH 5.0 and
6.5. Additional pH thresholds published since the 2008 ISA (Table 8-2) support this
range, and several new studies consider the role of DOC in controlling pH and
subsequent effects on biota. In the 2008 ISA and in new research, few or no fish species
are found in lakes and streams that have very low ANC (near zero; Figure 8-4 and
Table 8-3) and low pH (near 5.0). The number of fish species generally increases at
higher ANC and pH values. A1 is very toxic to fish, and thresholds to elevated
concentrations of this metal in acidified waters are summarized in Table 8-4.
IS.6.3.2 National-Scale Sensitivity, Biological Recovery, and Critical Loads
The extent and distribution of acid-sensitive freshwater ecosystems and sensitive regions
in the U.S. were well known at the time of the 2008 ISA. Measured data on lake and
stream ANC across the U.S. exhibit clear spatial patterns (Figure 8-11). Surface waters in
the U.S. that are most sensitive to acidification are largely found in the Northeast,
southern Appalachian Mountains, Florida, the upper Midwest, and the mountainous West
(Figure IS-12). Levels of acidifying deposition in the West are low in most areas and rare
in acidic surface waters, and the extent of chronic surface water acidification to date has
been very limited. However, episodic acidification occurs in both the East and West at
sensitive locations, and this is partly natural and partly caused by humans. Geographic
patterns in acidification sensitivity vary in response to spatial differences in geology,
hydrologic flow paths, presence and depth of glacial till, climate, and other factors
(Appendix 8.5.1). In the eastern U.S., acid-sensitive ecosystems are generally located in
upland, mountainous terrain underlain by weathering-resistant bedrock. Some of the most
in-depth studies of the effects of acid stress on fish were conducted in streams in
Shenandoah National Park in Virginia and in lakes in the Adirondack Mountains of New
York. Effects on fish have also been documented in acid-sensitive streams of the Catskill
Mountains of southeastern New York, and the Appalachian Mountains from
Pennsylvania to Tennessee and South Carolina.
78
-------
Surface Water Critical Loads for Acidity
S+N meq/m2/yr - 10th Percentile	v	\
12-50	\
51 -100	\
m 101 - 200	V
201-4,215 Conditions:
M n .	(1) ANC threshold: East = 50 peq/L, West = 20 peq/L
(2) Negative Critical Loads = 0.1 meq/m2/yr
	States	(3) NCLD v2.5 - 3/17/2015
ANC = acid-neutralizing capacity; meq = milliequivalent; yr = year.
Source: http://nadp.slh.wisc.edu/committees/clad.
Figure IS-12 Surface water critical loads for acidity in the U.S. 10th percentile
aggregation for 36-km2 grids with sulfur (S) and nitrogen (N).
Biological recovery in acid-affected areas is discussed in Section IS. 11. Typically,
biological recovery occurs only if chemical recovery (Appendix 7.1.5.1) is sufficient to
allow growth, survival, and reproduction of acid-sensitive plants and animals. Surface
water chemistry recovery varies by region, with the strongest evidence for improvement
in the Northeast and little or no recovery in central Appalachian streams. Acidification
and recovery of fresh waters will also be affected by the physical, chemical, and
biological modifications to acid inputs projected to occur with changes in annual mean
temperature and magnitude of precipitation (Appendix 8.5.3). As reported in the 2008
ISA and in new studies, biological recovery is slower than chemical recover}' in many
79
-------
systems (see Appendix 8.4). The time required for biological recovery is unknown and
only partial biological recovery may be possible.
Since the 2008 ISA, considerable CL research has focused on aquatic acidification in the
U.S. The CLs for deposition for aquatic acidification are expressed in eq/ha/yr of S, N, or
S + N because one or both pollutants can contribute to the observed effects. New
empirical CLs include 571 eq N/ha/yr in the Northeast and 286 eq N/ha/yr in the West to
prevent episodic acidification in high-elevation lakes (Table 8-7). Steady-state CLs have
been derived at many locations since the 2008 ISA (Table 8-8). Steady-state CLs of
acidifying deposition for lakes in the Adirondack Mountains (1,620 eq/ha/yr) and for the
central Appalachian streams (3,700 eq/ha/yr) were calculated to maintain a surface water
ANC of 50 (j,eq/L on an annual basis (NAPAP. 2011). CL values of less than
500 eq/ha/yr were calculated for one-third of streams in the Blue Ridge ecoregion, to
maintain stream ANC at 50 |_icq/L. For lakes in Class I and II wilderness areas in the
Sierra Nevada, CLs for acidifying deposition in 2008 were estimated at ANC values of 0,
5, 10, and 20 (j,eq/L, which span the range of minimum ANC values observed in Sierra
Nevada lakes. The median CL for granitic watersheds based on a critical ANC limit of
10 |_ieq/L was 149 eq/ha/yr. Slightly more than one-third of these lakes had estimated
rates of acidifying deposition higher than their CL.
In addition to the steady-state and empirical loads described above, CL estimates have
been derived from dynamic modeling (Appendix 8.5.4). For example, there is new work
on simulated past and future effects of N and S on stream chemistry in the Appalachians
and Adirondack Mountain lakes. In 12 watersheds in the Great Smoky Mountain
National Park, target levels of ANC to protect aquatic life were used and ranged from
minimal (0 j^ieq/L) to considerable protection (50 j^ieq/L). For the 12 study streams, target
levels ofNCV + SO42 deposition ranged from 270 to 3,370 eq/ha/yr to reach an ANC of
0 (j,eq/L by 2050 and 0 to 1,400 eq/ha/yr to reach an ANC of 50 |icq/L by 2050. However,
the majority of streams could not achieve the ANC target of 50 (ieq/L. Modeling also
suggests that complete recovery from acidification may not be possible by the year 2100
at all sites in the southern Blue Ridge region (Sullivan et al.. 201 lb) even if S emissions
cease entirely. In Shenandoah National Park, MAGIC modeling based on simulations of
14 streams identified a target load of about 188 eq S/ha/yr to achieve an ANC = 50 j^ieq/L
(preindustrial level based on hindcast simulations) in 2100 in sensitive streams. In a
dynamic modeling simulation in the Adirondack Mountains, about 30% of the lakes in
the region had a target load <500 eq/ha/yr to protect lake ANC to 50 j^ieq/L (Sullivan et
al.. 2012V Future decreases in SO42 deposition are suggested to be more effective in that
region in increasing Adirondack lake water ANC than equivalent decreases in NO3
deposition. In another modeling study of 20 Adirondack watersheds, estimates of
80
-------
preindustrial ANC for the study lakes ranged from 18 to 190 (ieq/L, and simulations
estimate that lake ANC has decreased by 26 to 100 j^ieq/L as a legacy of acidification.
IS.7 Estuarine and Near-Coastal Ecosystem Nitrogen Enrichment
For estuaries (areas where fresh water from rivers meets the salt water of oceans) and
near-coastal systems, causality determinations from the 2008 ISA are further supported
and strengthened by additional studies (Table IS-1). Estuaries support a large biodiversity
of flora and fauna and play a role in nutrient cycling. N from the atmosphere and other
sources contributes to increased primary productivity, leading to eutrophication
(Table 10-1), and N pollution is the major cause of harm to most estuaries in the U.S.
(Appendix 10). Source apportionment data in the 2008 ISA and newer studies indicate
that atmospheric contributions to estuarine N are heterogeneous across the U.S., ranging
from <10% to approximately 70% of total estuary N inputs (Table 7-9). In estuaries,
increasing nutrient over-enrichment leading to eutrophication is indicated by water
quality deterioration, resulting in numerous harmful effects, including areas of low
dissolved oxygen (DO) concentration (hypoxic zones), species mortality, and HABs.
New studies support the 2008 ISA's causal findings that increased N loading to coastal
areas can alter biogeochemical processes and lead to shifts in community composition,
reduced biodiversity, and mortality of biota. The current causal statement of biological
effects of N enrichment in estuarine ecosystems has been expanded to include total
primary production, altered growth, and total algal community biomass (Table IS-1).
IS.7.1 Estuary and Near-Coastal Biogeochemistry
In the 2008 ISA, the evidence was sufficient to infer a causal relationship between
reactive N deposition and biogeochemical cycling of N and C in estuarine and
near-coastal marine systems. Evidence reviewed in the 2008 ISA, along with new studies,
indicates elevated N inputs to coastal areas can alter key processes that influence N and C
cycling in near-coastal environments. As external organic matter loading to coastal areas
has increased in recent decades in many parts of the U.S., the varying rates of different N
cycling processes within estuaries themselves can also affect the magnitude of
eutrophication experienced as a result of external N enrichment. Nitrogen additions not
only cause the total pool of N to be larger but may also perturb N cycling in such a way
that the system may exacerbate eutrophication to a greater extent than expected based on
N additions alone. Research conducted since the 2008 ISA has shown that many of these
N cycling processes are more important in the estuarine environment than previously
understood. The removal of N through denitrification is a valuable ecosystem service in
81
-------
terms of constraining the extent and magnitude of eutrophication. Additional research has
established dissimilatory NOs reduction to NH4 (DNRA) as a more important N
reduction pathway in some estuaries. Ammonium produced via DNRA can lead to
enhanced productivity and respiration, which may exacerbate hypoxia. Recent studies
indicate that DNRA rates are higher in warmer months and can also take up a larger
percentage of total N reduction activity when temperatures are higher. The roles of
sedimentary microbial processes of denitrification and N2 production via anaerobic
ammonium oxidation (anammox) have been further characterized. New research has
shown that the community of N fixing microorganisms is more diverse in estuarine and
coastal waters than previously thought, and that N fixation occurs more widely than
previously assumed. Influence of benthic macrofauna on N cycling has received
increased research attention in part due to the potential for these organisms to mitigate
external N enrichment. Along with atmospheric anthropogenic CO2 inputs and other
factors, eutrophication from N loading may affect carbonate chemistry in coastal areas,
contributing to acidifying conditions in some circumstances such as where there is spatial
or temporal decoupling of production and respiration processes. Monitoring of coastal
areas shows that excess nutrient inputs continues to be a widespread problem in many
parts of the U.S. New research further supports conclusions of the 2008 ISA, and the
body of evidence is sufficient to infer a causal relationship between N deposition and
the alteration of biogeochemistry in estuarine and near-coastal marine systems.
IS.7.1.1 Nitrogen Enrichment
Estuarine biogeochemistry is complicated because it directly controls more than just the
N cycle; the response to N loading resulting in eutrophication affects the chemical
cycling of metals and DO (Appendix 7.2.3), redox conditions, pH (Appendix 7.2.4), and
ultimately energy transfer (e.g., food webs from microbes to humans). The response to N
loading is also tightly controlled by the availability of organic matter (i.e., C) and its
lability and reactivity. External organic matter loading to estuarine and coastal waters
appears to be increasing and these excess nutrient inputs are occurring within the context
of other stressors such as climate change (Appendix 7.2.6.12) and rising atmospheric
CO2, which further modify coastal biogeochemistry (Doncv. 2010). As reported in the
2008 ISA, estuaries are generally N limited, and have received sufficiently high levels of
N input from human activities (including deposition, agricultural runoff, and wastewater)
to cause eutrophication. Highly variable environments within estuaries are characterized
by a gradient of increasing salinity toward the ocean. As N moves downstream, some
fraction is taken up by phytoplankton or removed by microbial denitrification. Key
processes that influence N cycling include hypoxia, nitrification, denitrification, and
82
-------
decomposition. Until recently, it was generally believed that NH3 oxidation was
accomplished only by Proteobacteria in marine environments. New research has
discovered that some archaea can also oxidize NH3. These ammonia-oxidizing archaea
are dominant in some estuaries, while ammonia-oxidizing bacteria are more important in
others.
In the complex environment of the freshwater-to-ocean continuum, there are many
chemical and biological indicators of eutrophic condition. One approach is to measure
total nutrient loading and concentrations; however, these data need to be interpreted in
the context of the physical and hydrological characteristics that determine ecosystem
response. Water quality measures such as pH and DO, along with key biological
indicators such as chlorophyll a, phytoplankton abundance, HABs, macroalgal
abundance, and submerged aquatic vegetation (SAV; rooted vascular plants that do not
emerge above the water), can all be used to assess responses to nutrient loading
(Table 10-1). Nitrogen removal from the estuary is also influenced by faunal as well as
microbial communities.
Organic particles in coastal regions sink to the sediment-water interface where they
accumulate and decompose. Decomposition of these organic particles transforms
nutrients and depletes O2 in the water. Decreasing DO can create hypoxic (<2 mg/L of
dissolved O2) or anoxic zones inimical to fish and other aerobic life forms. Oxygen
depletion largely occurs only in bottom waters under stratified conditions, not throughout
the entire water column. This can result in seasonal hypoxia in shallow coastal regions,
particularly those that are receiving high inputs of nutrients from coastal rivers.
Development of hypoxia is increasingly a concern in estuaries across the U.S.
(Appendix 10.2.4).
Since the 2008 ISA, a number of papers have identified links between nutrient
enrichment and effects on estuarine carbonate chemistry, resulting in coastal acidification
or basification (Appendix 7.2.4). Eutrophication and acidification/basification are
complex biogeochemical processes that are driven by the same hydrological
(stratification) and biological (production/respiration) processes that can result in hypoxia
and enhanced organic matter loading. Acidification can occur by direct atmospheric
anthropogenic CO2 dissolution into the ocean. But under certain conditions N enrichment
can contribute to acidifying/basifying conditions, such as in systems with strong thermal
stratification or with spatial or temporal decoupling of production and respiration
processes. With increasing N inputs to coastal waters, CO2 in the water column is
produced from degradation of excess organic matter from changing land use, as well as
respiration of living algae and seagrasses, which in turn can make the water more acidic.
83
-------
Estuarine carbonate chemistry is complex, responding to a wide variety of natural,
anthropogenic, physical (mixing), chemical and biological drivers.
IS.7.1.2 Models
Since the 2008 ISA, several new applications of existing models have quantified
eutrophication processes in estuaries and near-coastal marine ecosystems. These have
included studies that focused primarily on N cycling or hypoxia. Other models of
estuarine eutrophication focus on N load apportionment, or on relationships between N
loads and ecological endpoints. Since the 2008 ISA, SPARROW has been used to
estimate total N loads within watersheds to determine sources of N to streams and rivers;
it has also been applied at regional and national scales. Additional models and tools that
include the contribution of N directly from the atmosphere have been applied to U.S.
estuaries, including the Watershed N Loading Model (NLM) and the Watershed
Deposition Tool (WDT). The latter was developed by the U.S. EPA to map atmospheric
deposition estimates to watersheds using wet and dry deposition data from CMAQ
(Schwede et al.. 2009). This tool links air and water quality modeling data for use in total
maximum daily load (TMDL) determinations and analysis of nonpoint-source impacts.
New model applications include studies that focused primarily on endpoints of N cycling,
hypoxia, and HABs. Models of coastal eutrophication are described in greater detail in
Appendix 7.2.8.
IS.7.1.3 National-Scale Sensitivity
Sensitivity of estuaries to biogeochemical changes associated with N enrichment varies
across the U.S. The biogeochemical sensitivity of estuaries and near coastal areas will be
discussed together with national-scale biological sensitivity to N enrichment in
Section IS.7.3.
IS.7.2 Biological Effects of Nitrogen Enrichment
In the 2008 ISA, evidence was sufficient to infer a causal relationship between N
deposition and the alteration of species richness, community composition, and
biodiversity in estuarine ecosystems. The strongest evidence for a causal relationship was
from changes in biological indicators of nutrient enrichment (chlorophyll a, macroalgal
[seaweed] abundance, HABs, DO, and changes in SAV; Table 10-1). Some indicators,
such as chlorophyll a, are directly linked to nutrient enrichment and provide evidence of
84
-------
early ecosystem response; other indicators, such as low DO and decreases in SAV,
indicate more advanced eutrophication. Phytoplankton are the base of the coastal food
web and increases in primary producer biomass and altered community composition
associated with increased N can lead to a cascade of direct and indirect effects at higher
trophic levels. At the time of the 2008 ISA, N was recognized as the major cause of harm
to the most estuaries in the U.S. Since 2008, new paleontological studies, observational
studies, and experiments have further characterized the effects of N on phytoplankton
growth and community dynamics, macroinvertebrate response, and other indices of
biodiversity. For this ISA, new information is consistent with the 2008 ISA and the
causal determination has been updated to reflect more specific categories of effects to
include total primary production, altered growth, and total algal community biomass.
This new research strengthens the evidence base and is consistent with the 2008 ISA
(Table IS-1) that the body of evidence is sufficient to infer a causal relationship
between N deposition and changes in biota including total primary production,
altered growth, total algal community biomass, species richness, community
composition, and biodiversity due to N enrichment in estuarine environments.
Since the 2008 ISA, additional evidence has shown that reduced forms of atmospheric N
play an increasingly important role in estuarine and coastal eutrophication and HAB
dynamics. New studies emphasize that N inputs interact with physical and hydrologic
factors to increase primary productivity and eutrophication in coastal areas.
Climate-related changes in temperature, precipitation, and wind patterns, as well as
extreme weather events, stronger estuary stratification, increased metabolism and organic
production, and rising sea-levels are all expected to modify coastal habitats
(Appendix 10.1.4.1).
IS.7.2.1 Primary Producers
Algae are the base of the coastal food web, and the 2008 ISA showed that changes in
chemical composition of N inputs can shift the algal community and cascade up the food
web. Chlorophyll a is a broadly recognized indicator of phytoplankton biomass and is
used as a proxy for assessing effects of estuarine nutrient enrichment. It can signal an
early stage of water quality degradation related to nutrient loading and is incorporated
into water quality monitoring programs and national-scale assessments including U.S.
EPA's National Coastal Condition Assessment (Appendix 7.2.7). Phytoplankton
sampling, microcosms studies, and sediment core analysis have shown changes in
phytoplankton community structure in estuaries with elevated N inputs (Appendix 10.3).
These shifts at the base of the food web to species that are not as readily grazed
(e.g., cyanobacteria, dinoflagellates) have a cascade of effects including poor trophic
85
-------
transfer and an increase in unconsumed algal biomass, which could stimulate
decomposition and O2 consumption, and thus increase the potential for hypoxia.
There is consistent and coherent evidence that the incidence of HAB outbreaks is
increasing in both freshwater and coastal areas, a problem that has been recognized for
several decades (Appendix 10.2.2). Of the 81 estuary systems for which data were
available for the National Estuarine Eutrophication Assessment (NEEA), 26 exhibited a
moderate or high symptom expression for nuisance or toxic algae (Brickcr et al.. 2007).
Since the 2008 ISA, HAB conditions and effects of HAB toxins on wildlife have been
further characterized (Appendix 10.2.2). Toxins released during HABs can be harmful to
fish and shellfish and may be transferred to higher trophic levels. The form of N affects
phytoplankton growth and toxin production of some HAB species. Increasing loads of
NH3+/NH4+ have been linked to the expansion of HABs and altered phytoplankton
community dynamics (Appendix 10.3.3). Cyanobacteria, and many chlorophytes and
dinoflagellates, may be better adapted to NH44", while diatoms generally thrive in the
presence of oxidized forms of N such as NO3 (Figure 10-7).
Macroalgal (seaweed) growth is also stimulated by increased N inputs, which increase
the dominance of faster growing benthic or pelagic macroalgae to the exclusion of other
species (Appendix 10.2.3). Studies published since the 2008 ISA provide further
evidence that macroalgae respond to the form of N, with some species showing greater
assimilation and growth rates with NH44" than with NO3 . Increased abundance of
macroalgae, which block light, and increased epiphyte loads on the surface of SAV may
reduce the growth and biomass of SAV. SAV, including the eelgrass Zostera marina, are
important ecological communities found within some coastal bays and estuaries that are
sensitive to elevated nutrient loading, and the loss of this habitat can lead to a cascade of
ecological effects because many organisms are dependent upon seagrasses for cover,
breeding, and as nursery grounds. Recently, the presence of seagrass beds was linked to
decreased bacterial pathogens of humans, fishes, and invertebrates in the water column
and lower incidence of disease in adjacent coral reefs (Appendix 10.2.5). The 2008 ISA
reported correlations between increased N loading and declines in SAV abundance, and
newer studies have further characterized this relationship. In a survey of southern New
England estuaries, reduced eelgrass extent was observed at increased watershed N
loading. New studies have characterized the role of invertebrate mesograzers, such as
small crustaceans and gastropods, in controlling algal growth, potentially buffering
eutrophication effects on seagrass communities (Appendix 10.3.7). Macroalgae may not
be a good indicator of eutrophication in some upwelling-influenced estuaries in the
Pacific Northwest because an increase in macroalgal biomass in these systems does not
appear to be associated with temporal declines in eelgrass (Appendix 10.2.3).
86
-------
IS.7.2.2
Bacteria and Archaea
Ammonia-oxidizing prokaryotes carry out nitrification in estuarine waters.
Ammonia-oxidizing archaea have been described relatively recently, and several studies
since the 2008 ISA have considered community responses of ammonia-oxidizing bacteria
and archaea. Community structure of ammonia-oxidizers is related to nutrient inputs and
affected by the form of available N (Appendix 10.3.4).
15.7.2.3	Invertebrates
In coastal areas with severe seasonal hypoxia, the community of benthic organisms shifts
toward shorter life spans and smaller body size (Appendix 10.2.4). Reduced species
density and diversity in the northern Gulf of Mexico are linked to persistent hypoxic
events. The form of N present has been shown to affect molluscan taxonomic
assemblages (Appendix 10.3.5). Shifts in algal composition and productivity can affect
growth of shellfish that feed on phytoplankton. Shellfish contribute to N and C cycling
and can improve water quality, and recent research has explored the use of these
organisms for coastal N remediation (Appendix 7.2.6.11). Harvest of shellfish for human
consumption removes nutrients from estuaries.
N enrichment is one of several factors linked to increased disease susceptibility,
bleaching, and reduced calcification rate in corals (Appendix 10.4.2). Several studies
have isolated effects of N, which affects corals via pathways that are distinct from P. The
threatened status of staghorn coral (Acropora cervicornis) and elkhorn coral (Acropora
palmata) under the U.S. Endangered Species Act has been linked to indirect N pollution
effects, specifically low DO, algal blooms that alter habitat, and other non-nutrient
stressors (Hernandez et al.. 2016). Increasing acidification of coastal waters, which may
be exacerbated by elevated N inputs under certain circumstances (Appendix 7.2.4), is
projected to alter marine habitat, have a wide range of effects at the population and
community level and affect food web processes. Although the interactions between
elevated CO2, decreasing pH, and nutrient inputs are complex, calcareous plankton,
oysters, clams, sea urchins, and coral that produce calcium carbonate shells may be
affected by long-term decreases in pH (Appendix 10.5).
15.7.2.4	Fish
Fish biodiversity is altered by increased N inputs and resulting changes in biological and
chemical indicators (Appendix 10.3.6). Many fish are unable to persist at DO levels
87
-------
below 2 mg/L (Figure 10-4). Recent studies in the southern Gulf of Saint Lawrence have
linked SAV loss to declines in fish biodiversity, although organisms did not change
positions within food webs. In laboratory conditions, turbidity associated with
eutrophication alters fish reproductive behaviors. Hypoxia has also recently been shown
to affect reproduction in fish. For example, hypoxia acts as an endocrine disruptor in
Atlantic croaker (Micropogonias undulatus; Appendix 10.2.4).
IS.7.3 National-Scale Sensitivity and Critical Loads
The NEEA, the most recent comprehensive survey of eutrophic conditions in U.S.
estuaries conducted by the National Oceanic and Atmospheric Administration, defined
eutrophication susceptibility as the natural tendency of an estuary to retain or flush
nutrients (Brickcr et al.. 2007). In estuaries that have longer water residence times,
nutrients are more likely to lead to eutrophic conditions (Appendix 10.1.4). As reported
in the 2008 ISA and newer studies, nutrient loading accelerates hypoxia, which is more
likely in marine waters with limited water exchange, water column stratification, and
high production and settling of C to bottom waters. Other factors identified in the
2008 ISA that increase estuary sensitivity to eutrophication include human population,
agricultural production, and the size of the estuary relative to its drainage basin. The
NEEA reported that the most eutrophic estuaries in the U.S. occur in the mid-Atlantic
region, and the estuaries with the lowest degree of eutrophication are in the North
Atlantic (Figure 10-2). Estuaries identified in the 2008 ISA as susceptible to
eutrophication include the Chesapeake Bay, Pamlico Estuary in North Carolina, Long
Island Sound, as well as along the continental shelf adjacent to the Mississippi and the
Atchafalaya River discharges to the Gulf of Mexico. New research at the regional scale
includes long-term studies of several coastal systems that are looking at trends in coastal
water quality and chemistry. A 23-year study of the Chesapeake Bay concluded that
water quality has decreased and chlorophyll a levels have increased since 1986, in part
due to long-term climate trends (see Appendix 10.2.5).
Since the 2008 ISA, there is additional information on the extent and severity of
eutrophication and hypoxia in sensitive regions. Areas of eutrophication-related hypoxia
are found on the U.S. eastern and western coasts and the Gulf of Mexico (Figure 10-5).
The 2008 ISA reported that the largest zone of hypoxic coastal water in the U.S. was the
northern Gulf of Mexico on the Louisiana-Texas continental shelf. In the summer of
2017, the hypoxic zone in the Gulf was the largest ever measured at 14,123 km2
[8,776 mi2; U.S. EPA (2017bVI. Atmospheric deposition to watersheds in the
Mississippi/Atchafalaya River Basin contributes approximately 16 to 26% of the total N
load to the Gulf of Mexico (Appendix 10.2.4). Long Island Sound also experiences
88
-------
periods of anoxia. In other U.S. coastal systems, hypoxia incidence is increasing, but DO
impacts are relatively limited temporally and spatially. In the Pacific Northwest, coastal
upwelling not related to anthropogenic sources can be a large source of nutrient loads,
and the advection of this upwelled water can introduce hypoxic water into estuaries.
The NEEA suggested that only a small fraction of the estuary systems evaluated showed
moderate to high SAV loss (Brickcr et al.. 2007). mostly in the mid-Atlantic region.
While seagrass coverage is improving in some estuaries, such as Tampa Bay (Tampa Bay
Case Study, Appendix 16), many estuaries continue to see declines in seagrass extent.
SAV is often at a competitive disadvantage under N enriched conditions because of the
fast growth of opportunistic macroalgae that preferentially take up NH4+ and can block
light from seagrass beds.
There are thresholds of response identified for some biological and chemical indicators of
N enrichment in estuaries (Appendix 10). The amount of chlorophyll a is an indicator of
phytoplankton biomass, and thus, a proxy for assessing estuarine nutrient enrichment. In
general, 0-5 (ig/L chlorophyll a is considered a good condition, concentrations between 5
and 20 (ig/L are classified as fair condition, and concentrations of >20 j^ig/L indicate poor
conditions (Table 10-2). A new response threshold of tidal-averaged total N
concentration of <0.34 mg/L has been identified for healthy eelgrass in Massachusetts
waters. Markedly decreased eelgrass coverage is observed at N loading rates
>100 kg N/ha/yr, and levels above 50 kg N/ha/yr are likely to impact SAV habitat extent
in shallow New England estuaries (Table 10-4). Greaver et al. (2011) identified the range
of 50-100 kg N/ha/yr total N loading as the empirical CL for loss of eelgrass based on
Latimer and Rego (2010). In terms of DO, concentrations of 0 mg/L are anoxic, 0-2 are
indicative of hypoxic conditions, and 2-5 mg/L are biologically stressful conditions
(Figure 10-4). Oxygen depletion largely occurs only in bottom waters under stratified
conditions, not throughout the entire water column.
The indicators of nutrient enrichment in coastal areas (chlorophyll a, HABs, macroalgal
abundance, DO, SAV, and benthic diversity) have been incorporated into indices of
coastal eutrophication. In the 2008 ISA, the Assessment of Estuarine Tropic Status
(ASSETS) categorical Eutrophication Condition index (ECI) developed for the NEEA
was used as an assessment framework for coastal U.S. estuaries (Bricker et al.. 2007).
Additional indices of estuarine functioning that incorporate biological indicators have
since been developed both in the U.S. and internationally (Appendix 10.2.6).
Comparisons of these frameworks have identified robust methods to measure estuarine
response, such as incorporation of annual data, frequency of occurrence, spatial coverage,
secondary biological indicators, and a multicategory rating scale.
89
-------
Since the 2008 ISA, N enrichment has been linked to coral bleaching and reduced
calcification rates (Appendix 10.4.2). Near-coastal coral reefs in the U.S. occur off south
Florida, Texas, Hawaii, and U.S. territories in the Caribbean and Pacific.
IS.8 Wetland Ecosystem Nitrogen Enrichment and Acidification
New evidence, including new CLs, supports and strengthens the causal findings from the
2008 ISA regarding N enrichment effects in wetlands (Table IS-1). In freshwater
wetlands and coastal wetland ecosystems, deposition of N and S does not tend to cause
acidification-related effects at levels currently common in the U.S. However, the 2008
ISA documented that wetlands can be sensitive to N enrichment and eutrophication
effects. Newer studies have characterized N effects on biogeochemistry, physiology,
biodiversity, national sensitivity, and CLs for freshwater and coastal wetlands; coastal
wetlands are typically tolerant of higher N loading than freshwater wetlands.
IS.8.1 Wetland Biogeochemistry
In the 2008 ISA, evidence was sufficient to infer a causal relationship between N
deposition and the alteration of wetland biogeochemical cycling. Although sources and
rates of N inputs vary widely among wetlands, N deposition contributes substantially to
total loading in many wetlands. This additional N alters C cycling, N cycling, and the
release of nutrients to hydrologically connected surface waters. New research together
with the information included in the 2008 ISA shows that the body of evidence is
sufficient to infer a causal relationship between N deposition and the alteration of
biogeochemical cycling in wetlands.
The 2008 ISA reported that N enrichment altered N cycling in wetland ecosystems.
Chemical indicators of N deposition in wetlands include NO? and NH4+ leaching, DON
leaching, N mineralization, denitrification rates, and N2O emissions. A wetland can act as
a source, sink, or transformer of atmospherically deposited N, and these functions vary
with season and hydrological conditions. Vegetation type, physiography, local hydrology,
and climate all influence source/sink N dynamics in wetlands. A new synthesis of global
wetland data showed that a wetland's reactive N removal and water quality improvement
is proportional to its reactive N load, and removal efficiency is 26% higher in nontidal
than tidal wetlands. Further, a new meta-analysis showed that N enrichment increases
wetland N2O emissions by 207%. New studies have also evaluated the effects of N
loading/N addition on other endpoints related to N cycling in peat bog, riparian,
mangrove, and salt marsh wetlands (see Appendix 11.3.1). The endpoints evaluated
90
-------
include ecosystem N retention, wetland export of N to surface waters, N fixation, N
mineralization, denitrification, emission of N2O, and bacterial abundance, activity, and
composition in wetland soils. The results of North American studies are summarized in
Figure 11-2. Across studies, N enrichment decreases the ability of wetlands to retain and
store N, which may diminish the wetland ecosystem service of improving water quality.
In the 2008 ISA, evidence from Canadian and European peatlands showed that N
deposition had negative effects on Sphagnum (moss) bulk density and mixed effects on
Sphagnum productivity depending on the history of deposition. There is new information
on how N deposition alters biogeochemical cycling of C in wetlands. Chemical indicators
of N deposition in wetlands include soil organic matter, total soil C or peat C, CO2
emissions, and CH4 emissions. Long-term C storage is an important ecosystem service of
wetlands for which measures of physical marsh stability can serve as a proxy, and
physical indicators of N deposition can include temperature, bulk density, physical
resistance, and soil water content. In addition, changes to plant growth rates and
productivity indicate altered C cycling in wetlands, and are summarized in Section IS.8.2.
The literature evaluates the effects of N deposition, N loading, or experimental N
addition on C cycling in bogs, fens, riparian or intermittent marshes, freshwater tidal
marshes, mangroves, and salt marshes (see Appendix 11.3.2). Significant effects ofN
loading upon biogeochemical cycling of C in North American wetlands (in which the N
addition was 500 kg N/ha/yr or lower) are summarized in Figure 11-3. N enrichment
decreases wetland retention of C, as indicated by new studies and a new meta-analysis
that show that N enrichment increases methane production in salt marshes. New studies
of marshes along the Gulf Coast and East Coast find that N enrichment also decreases the
bulk density of salt marshes, making marshes less resilient to physical stresses from tidal
or storm flooding, and may accelerate coastal marsh loss.
IS.8.2 Biological Effects of Wetland Nitrogen Enrichment/Eutrophication
In the 2008 ISA, evidence was sufficient to infer a causal relationship between N
deposition and the alteration of species richness, species composition, and biodiversity in
wetlands. New evidence is presented in the following sections regarding the effects of N
upon wetland plant physiology, architecture, demography, and biodiversity. The body of
evidence is sufficient to infer a causal relationship between N deposition and the
alteration of growth and productivity, species physiology, species richness,
community composition, and biodiversity in wetlands.
91
-------
IS.8.2.1
Growth, Productivity, and Physiology
In the 2008 ISA, evidence from Canadian and European bogs and fens showed that N
deposition had negative or mixed effects on Sphagnum (moss) productivity, depending on
history of deposition. In Canadian ombrotrophic peatlands experiencing deposition of
2.7-8.1 kg N/ha/yr, peat accumulation increased with N deposition, but accumulation
rates had slowed by 2004, indicating a degree of N saturation. Coastal wetlands
responded to N enrichment with increased primary production, which shifted microbial
and plant communities and altered pore water chemistry, although many of the studies in
coastal wetlands used N enrichment levels more like those of wastewater than
atmospheric deposition. New research on N enrichment effects on growth and
productivity was conducted in ombrotrophic bogs, intermittent wetlands, freshwater tidal
marsh, mangroves, and coastal salt marshes (see Appendix 11.4). Ecological endpoints
evaluated to assess N loading effects on growth and productivity include plant
aboveground biomass and productivity, plant belowground biomass of roots and
rhizomes, and growth rates, and are summarized along with N effects on C cycling in
Figure 11-3. The effects of N additions on plant physiology were not addressed in the
2008 ISA, but information regarding these effects is available for bogs and fens, riparian
wetlands, freshwater tidal marshes, mangroves, and salt marshes (see Appendix 11.5).
Ecological endpoints evaluated to assess N loading effects on plant physiology include
stoichiometry (i.e., nutrient concentrations and ratios of multiple nutrients in plant tissue),
nutrient acquisition efficiency (including insectivory rates in carnivorous plants), nutrient
use efficiency, and nutrient reabsorption efficiency. These endpoints are summarized in
Figure 11-4.
In general, across types of wetlands, nitrogen loading increases aboveground growth and
productivity while decreasing or not affecting belowground growth and productivity. In
bogs and fens, N deposition decreases growth of state-listed Sarracenia purpurea (purple
pitcher plant), and N enrichment increases aboveground productivity of emergent sedges
more than of peat-building moss species. These changes cascade up to affect biodiversity
in bogs and fens (see below, Section IS.8.2.2). In freshwater and tidal marshes, N
enrichment increases aboveground productivity while decreasing belowground
productivity, and this shift from belowground to aboveground plant productivity may
account for changes in wetland C storage (see Section IS.8.IV
Changes to plant physiology and stoichiometry vary by species tolerance to N and N
acquisition strategies. In bogs, N enrichment typically causes increased plant tissue N
concentrations, decreased N use efficiency, and decreased N resorption efficiency during
senescence. After several years of exposure to high rates of N loading, bog plants may
experience leafN saturation and limitation by other nutrients (e.g., P, K, and Ca,
92
-------
indicated by increasing reabsorption efficiencies), resulting in leaf damage in sensitive
species. S. purpurea (purple pitcher plant) decreases its dependence upon insectivory for
nutrition at N deposition rates of 4.4 kg N/ha/yr. In freshwater marshes, N enrichment
also increases plant tissue N concentrations while increasing P limitation and altering
resorption efficiencies.
Plant architecture was not addressed in the 2008 ISA, and demography was addressed
only for bogs and fens. Aboveground, plant architecture includes branching patterns, as
well as the size, shape, and position of leaves and flower organs. New studies find N
enrichment affects plant architecture in a salt marsh, in mangroves, in freshwater tidal
marshes, and in a riparian wetland (Appendix 11.6). In terms of plant demography, the
2008 ISA found positive population growth rates for S. purpurea at 0 or 1.4 kg N/ha/yr,
but population losses at 14 kg N/ha/yr. N deposition above 6.8 kg N/ha/yr increases
population extinction risk of S. purpurea. New studies show that N addition has
species-specific effects on reproduction of West Coast salt marsh plant species and that it
increases mortality across the global distribution of mangrove species (Appendix 11.7).
IS.8.2.2 Biodiversity
In the 2008 ISA, evidence was sufficient to infer a causal relationship between N
deposition and the alteration of species richness, species composition, and biodiversity in
wetlands. Notably, the 2008 ISA cited 4,200 native plant species in U.S. wetlands, 121 of
which are federally endangered. Given their relative area, wetlands provide habitat to a
disproportionally high number of rare plants. Many wetland species have adapted to N
limited conditions, including endangered species in the genera Isoetes (3 endangered
species) and Sphagnum (15 endangered species), as well as insectivorous plants such as
pitcher plants (Sarracenia spp.) and sundews (Drosera rotundifolia).
Coastal wetlands responded to N enrichment with increased primary production,
changing microbial and plant communities, and altered pore water chemistry, although
many of the studies available in 2008 used high N enrichment levels more similar to N
loading from wastewater than from atmospheric deposition. New research since 2008
across environmentally relevant N levels including N deposition gradient studies,
experimental N addition studies, and observational studies show that N enrichment
altered biodiversity in bogs and fens, intermittent wetlands, freshwater wetlands,
freshwater tidal wetlands, and coastal salt marshes (see Appendix 11.8).
New research from wetland ecosystems strengthens the 2008 causal statement. New
research confirms that, as in terrestrial systems, N addition can decrease the abundance
and richness of sensitive species while increasing the abundance and richness of tolerant
93
-------
species. In bogs and fens, N enrichment decreases the survival of insectivorous plants and
the cover of mosses, while increasing the cover of shrub species. In freshwater marshes,
N enrichment changes plant community composition, increases the abundance of and
stresses caused by invasive plant species, promotes the harmful algal species that produce
the toxin microcystin, and increases mosquito larvae that are vectors for zoonotic
diseases (see Figure 11-1). In freshwater tidal and coastal marshes, N enrichment changes
plant community composition, increases cover of invasive plant species, increases
herbivory by invertebrates, and increases herbivory by the invasive m am mal My ocas lor
coypus (nutria).
IS.8.2.3 National Sensitivity and Critical Loads for Wetlands.
Freshwater and coastal wetlands tend to have different sensitivity to added N. Broadly,
wetlands that receive a larger fraction of their total water budget in the form of
precipitation are more sensitive to the effects of N deposition. For example, bogs
(70-100% of hydrological input from rainfall) are more sensitive to N deposition than
fens (55-83% as rainfall), which are more sensitive than coastal wetlands (10-20% as
rainfall).
Since the 2008 ISA, an N CL for U.S. coastal wetlands has been established. The CL is
based on several different ecological endpoints, including plant community composition,
microbial activity, and biogeochemistry (63-400 kg N/ha/yr) and that this CL includes
total N loading values not just N deposition. Figure 11-6 shows a comparison of the N
CL for coastal wetlands with recent studies of ecological impacts of N (at N levels of
100-250 kg N/ha/yr).
Since the 2008 ISA, two N CLs for U.S. freshwater wetlands have been established. The
CL for wetland C cycling, quantified as altered peat accumulation and NPP, is between
2.7 and 13 kg N/ha/yr. The upper end of this CL range is based on measurements of wet
deposition only (10 to 13 kg N/ha/yr), and therefore, does not reflect total N loading.
There is also a CL to protect biodiversity based on morphology and population dynamics
of the purple pitcher plant (Sarracenia purpurea) between 6.8-14 kg N/ha/yr. A more
recent study across an N deposition gradient suggests that purple pitcher plant
populations experience negative effects ofN deposition at rates lower than this CL, but
the more recent research has not yet been incorporated into the CL framework. A
comparison of freshwater wetland CLs to observed ecological impacts of N from recent
studies (4.4-500 kg N/ha/yr) is given in Figure 11-7.
94
-------
IS.9
Freshwater and Wetland Ecosystem Sulfur Enrichment
New evidence from wetland and freshwater aquatic ecosystems strengthens and extends
the causal findings of the 2008 ISA regarding nonacidifying sulfur effects and provides
the basis for a new causal determination (Table IS-1). New research together with the
information included in the 2008 ISA shows that the evidence is sufficient to infer a
causal relationship between S deposition and the alteration of Hg methylation in surface
water, sediment, and soils in wetland and freshwater ecosystems. New evidence is
sufficient to infer a new causal relationship between S deposition and changes in
biota due to sulfide phytotoxicity, including alteration of growth and productivity,
species physiology, species richness, community composition, and biodiversity in
wetland and freshwater ecosystems.
SOx deposition can have chemical and biological effects other than acidification,
particularly in flooded wetland soils and aquatic ecosystems. The 2008 ISA described
qualitative relationships between SO42 deposition and a number of ecological endpoints,
including altered S cycling, sulfide phytotoxicity, internal eutrophication of aquatic
systems, altered methane emissions, increased mercury (Hg) methylation, and increased
Hg loading in animals, particularly fish. Table 12-11 summarizes the chemical
concentrations that alter ecological endpoints and the quantitative relationships
describing the effects of SO42 deposition. Recent research supports these relationships
between S deposition and ecological endpoints and provides the basis for SOx deposition
levels, water column SO42 concentrations, and water column sulfide concentrations
protective of plants and animals.
IS.9.1 Biogeochemistry
SOx deposition alters biogeochemical processes via S enrichment. The processes include
S cycling (see Appendix 12.2.1), P cycling (see Appendix 12.2.4), C cycling (see
Appendix 12.2.5), and Hg cycling (see Appendix 12.3). The primary chemical indicator
for nonacidifying or enrichment effects of S in wetland and aquatic ecosystems is surface
water SO42 concentration, as it is for acidifying effects. The 2008 ISA reported that
chemical reduction of SO42 was an important indicator of SOx effects on water
chemistry because the process generates ANC. There are no new studies on ANC
generation through SO42 reduction, although microbial SO42 reduction remains an
active area of research. In aquatic ecosystems for which atmospheric and terrestrial S
inputs are similar in magnitude to rates of microbial SO42 reduction, the products of
microbial SO42 transformation may be more reliable indicators of S enrichment effects
95
-------
than surface water SO42 concentrations. These chemical indicators include
methylmercury (MeHg), sulfide, and phosphate.
MeHg is the most persistent and toxic form of Hg in the natural environment. It is
measured in surface water or aquatic sediments (MeHg concentration or the percentage of
MeHg in total Hg) to predict its effects on biota. Several new studies demonstrate
significant positive relationships between surface water SO42 concentrations and water
or sediment MeHg concentrations (see Appendix 12.3.5). Another product of SO42
reduction, sulfide (measured as surface water or sediment pore water S2 concentrations),
is also a water quality indicator of deposition effects on biota. In freshwater ecosystems
with iron-rich sediments, sulfide may react with iron bound to phosphates in the sediment
to release phosphate into the water column, increasing primary productivity recent
literature refers to this process as internal eutrophication (Appendix 12.2.4).
In terms of S enrichment effects on carbon cycling, the 2008 ISA documented the
suppression of methane emissions in wetland soils by SO42 addition in several studies
and noted that 15 kg S/ha/yr suppressed methane emissions. Recent research has
confirmed that S enrichment increases the abundance or metabolic activity of
SO42 -reducing prokaryotes (SRPs), which under some conditions compete with
methanogens by suppressing their activity, and in turn, suppressing methane emissions
(Appendix 12.2.4). However, there are no new studies documenting S deposition effects
on methane emissions in U.S. ecosystems.
IS.9.2 Biological Effects of Sulfur Enrichment
Nonacidifying S effects upon biota include plant toxicity, changes in plant growth and
biodiversity, and increased Hg concentrations in biota. The toxicological effects of Hg
accumulation in animals were documented in the 2008 ISA and newer studies.
IS.9.2.1 Sulfur Nutrient and Toxicity to Plants
Plants and other organisms require S as an essential nutrient. The deposition of S can
affect plant protein synthesis by affecting S availability for S containing amino acids,
which in turn will affect N uptake. The 2008 ISA documented the effects of S042
toxicity on plant development and reproduction at very high S loads. There is no new
evidence of S deposition effects upon plant S nutrition or SO42 toxicity. The product of
microbial SO42 reduction, sulfide, is an important plant toxin, and the 2008 ISA
documented sulfide phytotoxicity in European systems. Together with new research
showing sulfide phytotoxicity in North American wetlands, the body of evidence is
96
-------
sufficient to infer a causal relationship between S deposition and changes in biota
due to sulfide phytotoxicity including alteration of growth and productivity, species
physiology, species richness, community composition, and biodiversity in wetland
and freshwater ecosystems.
The 2008 ISA showed that sulfide toxicity decreased the biomass of wetland plants and
aquatic macrophytes in mesocosms under aquatic S concentrations higher than current
U.S. concentrations. In Europe, research showed that a threshold value of <48 mg
S042 /L in surface water would protect the sensitive aquatic species Stratiotes aloides and
Potamogeton acutifolius (not native to contiguous US), as well as to protect
zosteriformis and Utricularia vulgaris, which are both native and widely distributed in
contiguous US. New research has demonstrated sulfide phytotoxicity effects at current
ambient sulfide concentrations in multiple ecosystems within the U.S. (Appendix 12.2.3).
Sulfide decreased total plant cover and cover of dominant species in a New York fen and
decreased the growth rate of Cladium jamaicense (sawgrass), a keystone species in the
Florida Everglades. Zizaniapalustris (wild rice) is an economically and culturally
important species sensitive to sulfide, and the Minnesota Pollution Control Agency has
developed a model for this species that calculates protective levels of water SO42
concentrations, given (specific) iron and DOC concentrations in water bodies. A recent
review identifies sulfide thresholds between 0.3-29.5 mg S27L for altered growth,
productivity, physiology, or increased mortality of 16 freshwater wetland emergent plant
and aquatic submerged macrophyte species native to North America (see Table 12-2).
IS.9.2.2 Sulfur Effects on Mercury Methylation
In the 2008 ISA, evidence was sufficient to infer a causal relationship between S
deposition and increased methylation of Hg in aquatic environments where the value of
other factors is within an adequate range for methylation. In the 2008 ISA,
sulfur-reducing bacteria (SRB) were identified as the organisms responsible for Hg
methylation. New evidence shows the ability to methylate Hg is more broadly distributed
phylogenetically, including both bacteria and archaea, which is why this document refers
to S042 -reducing mercury methylators as sulfur-reducing prokaryotes (SRPs) rather than
SRB (Appendix 12.3.2). In the 2008 ISA, wetland and lake-bottom sediments were
identified as habitat for mercury methylating SRPs. Recent research documents microbial
mercury methylation in lakes, in wetland sediments and moss, within periphyton, in
marine ecosystems, and within disturbed terrestrial forest soils (Appendix 12.3.2 and
Appendix 12.3.3). Microbial mercury methylation responsive to SOx deposition occurs in
freshwater lakes, freshwater wetlands, freshwater reservoirs, and freshwater agricultural
areas (Appendix 12.3.4). Between the 2008 ISA and new research, the body of evidence
97
-------
is sufficient to infer a causal relationship between S deposition and the alteration of
Hg methylation in surface water, sediment, and soils in wetland and freshwater
ecosystems.
Hg methylation is determined in part by surface water S042 . because many strains of
SRPs possess the recently identified gene pair hgcAB, and pair their metabolism of C
with both dissimilatory S042 reduction and mercury methylation (see Appendix 12.3.2
and Figure 12-5). Microbial methylation rates are determined by other environmental
requirements of SRPs, including seasonality and temperature, pH, salinity, amount of
organic matter in the water and sediments, and concentrations of iron and nitrate
(Appendix 12.3.3). New research demonstrates that Hg methylation occurs at current
ambient SO42 concentrations within U.S. water bodies. Multiple lines of evidence
support a relationship between SO42 surface water concentrations and MeHg
concentration or production in various freshwater systems. Linear relationships between
S042 concentrations and MeHg concentrations were observed in sediments of the South
River, VA, across peat bogs in Minnesota and Ontario, and across prairie pothole lakes in
Saskatchewan (Figure 12-17). In addition to the studies of lake and wetland sediments
reviewed in the 2008 ISA, studies employing lab incubations show that SO42 increases
Hg methylation in samples from Adirondack peat bogs, from South River, VA sediments,
from periphyton growing in North American lakes and wetlands, and from leaf packs in
Minnesota river water (Appendix 12.3.3.1). Experimental addition of S to field
mesocosms or whole ecosystems has shown that S enrichment as wet S deposition
increases MeHg in water, sediment, or biota, in Little Rock Lake, WI; Bog Lake Fen,
MN; the Experimental Lakes Area, Ontario; and the bog experiment at Degero Stormyr,
Sweden (Appendix 12.3.4.1). In observational studies of S and Hg deposition, fish Hg
concentrations decline with temporal declines in SOx deposition in Isle Royale (a Class I
area). Fish Hg concentrations correlate positively with Hg and S deposition across Texas
ecoregions, and a 12-year study found that fish Hg in Voyageurs National Park (a Class I
area) declined in lakes with decreasing S deposition only when lake DOC remained
constant (Appendix 12.3.5.1). New research is consistent and coherent with the research
presented in the 2008 ISA in demonstrating that sulfur enrichment from SOx deposition
stimulates mercury methylation in North American ecosystems. Current research
suggests that mercury methylation generally peaks between 10 and 100 mg SO42 /L in
surface water, and quantitative relationships between S and Hg, such as target values or
thresholds, are reported in Table 12-12.
98
-------
IS.9.2.3
Sulfur, Mercury, and Animal Species
Mercury is a developmental, neurological, endocrine, and reproductive toxin across
animal species. The 2008 ISA documented Hg accumulation in fish, songbirds, four turtle
species, insectivorous passerine birds, and the common loon (Gavia immer). Recent
research also documented Hg accumulation in insectivore songbirds, bats, and fish in
agricultural wetlands. The 2008 ISA reported that 23 states had issued fish advisories by
2007 in response to the U.S. EPA's fish tissue criterion of 0.3 (.ig MeHg/g fish (0.3 ppm),
set to protect human health. The 2008 ISA reported on the negative impacts of Hg on the
development, morphology, survival, or reproduction in the following fish species:
walleye (Stizostedion vitreum), grayling (Thymallus thymallus), mummichog (Fundulus
heteroclitus), rainbow trout (Oncorhynchus mykiss), fathead minnows (Pimephales
promelas), and zebrafish (Danio rerio). However, a recent report on Hg in streams of the
U.S. by the USGS summarizes current research indicating that birds, fish, and fish-eating
wildlife experience negative effects of Hg at lower concentrations than the 0.3 ppm
criterion set to protect human health on the basis of fish consumption.
The 2008 ISA documented a link between decreased S deposition and decreased fish
MeHg concentrations. Recent research in Voyageurs National Park (a Class I Area)
supports this finding, and there is supporting evidence from fish surveys of Texas
reservoirs across regions with different S deposition loads. There is also supporting
evidence from an S addition experiment in a peat bog in the Marcell Experimental Forest
in northern Minnesota, where increased S loading increased Hg concentrations in larval
Culex spp. (mosquitoes), which are an important food source for both aquatic and
terrestrial species (Appendix 12.4 and Figure 12-18). In addition to the studies that
consider S deposition, there are recent studies that consider SO42 concentrations in water
in relation to fish Hg concentrations in six lakes in South Dakota, and in the marshes of
the Everglades (Appendix 12.4). In the freshwater marshes of the Everglades, recent
work indicates a concentration of 1 mg/L S042 to keep water MeHg low
(Appendix 12.3.4.3) and protect fish from elevated Hg burdens in that system
(Figure 12-14).
IS.9.3 National-Scale Sensitivity and Critical Loads
The 2008 ISA identified ecosystems in the Northeast as particularly sensitive to Hg
methylation in response to S deposition because many watersheds in this region have
abundant wetlands and freshwater water bodies with high DOC and low pH. The U.S.
EPA national stream surveys found that MeHg in predator fish exceeded the Hg criterion
in a quarter of stream miles and half the lakes surveyed. Fish MeHg levels were highest
99
-------
in streams in watersheds with considerable wetland area, and surveys showed highest fish
MeHg concentrations in the southeastern U.S., suggesting that ecosystems sensitive to
SOx deposition effects on Hg methylation extend beyond the Northeast (Figure 12-15).
Recent studies confirm that Hg methylation is more widespread than was documented at
the time of the 2008 ISA. New research conducted in agricultural wetlands in California
suggests Hg methylation in these systems may provide a route to animal and human Hg
exposure through food, specifically MeHg concentrations in rice seeds.
There are no CLs for S to prevent sulfide phytotoxicity or Hg methylation, although there
are SO/" and sulfide water quality values that represent protective levels against toxic
effects of sulfide and Hg to biota (see Table 12-12). There are European CLs for Hg
concentrations in soil and fish tissue targeted to protect human health, drinking water
quality, and terrestrial soils, but these CLs are not framed in terms of SOx, Hg, or PM
deposition (see Appendix 12.6).
IS.10 Ecological Effects of Particulate Matter Other Than Nitrogen
(N) and Sulfur (S) Deposition
Since publication of the 2009 PM ISA, new literature builds upon the existing knowledge
of ecological effects associated with PM components other than those associated with N
and S deposition, especially metals and organics. In some instances, new techniques have
enabled further characterization of the mechanisms of PM on soil processes, vegetation,
and effects on fauna. New studies provide additional evidence for community-level
responses to PM deposition, especially in soil microbial communities. However,
uncertainties remain due to the difficulty in quantifying relationships between ambient
concentrations of PM and ecosystem response. Overall, the body of evidence is
sufficient to infer a likely causal relationship between deposition of PM and a
variety of effects on individual organisms and ecosystems, based on information from
the previous review and new findings in this review. However, the new findings are
limited in scope.
PM deposition comprises a heterogeneous mixture of particles differing in origin, size,
and chemical composition. Exposure to a given concentration of PM may, depending on
the mix of deposited particles, lead to a variety of toxic responses and ecosystem effects.
Effects of PM on ecological receptors can be both chemical and physical (U.S. EPA.
2009a. 2004). As described in the 2009 Integrated Science Assessment for Particulate
Matter (2009 PM ISA), particulates that elicit direct and indirect effects on ecological
receptors vary by size, origin, and chemical composition. Ecological outcomes are
attributed more to particle composition than to particle size (Grantz et al.. 2003).
100
-------
PM-associated metals and organics are linked to responses in biota; however, the
heterogeneous nature of PM composition and distribution coupled with the variability
inherent in natural environments confound assessment of the ecological effects of
particulates. Although most effects are from chemical composition of PM, there are some
effects of particle size such as changes to flux of solar radiation and soiling of leaves by
large coarse particles in areas near industrial facilities and unpaved roads. Atmospheric
deposition of PM from crustal material may be a source of base cations (especially Ca2+,
Mg2+, and K+) that can partially ameliorate the effects of acidifying deposition. Base
cations are important plant nutrients that in some locations are in short supply (U.S. EPA.
2009a).
In general, new studies on PM deposition to vegetation support findings in previous PM
reviews on altered photosynthesis, transpiration, and reduced growth. Since the 2009 PM
ISA, additional characterization of PM effects at the leaf surface has led to a greater
understanding of PM foliar uptake. Alterations in leaf fatty acid composition are
associated with metals transferred to plant tissues from PM deposition on foliar surfaces
(Appendix 15.4.2).
An important characteristic of fine particles (0.1 to 1.0 (jm) is their ability to affect the
flux of solar radiation increases in the diffuse component. A newly available research
method links changes in expression of proteins involved in photosynthesis to increases in
the diffuse component due to aerosols and PM. Although this method has not been
widely applied, it may represent an important way to study mechanistic changes to
photosynthesis in response to more diffuse radiation resulting from PM in the air column
(Appendix 15.2).
Several studies published since the 2009 PM ISA show PM chemical constituent effects
on soil physical properties and nutrient cycling. Previous findings in the PM ISA of
changes to microbial respiration and biomass are further supported by new studies.
Microbial communities respond to PM in various ways depending on their tolerance to
heavy metals and organics (Appendix 15.5.3).
In fauna, results from ecotoxicity assays with PM extracts using bacteria, rotifers,
nematodes, zebrafish, and earthworms support findings in the 2009 PM ISA that toxicity
is not related to the total mass of PM in the extract, but to the chemical components of the
PM. In nematodes exposed to PM from air filters, the insulin-signaling pathway was
identified as a possible molecular target. Use of wildlife as PM biomonitors has been
expanded to new taxa since the last PM review. Several studies in invertebrates and birds
report physiological responses to air pollutants, including PM (Appendix 15.6).
101
-------
For ecosystem-level effects, a gradient of response with increasing distance from PM
source was reported in the 2009 PM ISA. Newly available studies from long-term
ecological monitoring sites provide limited evidence for recovery in areas such as those
around former smelters due to the continued presence of metals in soils after operations
ceased. A novel experimental microecosystem using microbial communities living in
terrestrial mosses indicates that PM deposition alters responses of primary producers,
decomposers, and predators (Appendix 15.3).
IS.11 Recovery of Ecosystems from Nitrogen (N) and Sulfur (S)
Deposition in the U.S.
Evidence from across the U.S. of ecosystem recovery from N nutrient enrichment and
acidification corresponding to long-term trends in N and S emissions varies. Most studies
of recovery focus on ecosystem acidification recovery due to decreases in S emissions
and deposition. Overall N emissions and deposition have been increasing or relatively
steady, although a few areas have seen some decrease (Appendix 2.7). Consequently, the
amount of new information available and reported here on N enrichment recovery is
small.
IS.11.1 Overarching Concepts of Ecological Recovery from Acidification
Both chemical and biological indicators are used to assess the degree of ecological
degradation associated with environmental stressors and document responses in
ecosystems where improved conditions allow for recovery. Recovery can be documented
by measurement of indicators and projected/modeled recovery trajectories.
Chemical recovery of aquatic and terrestrial ecosystems is characterized by trends in
water quality indicators (NO3 , SO42 , pH, ANC, inorganic monomeric Al, MeHg)
towards inferred preindustrial values or, in the case of inorganic Al and MeHg, below
water quality threshold values protective of biota and human health. Preindustrial
conditions varied across the U.S. depending on climate, geology, and biological
communities, and preindustrial chemical indicator values are currently inferred from
models, paleolimnology samples, or historical samples. When evaluating ecosystem
recovery from acidification, it is important to note that different chemical pools within
the soil or water column may recover at different rates with the same decreases in
atmospheric deposition. For example, the soil solution Ca:Al ratio, SO42 . or NO3
respond more quickly than will total N. Indicators of slowly recovering pools (such as the
percentage of base saturation in the soil or soil C to N ratio) will have long response
102
-------
times with regard to changes in atmospheric deposition. An indicator such as
acid-neutralizing capacity (ANC), which is influenced by both fast and slow pools, has an
intermediate response time. Chemical indicators such as ANC or pH may not necessarily
follow a recovery path that mirrors the reverse of the acidification path due to dynamic
relationships among ANC, pH, DOC, and inorganic Al; depletion of soil base cation
pools; and/or pH-dependent S adsorption on soils. In addition, the ANC level that reflects
recovery of pH or Al may differ between the acidification and recovery phases
(Hcsthagcn et al.. 2008V
Biological recovery may follow chemical recovery of such water and soil quality
constituents; however, there may be a lag of decades between the onset of chemical
recovery and biological recovery ITJ.S. EPA (2008); Appendix 8]. As observed in some
of the early studies on formerly acidified systems, the biological recovery trajectory may
exhibit hysteresis, where a system does not follow the same path from acidification to
recovery (Frost et al.. 2006). Complete biological recovery would entail a return to the
same species make-up, richness, and abundance as existed in the ecosystem in question
prior to the advent of human-caused acidic deposition (around the year 1860 in North
American ecosystems). In a practical sense, complete biological recovery is probably not
attainable at most acidified locations within a reasonable management time frame
(perhaps 100 years) because soil reserves of base cations at many locations have been
depleted in response to many decades of acidic deposition and because other stressors, in
addition to acidic deposition, have also altered ecosystem structure and/or function or
will do so in the coming decades. Such stressors include changes in climate, land use, and
other perturbations. More commonly, partial biological recovery may be possible.
Ecosystems deemed to be on a recovery trajectory are those found to be moving towards
a mix of species presence and abundance that approximates the undisturbed state. There
is substantial evidence that recovery rates from acidification differ between taxonomic
groups [e.g., rotifers vs. crustaceans; Frost et al. (2006); Mallev and Chang (1994)1. In
general, recovery in freshwater ecosystems is characterized by populations of plankton
and benthic invertebrates prior to the recovery of fish populations, although most
biological communities studied to date have not returned to preacidification conditions,
even after recovery of chemical parameters.
IS.11.2 Acidification Recovery in the U.S.
Long-term monitoring has been very important in tracking the ecological response to N
and acidifying deposition (Appendix 7 and Appendix 4.4). Experimental liming studies
have also provided some evidence for biological recovery, although these types of studies
are limited in the U.S. (Appendix 4.3.4 and Appendix 8.4.6). The historical focus on
103
-------
aquatic acidification has resulted in more data to evaluate recovery in aquatic than
terrestrial ecosystems (Appendix 7.1.3). Fewer studies have tracked the potential
recovery of terrestrial ecosystems; however, since the early 1990s, increasing evidence
indicates that soils in some areas are beginning to recover, yet most sensitive regions
continue to acidify in response to deposition (Appendix 4.6.1). In areas where N and S
deposition has decreased, chemical recovery must first create physical and chemical
conditions favorable for growth, survival, and reproduction of the pre-1860 assemblage
for biological recovery to occur.
The northeastern U.S. and southern Appalachians are two regions of the U.S. where a
large body of research has evaluated recovery. In the Northeast, evidence for chemical
recovery is primarily from soils (Appendix 4.6.1) and freshwater lakes and streams
(Appendix 7.1.5.1). In regard to biological recovery (Appendix 8.4), newer studies have
documented some evidence for zooplankton recovery and the successful reintroduction of
brook trout in previously acidified Adirondack water bodies or recolonization of
previously acidic lakes from refugia (Appendix 8.6.6). In addition to decreased
acidification, a few studies report declines in methylmercury concentrations in biota or
water in response to decreasing S, which is suggestive of ecosystem recovery
(Appendix 12.5).
In contrast to the northeastern U.S., there is little evidence for recovery in the southern
Appalachian Mountain region (Appendix 4.6.1 and Appendix 16.3). This area is
characterized by an abundance of low-ANC streams situated on acidic, highly weathered
soils. Streams in this region are strongly affected by SO42 adsorption on soils, and
long-term monitoring studies suggest that soil base cation depletion has prevented
chemical recovery (Appendix 7.1.5.1.4). Biogeochemistry modeling scenarios suggest
that even with large decreases in SO42 deposition, it may take decades for soil base
cation levels to recover in this region.
New studies continue to support findings in the 2008 ISA that biological response to
water chemistry recovery varies among taxa and water bodies, and that most biological
communities studied have not returned to preacidification conditions, even after recovery
of chemical parameters (Appendix 8.4). Since the 2008 ISA, research has demonstrated
that the DOC of many lakes and streams has risen, with the source of the DOM and
associated DOC likely to be the soils in the terrestrial watershed (Table IS-2;
Appendix 4.3.9 and Appendix 7.1.2.9). The mechanism causing the observed increase in
DOC is unclear; it may be a combination of soil recovery from acidification, changes in
climate (e.g., temperature and precipitation), and N deposition, among other mechanisms.
DOC interacts like a weak acid; therefore, DOC concentration may affect pH and ANC
levels and constrain the extent of recovery from acidification. At the same time, the
104
-------
acidic properties of DOC make it a host for binding trace metals such as toxic inorganic
A1 (for additional discussion on inorganic A1 and DOM see Appendix 4.3.5) and
decreases the toxicity of dissolved A1 to aquatic organisms. Overall, current research
indicates DOC increases are inconsistent across surface waters in the U.S., with large
increases in DOC with acidification recovery in some locations and no increases in other
recovering sites.
IS.11.3 Nitrogen (N) Driven Nutrient Enrichment Recovery in the U.S.
Most freshwater systems sensitive to nutrient effects of atmospheric deposition of N have
shown no evidence for biological recovery, although decreases in NO3 concentrations
consistent with declines in N deposition have been reported in some regions of the U.S.,
notably the Appalachian, Adirondack, and Rocky Mountains (Appendix 7.1.5). Some
estuaries have shown improvements in biological indicators, such as increases in the
extent of SAV, in response to decreases in N inputs from atmospheric deposition and in
wastewater and agricultural runoff. For an example, see the Tampa Bay case study
(Appendix 16). In other coastal areas of the U.S., biological indicators of nutrient
enrichment have remained relatively unchanged or declined. In the well-studied
Chesapeake Bay watershed where extensive restoration efforts have been implemented,
water quality and measures of ecological condition have shown little improvement during
a 23-year period (Williams et al.. 2010V The one exception to the pattern of no
improvement in water quality was an observed increase in the amount of SAV
(Appendix 10.2.5).
IS.12 Climate Modification of Ecosystem Response to Nitrogen (N)
and Sulfur (S) Deposition
Nitrogen and S deposition occur in many ecosystems concurrently experiencing multiple
stressors, including human-driven climate change. Climate change effects on U.S.
ecosystems were recently summarized in the U.S. National Climate Assessment
(Galloway et al.. 2014; Groffman et al.. 2014). Each appendix of the ISA evaluating N
enrichment or acidification includes a section on how climate modifies the ecosystem
response. In the context of this section of the ISA, climate refers to meteorological factors
over a 5-year horizon (because NAAQS are reviewed every 5 years) in contrast to
long-term climate change, or associated changes to CO2 concentrations. Additionally, to
serve as a foundation for the discussion, text in Appendix 13 is excerpted from Greaver et
al. (2016). a current review of how climate (e.g., temperature and precipitation) modifies
ecosystem response to N that focuses on empirical observations.
105
-------
Anthropogenic emissions of greenhouse gases are likely to cause a global-average
temperature increase of 1.5 to 4.0°C and a significant shift in the amount and distribution
of precipitation by the end of the 21st century (Collins et al.. 2013). Recent work has
focused on the effects of anthropogenic N on the Earth's radiative forcing (Pinder et al..
2012) and how temperature and precipitation alter ecological responses to N exposure
(Grcavcr et al.. 2016). Most work is conducted on the effects of climate interactions with
N or acidifying deposition (N + S); relatively little work is conducted on how climate
modifies ecosystem response to S nutrient-related effects.
Understanding climate effects on ecosystems is a rapidly expanding field with many new
empirical studies, meta-analyses, and modeling work published since the 2008 ISA.
General patterns of how climate affects some biogeochemical processes are known and
how climate alters growth rates and biodiversity of some species have been identified,
Figure 13-1 is an example of how processes relevant to N enrichment and acidification
may be altered with either wetter or drier conditions. In addition to the excerpt from
Greaver et al. (2016). additional studies are summarized for effects of climate on N
transport and transformation (Table 13-1), N and C cycling (Table 13-2), acidification
(Table 13-3), and biodiversity (Table 13-4). Our understanding of the effects of climate
on ecosystem response to N and S deposition varies; for many ecological endpoints, data
are insufficient to quantify either the direction or magnitude of how climate may alter
ecosystem response with certainty.
IS.13 Ecosystem Services
"Ecosystem services" refers to the concept that ecosystems provide benefits to people,
directly or indirectly (Costanza et al.. 2017). and that ecosystems produce socially
valuable goods and services deserving of protection, restoration, and enhancement (Boyd
and Banzhaf. 2007). The concept of ecosystem services recognizes that human
well-being and survival are not independent of the rest of nature, and that humans are an
integral and interdependent part of the biosphere (Costanza et al.. 2017). In some cases,
and in line with more conventional economic thinking, ecosystem services analysis can
result in attaching monetary values to ecosystem outcomes. However, because ecosystem
services are often public goods their benefits can be difficult to monetize. We emphasize
that this practical difficulty in no way implies that ecosystem service benefits are small or
without value. At a minimum, ecosystem services analysis involves discussion and,
ideally, quantification of ecological outcomes understood by households, communities,
and businesses. Explicitly linking ecosystem services to social and economic welfare
measures has proven difficult because of the broad definition of ecosystem services and
the numerous types of services that could be affected. An analysis of ecosystem services
106
-------
specifically altered by NOx, SOx, and PM would translate the effects of ambient
concentrations and deposition into biological, physical, or monetary metrics that give
insight to public welfare effects.
For acidification, the ecosystem service literature since the 2008 ISA includes studies that
better characterize ecosystem service valuation by pairing biogeochemical modeling and
benefit transfer equations informed by willingness-to-pay surveys, especially for the
Adirondacks and Shenandoah regions (Appendix 14). Aside from valuation studies, there
is an improved understanding of the numerous causal pathways by which N and S
deposition may affect ecosystem services, supported by studies that relate deposition to
final ecosystem services under the FEG-CS (Bell et al.. 2017; Clark et al.. 2017; Irvine et
al.. 2017; O'Dea et al.. 2017; Rhodes et al.. 2017). However, for many regions and
specific services, poorly characterized dose-response between deposition, ecological
effect, and services are the greatest challenge in developing specific data on the economic
benefits of emission reductions (NAPAP. 2011).
In the 2008 ISA there were no publications that specifically evaluated the effects of N
deposition on ecosystem services associated with N driven eutrophication. Since then
several comprehensive studies have been published on the ecosystem services related to
N pollution in the U.S. (Appendix 14). These include an evaluation of services affected
by multiple N inputs (including N deposition) to the Chesapeake, a synthesis of the
cost-benefits on N loading across the nation, and analysis of the amount of N that leaked
out of its intended application area causing effects on adjacent ecosystems and ecosystem
services, two calculations of the social cost of nitrogen (Minnesota and the Mississippi
Alluvial Valley), and an estimate of the cost to remove N from the White River Basin in
Indiana (this work specifically identified the costs of the atmospheric portion of total N
loading). The estimate of the total number of ecosystem services affected by N is better
quantified by the new studies that use FEG-CS (Bell et al.. 2017; Clark et al.. 2017;
Irvine et al.. 2017; O'Dea et al.. 2017; Rhodes et al.. 2017). In these analyses, CL
exceedances for N related air pollution were used as a model stressor from which a total
of 1,104 unique chains linking stressor to beneficiary were identified.
The conclusions considering the full body of literature are that (1) there is evidence that
N and S emissions/deposition have a range of effects on U.S. ecosystem services and
their social value; (2) there are some economic studies that demonstrate such effects in
broad terms; however, it remains methodologically difficult to derive economic costs and
benefits associated with specific regulatory decisions/standards; and (3) there is an
improved understanding of the numerous causal pathways by which N and S deposition
ay affect ecosystem services, though most of these causal relationships remain to be
quantified.
107
-------
IS.14
Key Scientific Uncertainties
Evaluation of uncertainty is an important part of ecosystem assessment. Uncertainty
refers to the absence of information and is a way to describe how certain we are in
scientific knowledge. As described by Curry and Webster (2011). the nature of
uncertainty can be expressed by the distinction between ontic uncertainty and epistemic
uncertainty. Ontic uncertainty is associated with inherent variability or randomness and is
an irreducible form of uncertainty. Epistemic uncertainty is associated with imperfections
of knowledge, which may be reduced by further research and empirical investigation.
Walker et al. (2003) [as summarized in Curry and Webster (2011)1 characterized
uncertainty as a progression from deterministic understanding to total ignorance:
"Statistical uncertainty is the aspect of uncertainty that is described in
statistical terms. An example of statistical uncertainty is measurement
uncertainty, which can be due to sampling error or inaccuracy or
imprecision in measurements.
"Scenario uncertainty implies that it is not possible to formulate the
probability of occurrence of one particular outcome. A scenario is a
plausible but unverifiable description of how the system and/or its
driving forces may develop overtime. Scenarios may be regarded as a
range of discrete possibilities with no a priori allocation of likelihood.
"Recognized ignorance refers to fundamental uncertainty in the
mechanisms being studied and a weak scientific basis for developing
scenarios. Reducible ignorance may be resolved by conducting further
research, whereas irreducible ignorance implies that research cannot
improve knowledge."
The understanding and reporting of uncertainty is not consistent across scientific
disciplines, and uncertainty may be quantified by various methods. Csavina et al. (2017)
provided an overview of terminology and definitions of 41 different terms used to
describe uncertainty. Here we provide a summary of some of the key methods that may
be used to evaluate the uncertainty of the relationships between NOx, SOx, and PM
pollutants and ecological effects. This summary presents uncertainties associated with
several specific concepts, including source emissions measurements, atmospheric
deposition estimates, empirical measurements of CLs, models used to estimate CLs, and
uncertainties in the aquatic acidification index. Quantified estimates of uncertainty vary
according to the number of decision points (Section IS. 14.2.3). including the method used
and the input parameters under consideration; therefore, the analyses and discussion of
108
-------
quantified uncertainty values will occur in the Risk and Exposure Assessment as scoped
in the 2017 IRP (U.S. EPA. 2017a).
IS.14.1 Atmospheric Science
Estimating atmospheric deposition involves quantification of emissions, atmospheric
concentrations, and deposition fluxes of the various species that make up atmospheric
SOx, NOy, and NHx. This is accomplished with environmental measurements, model
predictions, or hybrid approaches that combine measurements and modeling methods.
There are a wide range of uncertainties across the environmental measurements and
model parameters used to estimate atmospheric deposition fluxes. The largest
uncertainties are those for dry deposition and ammonia emissions, whether measured or
modeled. The smallest uncertainties are associated with ambient concentration
measurements and continuously monitored stationary emissions like electric power
plants.
IS.14.1.1 Emissions Uncertainty
Quantitative uncertainty estimates are not documented in the National Emissions
Inventory (NEI), but uncertainties are often evaluated through separate efforts by
comparing inventory predictions with measured long-term trends, statistical source
apportionment methods, inverse chemical transport modeling, and comparison with
satellite data (Appendix 2.2.2). SO2 and NOx emission uncertainties for
electricity-generating units, the major source of SO2 and an important source of NOx, are
in the 10-15% range because emissions are usually continuously monitored
(Appendix 2.2.3). NOx emission uncertainties for mobile sources, the largest source of
NOx, arise from differences in engine type, size, age, and maintenance, as well as fuel
composition and emission control equipment. Overestimation of NOx emissions from
mobile sources was proposed as an explanation for modeled NOx concentration bias in
several studies. However, mixed results have been observed across several studies when
modeled concentrations were compared with measurements. Estimates of NOx emissions
uncertainties are in the 10-20% range for on-road gasoline and diesel vehicles, and up to
30% for off-road vehicles like ships, airplanes, and locomotives (Appendix 2.2.3). Spatial
and temporal variability in soil NOx emissions can lead to uncertainty in emissions
estimates. Soil emissions occur mainly during summer and across the U.S., but some
areas, such as the central Corn Belt of the U.S., release more NOx emissions than others
(Appendix 2.2.3).
109
-------
In contrast, total NH3 emissions uncertainties appear to be greater, underestimated by as
much as a factor of two or more according to several recent studies (Appendix 2.2.3). The
predominant sources, livestock operations and fertilizer application, exhibit large
temporal and regional variability due to differences in climate conditions and farming
practices. As a result, detailed models are required for estimating NH3 emissions
(Appendix 2.2.2), but data on local environmental conditions and farming practices
necessary for good model performance are often not available. Large discrepancies
between modeled and measured N concentrations and deposition rates have been
attributed to uncertainties in NH3 emissions (Appendix 2.2.3). Activity rates, including
those for mobile source emissions, are also difficult to quantify, contributing to
uncertainty in NH3 emission estimates (Appendix 2.2.3).
IS.14.1.2 Atmospheric Measurement Uncertainty
Uncertainties in concentration and deposition measurements from network-based
measurements are generally under 20%, and surface concentration uncertainties from
satellite-based measurements typically somewhat higher. Concentration and deposition
data are derived from several specialized national monitoring networks, including the
national SO2 monitoring network, the NCore network for multipollutant concentration
monitoring including NOy, the Ammonia Monitoring Network, CASTNet for estimating
dry deposition, and the National Trends Network for wet deposition (Appendix 2.4.1).
Uncertainties are estimated from reports of precision in data quality reports where
available, and otherwise from network data quality objectives.
For air concentration measurements used to estimate dry deposition, CASTNet measured
precision was 2-5% for SO42 , 5-13% forN03~, and 2-6% for NH3 in 2016
(Appendix 2.4.5). Additional uncertainty is associated with estimating dry deposition
from NTN concentration data. Uncertainties of 30% for SO2 and 40% for HN03 have
been reported using a simple inferential approach (Clarke et al.. 1997V However, single
site determinations are of limited use because dry deposition fluxes are determined by
several factors and can vary considerably over small spatial scales. In most recent efforts,
dry and total deposition on a regional or national scale is usually modeled with CTMs
(Section IS. 14.1.3).
Precipitation concentration measurement precision and estimated wet deposition
precision in the National Trends Network were less than 7% for SO42 and N03 and less
than 20% for NH3. PRISM (Parameter-elevation Regression on Independent Slopes
Model) enhances spatial resolution using National Trends Network data to improve the
creation of wet deposition maps (Appendix 2.6). Uncertainty for PRISM data sets has
110
-------
been evaluated in the literature using cross validation and a 70% prediction interval for
different data sets. NH3 air concentration measurement methods used in AMoN were
evaluated and found to have a precision of 10% (Appendix 2.4.3). Minimum performance
specifications for SO2 monitoring from the national SO2 monitoring network include a
precision of 2.0% (Appendix 2.4.4). Data quality objectives for NOy in the NCore
network include a precision of 15% (Appendix 2.4.2). Uncertainty in satellite-based
measurements depend on vertical profile, cloud fraction, cloud-top pressure, surface
reflectivity, and extent of aerosol scattering. Estimates of 20% for NO2 (Appendix 2.4.2)
and 10-45% for SO2 (Appendix 2.4.4) have been reported for cloud-free conditions.
IS.14.1.3 Atmospheric Modeling Uncertainty
The Community Multiscale Air Quality modeling system is probably the most widely
used model in the U.S. for estimating atmospheric deposition. CMAQ accurately
modeled total SOx, but partitioning resulted in overpredicting SO2 and underpredicting
SO42 . In a recent CMAQ evaluation, SO2 concentrations were overestimated by 39 to
47%, and SO42 concentrations were underestimated by 9 to 17%, as annual averages
over a range of 4 years compared to surface-based measurements. In addition,
atmospheric NO3 concentrations were overestimated by 22 to 26%, as annual averages
over a range of 4 years compared to surface-based measurements (Appendix 2.5.3).
Mixed results have been observed in several recent comparisons of CMAQ wet
deposition estimates to network-based measurements, with average differences in
modeled results and measurements ranging from <15 to 99% for NO3 . SO42 , and <15 to
60% for NH3 (Appendix 2.5.3). Modeling methods for estimating dry and total deposition
are still under development, and uncertainties have not been extensively evaluated or
quantified. Recent sensitivity analysis results found less than 5% differences in total
deposition estimates because of compensation of competing model processes, but
extensive comparison of model results and measurements are not available
(Appendix 2.5.3).
Horn et al. (2018) used deposition and forest inventory data (from 2000 to 2016) to assess
the relationship between deposition and growth and survival of 71 tree species across the
contiguous U.S. in a correlational analysis. Authors attempted to reduce uncertainty by
accounting for other variables, either directly in their model or by quantifying and
avoiding instances with high collinearity. The authors isolated the effects of N deposition
from S deposition by adding S deposition explicitly into their models. Using variance
inflation factors (VIFs), they also quantified the collinearity of N and S deposition against
a suite of environmental variables that might have an effect. The analysis focused on the
relationships of tree growth and survival to N and S deposition where the VIF was less
111
-------
than or equal to 3. VIF values of 3-10 have been proposed as thresholds above which
there is a potential for high collinearity (Horn et al.. 2018). To what extent variables not
included could have varied with deposition and had an effect, including ozone and
drought, remained a key uncertainty.
Clark et al. (2018) analyzed exceedances of multiple types of CLs for the contiguous U.S.
since 1800 and projecting out to 2025. The study authors discussed the uncertainty
around CMAQ deposition estimates using CMAQ estimates starting in 1980. They noted
that CMAQ may underestimate hot spots of deposition in space (e.g., concentrated
deposition because of an orographic effect) or in time (e.g., from cloudbursts). CLs are an
ecosystem response to deposition, and so any errors associated with deposition estimates
would propagate through CLs. Fenn et al. (2010) found that CMAQ estimates and N in
throughfall were similar under low throughfall conditions, but CMAQ underestimated N
deposition when throughfall was high. Clark et al. (2018) noted that CMAQ is corrected
using NADP data, but NADP sites do not provide complete spatial coverage. Remote
sites are likely underrepresented.
In addition to measurable uncertainties associated with measurement precision or
comparisons between models and measurements, there are also structural uncertainties
due to incomplete understanding of the underlying science related to atmospheric
deposition that are not possible to quantify. The main structural uncertainties associated
with deposition estimates are canopy effects on NOx (including both bidirectional gas
exchange and canopy reactions), bidirectional exchange of NH3 with biota and soils, and
processes determining transference ratios that relate average concentration to deposition
(Appendix 2.5).
IS. 14.2 Ecological Effects
Evaluation of ecological effects caused by acidification or eutrophication involves a suite
of parameters and dose-response functions, both empirical and modeled. The quantitative
uncertainty of empirically observed variables in ecology is determined by using statistics.
A suite of mathematical statistical models is available to describe the variability among
empirical observations and the strength of a cause and ecological effect relationship, the
appropriate method to apply depends on the experimental design. Statistics for empirical
data include calculation of probability, distributions, standard deviation, variance, /-tests.
ANOVA, linear regression, spatial statistics, Bayesian analysis, and multivariate analysis,
among others. In general, ecological endpoints determined by empirical studies to be
affected by deposition were reported in the ISA if they were statistically significant; this
means the magnitude of effect was larger than the estimated uncertainty.
112
-------
Models of chemical and ecological processes, including biogeochemistry, provide
representations of biological and geochemical interactions through mathematical
expressions. The models used to characterize aquatic and terrestrial biogeochemistry
response to N and S deposition can be complex, including many interacting variables.
Model results are often compared to empirically collected data to confirm the model.
Each of the input variables used in a biogeochemical model entails uncertainty. Model
uncertainty is governed, in part, by how close the model predictions are to actual
observations. Uncertainty in modeled results may arise from limitations in input data or
from limitations in model assumptions. Statistical inference methodologies enable
uncertainty analysis and determine the strength of the relation between a given uncertain
input and the output (i.e., sensitivity analysis). For biogeochemistry models these
methods include first-order sensitivity index, Monte Carlo technique, extended Fourier
amplitude sensitivity test, Morris one-factor-at-a-time, and Bayesian analysis.
IS.14.2.1 Empirical Critical Loads
Empirical N CLs for terrestrial and aquatic ecosystems reported in this ISA have been
estimated using empirical data sets. The exact effects threshold may be determined using
expert judgement. For example, if three levels of N addition are applied to a study site
(10, 20, and 30 kg N/ha/yr) and an effect is noticed at 20 kg N/ha/yr, then the CL is
estimated at <20 kg N/ha/yr. Another approach would be to fit a mathematical function to
the observations, and a scientific judgement made to identify the level of deposition
and/or N addition, or threshold, at which the ecological effect is considered to occur and
which is likely to be biologically adverse.
There are some challenges associated with developing CLs that can result in uncertainty.
First, because biological responses are often continuous, there can be a lack of an obvious
cutoff between adverse and nonadverse effects. As a result, individual author groups have
selected different response thresholds. For example, N CLs for lichens have been
calculated for (1) deposition values associated with thallus N concentrations above the
97% distribution quantile observed for clean sites (Fenn et al.. 2008). (2) community
composition shifts from oligotroph to eutroph dominance (Fenn et al.. 2008). (3) low
probability of detecting regionally distributed sensitive species (Root et al.. 2015; Geiser
et al.. 2010). or (4) extirpation of oligotrophs (Fenn et al.. 2008). Secondly, clean site data
can be lacking in some ecoregions. For instance, few empirical data are available for sites
in the eastern U.S. with deposition rates <4 kg N/ha/yr. This makes it difficult to quantify
physiological or community compositional conditions that may have occurred in this
region at deposition rates of 1-4 kg N/ha/yr.
113
-------
The Pardo et al. (2011a) study provided a compilation of terrestrial and aquatic N CLs
reported since the 2008 ISA. Uncertainty in the derivation of empirical CLs for N input
as presented by Pardo etal. (2011a) arises in estimating the ambient (and perhaps
historical) deposition loads and in estimating the biological effects caused by those
deposition levels. According to Pardo etal. (2011a). sources of uncertainty in N
deposition estimates forN CLs at the Ecoregion Level 1 scale include "(1) the difficulty
of quantifying dry deposition of nitrogenous gases and particles to complex surfaces;
(2)	sparse data, particularly for arid, highly heterogeneous terrain (e.g., mountains); and
(3)	sites with high snowfall or high cloud water/fog deposition, where N deposition tends
to be underestimated." Examples of high uncertainty include high-elevation sites in the
Rockies and Sierra Nevada mountains, due in part to highly uncertain estimates of dry
deposition (Appendix 2). For sensitive receptors such as phytoplankton, shifts in
high-altitude lakes, N deposition model bias may be close to, or exceed, predicted CL
values (Williams et al.. 2017a).
Physical, chemical, and ecological variability across lakes affect their response to N
deposition and contribute to uncertainty of CL estimates (Appendix 9.1.1.2). A review by
Bowman et al. (2014) noted that current N CLs for sensitive alpine systems may not be
protective under future climate scenarios of warmer summer temperatures and a shorter
duration of snow cover.
Between the publication of Pardo et al. (2011a) and the cutoff date for literature in this
ISA (May 2017), some additional aquatic and terrestrial N CLs have been published
(Appendix 4; Appendix 6.5). Simkin et al. (2016) was not based on field addition or N
gradient of deposition studies; instead, the methods were a spatial analysis of plant
diversity using a large data set of over 15,000 forest, shrubland, and herbaceous sites
across the U.S. Atmospheric N deposition varied nearly 20-fold across the site gradient.
The study authors found that N deposition was negatively correlated with plant species
richness at many locations, but positively correlated at others with most of the positive
correlations in areas with low N deposition averaging 3 kg N ha/yr or less. Simkin et al.
(2016) also estimated the uncertainty surrounding the mean CL estimates. For open
canopy ecosystems, for example, they estimated a mean of 8.7 kg N ha/yr and provided
95% confidence intervals, which can be used as estimates of uncertainty, of 6.4 to
11.3 kg N ha/yr. For closed canopy systems, the mean of 13.4 kg N ha/yr was surrounded
by a 95% confidence interval of 6.8 to 22.2 kg N ha/yr.
Clark et al. (2018) noted that many of the CLs used are empirically derived. Some of the
uncertainties with these CLs are that they are often from one or two studies at a given
location or area and extrapolated to a larger area, such as an entire Level 1 ecoregion.
Thus, there is uncertainty about how representative these are for larger areas. As noted in
114
-------
Horn et al. (2018). there also can be covariates unaccounted for that could affect
estimates of CLs. CLs also do not generally account for historical effects that already
might have affected the ecosystem. There are also uncertainties regarding process-based
CLs, such as the terrestrial acidification CL. Clark et al. (20IS) specifically pointed
towards the existence of poor estimates of soil weathering despite the importance of soil
weathering estimates for acidification CLs.
The majority of studies that evaluate terrestrial N CLs for N enrichment effects are based
on observed response of a biological receptor to N deposition (or N addition as a proxy
for deposition), without a known soil chemistry threshold that causes the biological
effect. In contrast, CLs for acidification are typically based on the deposition amount that
gives rise to a soil chemical indicator value which is known to cause an adverse
biological effect. The link between soil chemical indicator and biological effect is based
on empirical evidence (Appendix 5). The relationship between deposition and the
biogeochemistry that causes effects on soil chemistry is typically modeled (Appendix 4;
Section IS. 14.2).
IS.14.2.2 Modeled Critical Loads
IS.14.2.2.1 Terrestrial and Aquatic Acidification: Biogeochemistry
A variety of process models have been used to estimate past and future resource
conditions under scenarios of acidification/recovery responses and critical and target
loads, both aquatic and terrestrial. Models include simple approaches such as the simple
mass-balance equation (SMBE), and dynamic models, such as PnET-BGC and ForSAFE,
MAGIC, VSD, and VSD+ (Appendix 4.5). CLs for terrestrial and aquatic acidification
are calculated by the model to determine the amount of deposition that alters soil or water
chemistry to a threshold value known to have detrimental effects on a biological receptor.
Each of the several well-established models of terrestrial biogeochemistry used to
evaluate soil acidification (Appendix 4.5) rely heavily on input or simulated values for
base cation weathering (BCw) rate, one of the most influential yet difficult to estimate
parameters in the calculation of critical acid loads of N and S deposition for protection
against terrestrial acidification (Appendix 4.5.1.1). Obtaining accurate estimates of
weathering rates is difficult because weathering is a process that occurs over very long
periods of time, and the estimates on an ecosystem's ability to buffer acid deposition rely
on accurate estimates of weathering. Various approaches can be used to estimate BCw,
including the empirical soil clay approach, the PROFILE model [e.g., Phelan et al.
(2014)1. the F-factor approach (U.S. EPA. 2009c). and calibration of a dynamic model
115
-------
such as MAGIC [e.g., Povak et al. (2014); McDonnell et al. (2014)1. There are new
studies on estimating BCw, including evaluation of uncertainty (Whitfield et al.. 2018;
Futter et al.. 2012). When applying PROFILE to upland forests in the U.S., Whitfield et
al. (2018) found the greatest uncertainty in BCw estimate was due to the particle size
class-based method used to estimate the total specific surface area on which weathering
reactions can take place.
The uncertainty of forest soil CLs for acidification in U.S. calculated using simple
mass-balance equations (SMBE) was investigated by Li and McNultv (2007). The results
included a quantification of how 17 of the model's parameters contributed to the
uncertainty and indicated that uncertainty in the CLs came primarily from components of
base cation weathering and acid-neutralizing capacity, whereas the most critical
parameters were BCw base rate, soil depth, and soil temperature. The study authors
concluded that improvements in estimates of these factors are crucial to reducing
uncertainty and successfully scaling up SMBE for national assessments (see
Appendix 4.6).
Several dynamic models are commonly used to model terrestrial soil acidification
(Appendix 4.5). Tominaga et al. (2009) conducted a Monte Carlo multiple-model
evaluation of the dynamic models MAGIC, SAFE, and VSD and found that given the
same deposition scenario, the three models (without calibration) simulate changes in soil
and soil solution chemistry differently, but the basic patterns were similar. The study
authors also found the greatest differences in model outputs were attributed to the cation
exchange submodel. Bonten et al. (2015) compared how well the common types of
dynamic models used to evaluate terrestrial soils (VSD, MAGIC, ForSAFE, and
SMARTml) quantified several variables including soil S, soil pH, soil ANC, BC, base
saturation, and Al (Appendix 4.5.3).
Uncertainty analysis of a dynamic model (VSD) used for CL based on soil chemistry
chemical limits showed that the main drivers of uncertainty were largely dependent on
the chemical criterion selected [Appendix 5.5.3.3; Reinds and de Vries (2010)1. For
example, base cation weathering, deposition, and the parameters describing the H-Al
equilibrium in the soil solution were the main sources of uncertainty in the estimates of
maximum CLs for S (Clmax[S]) based on the Al:Bc criterion of 1.0, and uncertainty in
Clmax(S) based on ANC was completely determined by base cation inputs. The
denitrification fraction was the most important source of uncertainty for the maximum
CLs of N (Clmax[N]). Calibration of VSD reduced the levels of uncertainty for all CLs
and criteria.
Fakhraei et al. (2017b) reviewed sensitivity and uncertainty analysis techniques
(e.g., first-order sensitivity index, Monte Carlo technique, extended Fourier amplitude
116
-------
sensitivity test, Morris one-factor-at-a-time, and Bayesian analysis) in the context of a
biogeochemistry model. The study authors applied these techniques to determine the
uncertainty and sensitivity of the PnET-BGC model calculation of TMDLs of acidifying
deposition that occur in high-elevation, acid-impaired streams in GSMNP (Fakhraei et
al.. 2017a). Sensitivity analyses showed that modeled estimates of maximum allowable
acidifying deposition loads were most sensitive to uncertainty in model input parameters
of air temperature, precipitation quantity, and rate of calcium weathering. Importantly, as
more uncertainty was incorporated into model input parameters (±5 to ±10 to ±20%
uncertainty), estimates of allowable deposition loads to protect aquatic ecosystem
recovery decreased in magnitude (Fakhraei et al.. 2017a).
15.14.2.2.2	Biogeochemistry and Plant Biodiversity Linked Modeling
Plant biodiversity models, such as VEG and PROPS, have been coupled to dynamic
biogeochemical models, such as ForSAFE and VSD+ (Mcdonnell et al.. 2018b;
Mcdonnell et al.. 2018a; Phelan et al.. 2016). ForSAFE-VEG is an older and more
broadly applied model than VSD + PROPS. There are some key differences between
VEG and PROPS. Plant species in the VEG component of ForSAFE-VEG are defined by
mathematical equations based on expert opinion regarding such parameters as plant needs
for moisture, sunlight, and N supply to represent unobservable fundamental niches. In the
PROPS, statistical relationships based on empirical data are used to characterize plant
species, which are more likely to approximate real-world niches influenced by
competition among species. These model chains are subject to the same constraints and
uncertainties as the biogeochemical models on their own, plus those of the plant response
modules.
15.14.2.2.3	Aquatic Eutrophication Modeling
Many of the models that estimate N loads to the coastal zone from land-based inputs
(agricultural practices, sewage, atmospheric deposition, natural lands) and freshwater
inflow have been compared, and there is a good deal of knowledge about their limitations
and uncertainties (McCrackin et al.. 2013; Alexander et al.. 2008). A National Research
Council review determined that these models are hydrodynamically complex and tend to
be site specific. Thus, they are difficult to apply broadly (NRC. 2000).
The SPARROW model application used only wet N deposition. A large amount of N
from nonpoint source urban influences (most likely due primarily to the dry deposition of
exhaust N gases) often approximately doubles the importance of N deposition as an N
source to higher order river systems (Howarth. 2008a. b).
117
-------
IS.14.2.3 Additional Key Considerations for Critical Loads
The choice of model for CL estimation, or for scenario projection, depends largely on the
availability of time, data, and resources. Major decisions inherent in the modeling efforts
include:
•	Empirical observation or application of a model
•	Steady-state or dynamic model
•	Statistical or process-based model
•	Protection against acidification or nutrient N enrichment
•	Site-specific, regional, or national spatial scale
•	Resources to be protected (i.e., stream, lake, soil, vegetation, aquatic biota)
•	Chemical indicator(s) of adverse effects (e.g., water ANC, water NO3 , soil BS)
•	Critical level(s) for selected indicator(s)
•	Time frame of evaluation (i.e., ambient, 2050, long-term steady state)
Each of these decision points introduces additional uncertainties, data needs, and
potential assessment errors. U.S. EPA (2008) summarized CL research and monitoring
needs identified by U.S. EPA (2006a) at the time of the previous (2009) U.S. EPA Risk
and Exposure Assessment.
IS.14.3 Aquatic Acidification Index
Detailed analysis of uncertainty in the AAI equation can be found in Appendix F of the
2011 Policy Assessment for the Review of the Secondary National Ambient Air Quality
Standards for Oxides of Nitrogen and Oxides of Sulfur (U.S. EPA. 2011). The AAI is
made up of components including ecosystem effects; dose-response relationships;
underlying ecosystem sensitivity to acid deposition, biogeochemical, atmospheric and
deposition processes; and characterization of ecosystem services. Some degree of
uncertainty exists in all of the components of the AAI. Overall, the 2011 Policy
Assessment found, on balance, low uncertainty in the information and processes
associated with linkages from ecological effects to atmospheric conditions through
deposition and ecosystem modeling. However, it acknowledged the need to improve
certainty of several components including nitrogen and sulfur deposition processes in
CMAQ, natural emissions of NOx from lightning processes, and improving the amount
of samples of CL estimates at several ecoregions (U.S. EPA. 2011).
118
-------
IS. 15 References
Alexander. RB; Smith. RA; Schwarz. GE; Bover. EW; Nolan. JV; Brakebill. JW. (2008).
Differences in phosphorus and nitrogen delivery to the Gulf of Mexico from the
Mississippi River Basin. Environ Sci Technol 42: 822-830.
http://dx.doi.org/10.1021/esQ716103
Baker. JP; Bernard. DP: Christensen. SW: Sale. MJ; Freda. J: Heltcher. K; Marmorek. D;
Rowe. L; Scanlon. P; Suter. G: Warren-Hicks. WJ: Welbourn. PM. (1990). Biological
effects of changes in surface water acid-base chemistry. (NAPAP Report 13). Washington,
DC: National Acid Precipitation Assessment Program.
https://ntrl.ntis. gov/NTRL/dashboard/searchResults.xhtml?searchQuerv=PB92100494
Baldigo, BP: Lawrence. GB; Bode. RW: Simonin. HA: Roy. KM: Smith. AJ. (2009). Impacts
of acidification on macroinvertebrate communities in streams of the western Adirondack
Mountains, New York, USA. Ecol Indicat 9: 226-239.
http://dx.doi. org/10.1016/i. ecolind.2008.04.004
Balvanera. P; Pfisterer. AB; Buchmann. N: He. JS: Nakashizuka. T; Raffaelli. D: Schmid. B.
(2006). Quantifying the evidence for biodiversity effects on ecosystem functioning and
services. Ecol Lett 9: 1146-1156. http://dx.doi.org/10.1111/i. 1461-0248.2006.00963,x
Barnthouse. LW: Munns. RM. Jr: Sorensen. MT. (2008). Population-level ecological risk
assessment. Pensacola, FL: Taylor & Francis.
http://www.crcpress.com/product/isbn/9781420Q53326
Baron. JS: Driscoll. CT: Stoddard. JL: Richer. EE. (2011). Empirical critical loads of
atmospheric nitrogen deposition for nutrient enrichment and acidification of sensitive US
lakes. Bioscience 61: 602-613. http://dx.doi.org/10.1525/bio.2011.61.8.6
Bartell. SM. (2007). Ecosystem effects modeling. In GW Suterll (Ed.), Ecological risk
assessment (2nd ed., pp. 413-432). Boca Raton, FL: CRC Press.
Bell, MP; Phelan, J: Blett, TF; Landers, D; Nahlik, AM; Van Houtven, G; Davis, C; Clark,
CM: Hewitt, J. (2017). A framework to quantify the strength of ecological links between
an environmental stressor and final ecosystem services. Ecosphere 8.
http://dx.doi.org/10.1002/ecs2.1806
Bonten, LTC; Reinds, GJ; Groenenberg, JE; de Vries, W; Posch, M; Evans, CD: Belyazid, S;
Braun, S: Moldan, F; Sverdrup, HU: Kurz, D. (2015). Dynamic geochemical models to
assess deposition impacts and target loads of acidity for soils and surface waters. In W de
Vries; JP Hettelingh; M Posch (Eds.), Critical loads and dynamic risk assessments:
Nitrogen, acidity and metals in terrestrial and aquatic ecosystems (pp. 225-251). Bilthoven,
The Netherlands: Springer Netherlands. http://dx.doi.org/10.1007/978-94-017-95Q8-l 8
Bowman, WD: Nemergut, PR: McKnight, DM: Miller, MP: Williams, MW. (2014). A slide
down a slippery slope - Alpine ecosystem responses to nitrogen deposition. Plant Ecol
Divers 8: 727-738. http://dx.doi.org/10.1080/17550874.2Q14.984786
119
-------
Boyd. J; Banzhaf. S. (2007). What are ecosystem services? The need for standardized
environmental accounting units. Ecol Econ 63: 616-626.
Bricker. S; Longstaff. B; Dennison. W; Jones. A; Boicourt. K; Wicks. C; Woerner. J. (2007).
Effects of nutrient enrichment in the nation's estuaries: A decade of change: National
estuarine eutrophication update. Silver Spring, MD: National Centers for Coastal Ocean
Science, https://repositorv.librarv.noaa.gov/view/noaa/17779
CAA (Clean Air Act). (1990). Clean Air Act, as amended by Pub. L. No. 101-549, section
108: Air quality criteria and control techniques, 42 USC 7408.
http://www.law.cornell.edu/uscode/text/42/7408
Caires. AM: Chandra. S: Havford. BL; Wittmann. ME. (2013). Four decades of change:
Dramatic loss of zoobenthos in an oligotrophic lake exhibiting gradual eutrophication.
Freshw Sci 32: 692-705. http://dx.doi.org/10.1899/12-064.1
Cardinale. BJ; Duffy. JE; Gonzalez. A: Hooper. DU; Perrings. C: Venail. P; Narwani. A:
Mace. GM; Tilman. D; Wardle. DA: Kinzig. AP; Daily. GC: Loreau. M; Grace. JB;
Larigauderie. A: Srivastava. DS: Naeem. S. (2012). Biodiversity loss and its impact on
humanity [Review], Nature 486: 59-67. http://dx.doi.org/10.1038/naturel 1148
Cardinale. BJ: Matulich. KL; Hooper. DU: Byrnes. JE: Duffy. E; Gamfeldt. L; Balvanera. P;
O'Connor. MI; Gonzalez. A. (2011). The functional role of producer diversity in
ecosystems. Am J Bot 98: 572-592. http://dx.doi.org/10.3732/aib.1000364
Charles. DF; Christie. S. (1991). Acidic deposition and aquatic ecosystems: Regional case
studies. In DF Charles; S Christie (Eds.). New York, NY: Springer-Verlag.
http ://dx.doi .org/10.1007/978-1-4613 -903 8-1
Clark. CM; Bell. MD; Boyd. JW; Compton. JE; Davidson. EA; Davis. C; Fenn. ME; Geiser.
L; Jones. L; Blett. TF. (2017). Nitrogen-induced terrestrial eutrophication: Cascading
effects and impacts on ecosystem services. Ecosphere 8: e01877.
http://dx.doi.org/10.1002/ecs2.1877
Clark. CM; Phelan. J; Doraiswamy. P; Buckley. J; Caika. JC; Dennis. RL; Lynch. J; Nolte.
CG; Spero, TL. (2018). Atmospheric deposition and exceedances of critical loads from
1800-2025 for the conterminous United States. Ecol Appl 28: 978-1002.
http://dx.doi.org/10.1002/eap.1703
Clarke. JF; Edgerton. ES; Martin. BE. (1997). Dry deposition calculations for the clean air
status and trends network. Atmos Environ 31: 3667-3678.
http://dx.doi.org/10.1016/S1352-2310(97V)0141-6
Collins. M; Knutti. R; Arblaster. J; Dufresne. JL; Fichefet. T; Friedlingstein. P; Gao. X;
Gutowski. WJ. Jr; Johns. T; Krinner. G; Shongwe. M; Tebaldi. C; Weaver. AJ; Wehner.
M. (2013). Long-term climate change: Projections, commitments and irreversibility. In TF
Stocker; D Qin; GK Plattner; MMB Tignor; SK Allen; J Boschung; ANauels; Y Xia; V
Bex; PM Midgley (Eds.), Climate change 2013: The physical science basis Contribution of
working group I to the fifth assessment report of the Intergovernmental Panel on Climate
Change (pp. 1029-1136). Cambridge, UK: Cambridge University Press.
http://www.ipcc.ch/report/ar5/wgl/
120
-------
Costanza. R; De Groot R: Braat L; Kubiszewski. I; Fioramonti. L; Sutton. P; Farber. S;
Grasso. M. (2017). Twenty years of ecosystem services: How far have we come and how
far do we still need to go? Ecosyst Serv 28: 1-16.
http://dx.doi.Org/10.1016/i.ecoser.2017.09.008
Csavina. J: Roberti. JA; Taylor. JR; Loescher. HW. (2017). Traceable measurements and
calibration: a primer on uncertainty analysis. Ecosphere 8.
http://dx.doi.org/10.1002/ecs2.1683
Curry. JA: Webster. PJ. (2011). Climate science and the uncertainty monster. Bull Am
Meteorol Soc 92: 1667-1682. http://dx.doi.Org/10.1175/2011BAMS3139.l
Dias. T; Clemente. A: Martins-Loucao. MA: Sheppard. L; Bobbink. R; Cruz. C. (2014).
Ammonium as a driving force of plant diversity and ecosystem functioning: observations
based on 5 years' manipulation of N dose and form in a Mediterranean ecosystem. PLoS
ONE 9: e92517. http://dx.doi.org/10.1371/iournal.pone.0Q92517
Doak. DF; Bigger. D; Harding. EK; Marvier. MA: O'Mallev. RE: Thomson. D. (1998). The
statistical inevitability of stability-diversity relationships in community ecology. Am Nat
151: 264-276. http://dx.doi.org/10.1086/286117
Donev. SC. (2010). The growing human footprint on coastal and open-ocean biogeochemistry
[Review], Science 328: 1512-1516. http://dx.doi.Org/10.l 126/science.l 185198
Driscoll. CT; Lawrence. GB; Bulger. AJ; Butler. TJ; Cronan. CS: Eagar. C: Lambert. KF;
Likens. GE; Stoddard. JL; Weathers. KC. (2001). Acidic deposition in the northeastern
United States: Sources and inputs, ecosystem effects, and management strategies.
Bioscience 51: 180-198. http://dx.doi.org/10.1641/00Q6-
3568(200 n05ir0180:ADITNU12.0.CQ:2
Duarte. N: Pardo. LH; Robin-Abbott. MJ. (2013). Susceptibility of forests in the northeastern
USA to nitrogen and sulfur deposition: Critical load exceedance and forest health. Water
Air Soil Pollut 224: 1355. http://dx.doi.org/10.1007/sll270-Q12-1355-6
Duffy. JE; Cardinale. BJ; France. KE; Mclntyre. PB; Thebault. E; Loreau. M. (2007). The
functional role of biodiversity in ecosystems: Incorporating trophic complexity. Ecol Lett
10: 522-538.
Fakhraei. H; Driscoll. CT: Renfro. JR: Kulp. MA: Blett. TF; Brewer. PF; Schwartz. JS.
(2016). Critical loads and exceedances for nitrogen and sulfur atmospheric deposition in
Great Smoky Mountains National Park, United States. Ecosphere 7: 1-28.
http://dx.doi.org/10.1002/ecs2.1466
Fakhraei. H: Driscoll. CT: Selvendiran. P: DePinto. JV: Bloomfield. J: Quinn. S: Rowell. HC.
(2014). Development of a total maximum daily load (TMDL) for acid-impaired lakes in
the Adirondack region of New York. Atmos Environ 95: 277-287.
http://dx.doi.Org/10.1016/i.atmosenv.2014.06.039
Fakhraei. HK: Driscoll. CT: Kulp. MA: Renfro. JR: Blett. TF: Brewer. PF: Schwartz. JS.
(2017a). Sensitivity and uncertainty analysis of PnET-BGC to inform the development of
Total Maximum Daily Loads (TMDLs) of acidity in the Great Smoky Mountains National
Park. Environ Modell Softw 95: 156-167. http://dx.doi.Org/10.1016/i.envsoft.2017.06.013
121
-------
Fakhraei. HK; Driscoll. CT; Kulp. MA; Renfro. JR; Blett TF; Brewer. PF; Schwartz. JS.
(2017b). Sensitivity and uncertainty analysis of PnET-BGC to inform the development of
Total Maximum Daily Loads (TMDLs) of acidity in the Great Smoky Mountains National
Park: Supplementary materials [Supplemental Data], Environ Modell Softw 95.
Fenn. ME; Allen. EB; Weiss. SB; Jovan. S; Geiser. LH; Tonnesen. GS; Johnson. RF; Rao.
LE; Gimeno. BS; Yuan. F; Meixner. T; Bytnerowicz. A. (2010). Nitrogen critical loads
and management alternatives for N-impacted ecosystems in California [Review], J Environ
Manage 91: 2404-2423. http://dx.doi.Org/10.1016/i.ienvman.2010.07.034
Fenn. ME; Driscoll. CT; Zhou. Q; Rao. LE; Meixner. T; Allen. EB; Yuan. F; Sullivan. TJ.
(2015). Use of Combined Biogeochemical Model Approaches and Empirical Data to
Assess Critical Loads of Nitrogen. In Environmental Pollution Series.
http://dx.doi.org/10.1007/978-94-017-95Q8-l 10
Fenn. ME; Jovan. S; Yuan. F; Geiser. L; Meixner. T; Gimeno. BS. (2008). Empirical and
simulated critical loads for nitrogen deposition in California mixed conifer forests. Environ
Pollut 155: 492-511. http://dx.doi.Org/10.1016/i.envpol.2008.03.019
Fenn. ME; Lambert. KF; Blett. TF; Burns. DA; Pardo. LH; Lovett. GM; Haeuber. RA; Evers.
DC; Driscoll. CT; Jefferies. DS. (2011). Setting limits: Using air pollution thresholds to
protect and restore U.S. ecosystems. (Issues in Ecology No. 14). Washington, DC:
Ecological Society of America, https://www.esa.org/esa/wp-
content/uploads/2013/03/issuesinecology 14.pdf
Frost. TM; Fischer. JM; Klug. JL; Arnott. SE; Montz. PK. (2006). Trajectories of zooplankton
recovery in the Little Rock Lake whole-lake acidification experiment. Ecol Appl 16: 353-
367. http://dx.doi.org/10.1890/04-180Q
Futter. MN; Klaminder. J; Lucas. RW; Laudon. H; Kohler. SJ. (2012). Uncertainty in silicate
mineral weathering rate estimates: Source partitioning and policy implications. Environ
Res Lett 7: 1-8. http://dx.doi.Org/10.1088/1748-9326/7/2/024025
Galloway. JN; Aber. JD; Erisman. JW; Seitzinger. SP; Howarth. RW; Cowling. EB; Cosby.
BJ. (2003). The nitrogen cascade. Bioscience 53: 341-356.
Galloway. JN; Cowling. EB. (2002). Reactive nitrogen and the world: 200 years of change
[Review], Ambio 31: 64-71. http://dx.doi.org/10.1639/0Q44-
7447(2002^)031 r0064:RNATWY12.0.CQ:2
Galloway, JN; Schlesinger, WH; Clark, CM; Grimm, NB; Jackson, RB; Law, BE; Thornton,
PE; Townsend, AR; Martin, R. (2014). Ch. 15: Biogeochemical cycles. In JM Melillo; TC
Richmond; GW Yohe (Eds.), Climate change impacts in the United States: The third
National Climate Assessment (pp. 350-368). Washington, DC: U.S. Global Change
Research Program. http://dx.doi.org/10.7930/J0X63JTQ
Gamfeldt, L; Lefcheck, JS; Byrnes, JEK; Cardinale, BJ; Duffy, JE; Griffin, JN. (2015).
Marine biodiversity and ecosystem functioning: what's known and what's next? Oikos 124:
252-265. http://dx.doi.org/10.111 l/oik.01549
122
-------
Gamfeldt L; Snail T; Bagchi, R; Jonsson. M; Gustafsson. L; Kiel lander. P; Ruiz-Jaen. MC;
Froberg, M; Stendahl, J; Philipson, CD; Mikusinski, G; Andersson, E; Westerlund, B;
Andren, H; Moberg, F; Moen, J: Bengtsson, J. (2013). Higher levels of multiple ecosystem
services are found in forests with more tree species. Nat Commun 4: 1340.
http://dx.doi.org/10.1038/ncomms2328
Geiser. LH; Jo van. SE; Glavich. DA; Porter. MK. (2010). Lichen-based critical loads for
atmospheric nitrogen deposition in Western Oregon and Washington Forests, USA.
Environ Pollut 158: 2412-2421. http://dx.doi.Org/10.1016/i.envpol.2010.04.001
Grantz. DA; Garner. JHB; Johnson. DW. (2003). Ecological effects of particulate matter.
Environ Int 29: 213-239. http://dx.doi.org/10.1016/S0160-4120(02^)00181-2
Greaver. TL; Clark. CM; Compton. JE; Vallano. D; Talhelm. AF; Weaver. CP; Band. LE;
Baron. JS; Davidson. EA; Tague. CL; Felker-Quinn. E; Lynch. JA; Herrick. JD; Liu. L;
Goodale. CL; Novak. KJ; Haeuber. RA. (2016). Key ecological responses to nitrogen are
altered by climate change. Nat Clim Chang 6: 836-843.
http://dx.doi.org/10.1038/NCLIMATE3088
Greaver. TL; Liu. L; Bobbink. R. (2011). Wetlands. In Assessment of nitrogen deposition
effects and empirical critical loads of nitrogen for ecoregions of the United States (General
Technical Report NRS-80) (pp. 193-208). Newtown Square, PA: U.S. Department of
Agriculture, Forest Service, Northern Research Station.
http://www.nrs.fs.fed.us/pubs/gtr/gtr-nrs-80chapters/17-greaver.pdf
Groffman. PM; Kareiva. P; Carter. S; Grimm. NB; Lawler. J; Mack. M; Matzek. V; Tallis. H.
(2014). Ch. 8: Ecosystems, biodiversity, and ecosystem service. In JM Melillo; TC
Richmond; GW Yohe (Eds.), Climate change impacts in the United States: The third
National Climate Assessment (pp. 195-219). Washington, DC: U.S. Global Change
Research Program. http://dx.doi.org/10.7930/J0TD9V7H
Hernandez, PL; Vallano, DM; Zavaleta, ES; Tzankova, Z; Pasari, JR; Weiss, S; Selmants,
PC; Morozumi, C. (2016). Nitrogen pollution is linked to US listed species declines.
Bioscience 66: 213-222. http://dx.doi.org/10.1093/biosci/biw003
Hesthagen, T; Fiske, P; Skielkvale, BL. (2008). Critical limits for acid neutralizing capacity
of brown trout (Salmo trutta) in Norwegian lakes differing in organic carbon
concentrations. Aquatic Ecology 42: 307-316. http://dx.doi.org/10.1007/slQ452-008-9191-
x
Hooper, DU; Adair, EC; Cardinale, BJ; Byrnes, JEK; Hungate, BA; Matulich, KL; Gonzalez,
A; Duffy, JE; Gamfeldt, L; O'Connor, MI. (2012). A global synthesis reveals biodiversity
loss as a major driver of ecosystem change. Nature 486: 105-U129.
http://dx.doi.org/10.1038/naturel 1118
Horn, KJ; Thomas, RO; Clark, CM; Pardo, LH; Fenn, ME; Lawrence, GB; Perakis, SS;
Smithwick, EAH; Baldwin, D; Braun, S; Nordin, A; Perry, CH; Phelan, JN; Schaberg, PG;
St. Clair, SB; Warbv, R; Watmough, S. (2018). Growth and survival relationships of 71
tree species with nitrogen and sulfur deposition across the conterminous. PLoS ONE 13:
Article #e0205296. http://dx.doi.org/10.1371/iournal.pone.0205296
123
-------
Howarth. RW. (2008a). Coastal nitrogen pollution: A review of sources and trends globally
and regionally. Harmful Algae 8: 14-20. http://dx.doi.Org/10.1016/i.hal.2008.08.015
Howarth. RW. (2008b). Estimating atmospheric nitrogen deposition in the Northeastern
United States: Relevance to Narragansett Bay. In A Desbonnet; BA Cost-Pierce (Eds.),
Science for ecosystem-based management: Narragansett Bay in the 21st century (pp. 47-
65). New York: Springer New York, http://dx.doi.org/10.1007/978-0-387-35299-2 3
IPCC (Intergovernmental Panel on Climate Change). (2007). Climate change 2007: Impacts,
adaptation and vulnerability. Cambridge, UK: Cambridge University Press.
https://www.ipcc.ch/publications and data/publications ipcc fourth assessment report
wg2 report impacts adaptation and vulnerabilitv.htm
Irvine. IC: Greaver. T; Phelan. J: Sabo. RD; Van Houtven. G. (2017). Terrestrial acidification
and ecosystem services: effects of acid rain on bunnies, baseball, and Christmas trees.
Ecosphere 8. http://dx.doi.org/10.1002/ecs2.1857
Isbell. F; Calcagno. V: Hector. A: Connolly. J: Harpole. WS: Reich. PB; Scherer-Lorenzen.
M; Schmid. B; Tilman. D; van Ruiiven. J: Weigelt. A: Wilsev. BJ; Zavaleta. ES: Loreau.
M. (2011). High plant diversity is needed to maintain ecosystem services. Nature 477: 199-
202. http://dx.doi.org/10.1038/nature 10282
Isbell. F; Craven. D; Connolly. J: Loreau. M; Schmid. B; Beierkuhnlein. C: Bezemer. TM;
Bonin. C: Bruelheide. H; de Luca. E; Ebeling. A: Griffin. JN; Guo. Q: Hautier. Y; Hector.
A: Jentsch. A: Krevling. J: Lanta. V: Manning. P; Meyer. ST: Mori. AS: Naeem. S:
Niklaus. PA: Pollev. HW: Reich. PB: Roscher. C: Seabloom. EW: Smith. MP: Thakur.
MP: Tilman. D; Tracy. BF; van Per Putten. WH: van Ruiiven. J: Weigelt. A: Weisser.
WW: Wilsev. B; Eisenhauer. N. (2015). Biodiversity increases the resistance of ecosystem
productivity to climate extremes. Nature 526: 574-U263.
http://dx.doi.org/10.1038/naturel5374
Johnson, PT; Ostfeld, RS; Keesing, F. (2015). Frontiers in research on biodiversity and
disease [Review], Ecol Lett 18: 1119-1133. http://dx.doi.org/10.Ill 1/ele. 12479
Johnson. PTJ; Townsend. AR; Cleveland. CC: Glibert. PM; Howarth. RW: McKenzie. VJ:
Reimankova. E: Ward. MH. (2010). Linking environmental nutrient enrichment and
disease emergence in humans and wildlife. Ecol Appl 20: 16-29.
http://dx.doi.Org/10.1890/08-0633.l
Kleijn. D: Bekker. RM; Bobbink. R: De Graaf. MCC: Roelofs. JGM. (2008). In search for
key biogeochemical factors affecting plant species persistence in heathland and acidic
grasslands: a comparison of common and rare species. J Appl Ecol 45: 680-687.
http://dx.doi.org/10.1111/i. 1365-2664.2007.01444.X
Lacoul. P; Freedman. B: Clair. T. (2011). Effects of acidification on aquatic biota in Atlantic
Canada. Environ Rev 19: 429-460. http://dx.doi.Org/10.l 139/al 1-016
Latimer. JS: Rego. SA. (2010). Empirical relationship between eelgrass extent and predicted
watershed-derived nitrogen loading for shallow New England estuaries. Estuar Coast Shelf
Sci 90: 231-240. http://dx.doi.Org/10.1016/i.ecss.2010.09.004
124
-------
Lawrence. GB; Dukett JE; Houck. N; Snyder. P; Capone. S. (2013). Increases in dissolved
organic carbon accelerate loss of toxic A1 in Adirondack lakes recovering from
acidification. Environ Sci Technol 47: 7095-7100. http://dx.doi.org/10.1021/es40Q4763
Li. H; McNultv. SG. (2007). Uncertainty analysis on simple mass balance model to calculate
critical loads for soil acidity. Environ Pollut 149: 315-326.
http://dx.doi.Org/10.1016/i.envpol.2007.05.014
Maestre. FT: Quero. JL; Gotelli. NJ: Escudero. A: Ochoa. V: Delgado-Baquerizo. M; Garcia-
Gomez. M; Bowker. MA: Soliveres. S: Escolar. C: Garcia-Palacios. P; Berdugo. M;
Valencia. E; Gozalo. B; Gallardo. A: Aguilera. L; Arredondo. T; Blones. J: Boeken. B;
Bran. D; Conceicao. AA; Cabrera. O: Chaieb. M; Derak. M; Eldridge. DJ: Espinosa. CI:
Florentino. A: Gaitan. J: Gatica. MG: Ghiloufi. W: Gomez-Gonzalez. S: Gutierrez. JR:
Hernandez. RM: Huang. X: Huber-Sannwald. E: Jankiu. M: Miriti. M: Monerris. J: Mau.
RL: Morici. E: Naseri. K: Ospina. A: Polo. V: Prina. A: Pucheta. E: Ramirez-Collantes.
DA: Romao. R: Tighe. M: Torres-Diaz. C: Val. J: Veiga. JP: Wang. D: Zaadv. E. (2012).
Plant species richness and ecosystem multifunctionality in global drylands. Science 335:
214-218. http://dx.doi.org/10.1126/science. 1215442
Mallev. DF: Chang. PSS. (1994). Assessing the health of a zooplankton community in a small
Precambrian Shield lake during recovery from experimental acidification. J Aquat Ecosys
Health 3: 273-286. http://dx.doi.org/10.1007/BF0Q115286
McCann. KS. (2000). The diversity-stability debate. Nature 405: 228-233.
http://dx.doi.org/10.1038/35012234
McCrackin. ML: Harrison. JA; Compton. JE. (2013). A comparison of NEWS and
SPARROW models to understand sources of nitrogen delivered to US coastal areas.
Biogeochemistry 114: 281-297. http://dx.doi.org/10.1007/slQ533-012-9809-x
Mcdonnell. TC: Belvazid. S: Sullivan. TJ; Bell. M; Clark. C: Blett. T; Evans. T; Cass. W:
Hyduke, A; Sverdrup, H. (2018a). Vegetation dynamics associated with changes in
atmospheric nitrogen deposition and climate in hardwood forests of Shenandoah and Great
Smoky Mountains National Parks, USA. Environ Pollut 237: 662-674.
http://dx.doi.Org/10.1016/i.envpol.2018.01.l 12
Mcdonnell, TC: Reinds, GJ; Sullivan, TJ: Clark, CM; Bonten, LTC; Mol-Dijkstra, JP:
Wamelink, GWW: Dovciak, M. (2018b). Feasibility of coupled empirical and dynamic
modeling to assess climate change and air pollution impacts on temperate forest vegetation
of the eastern United States. Environ Pollut 234: 902-914.
http://dx.doi.Org/10.1016/i.envpol.2017.12.002
McDonnell, TC: Sullivan, TJ: Hessburg, PF: Reynolds, KM: Povak, NA: Cosby, BJ: Jackson,
W: Salter, RB. (2014). Steady-state sulfur critical loads and exceedances for protection of
aquatic ecosystems in the U.S. Southern Appalachian Mountains. J Environ Manage 146:
407-419. http://dx.doi.Org/10.1016/i.ienvman.2014.07.019
McNultv, SG: Cohen, EC: Myers, JAM: Sullivan, TJ: Li, H. (2007). Estimates of critical acid
loads and exceedances for forest soils across the conterminous United States. Environ
Pollut 149: 281-292. http://dx.doi.Org/10.1016/i.envpol.2007.05.025
Naeem, S; Li, SB. (1997). Biodiversity enhances ecosystem reliability. Nature 390: 507-509.
125
-------
Nanus. L; McMurrav. JA; Clow. DW; Saros. JE; Blett T; Gurdak. JJ. (2017). Spatial
variation of atmospheric nitrogen deposition and critical loads for aquatic ecosystems in
the Greater Yellowstone Area. Environ Pollut 223: 644-656.
http://dx.doi.Org/10.1016/i.envpol.2017.01.077
NAPAP (National Acid Precipitation Assessment Program). (2011). National acid
precipitation assessment program report to congress 2011: An integrated assessment.
Washington, DC: U.S. Environmental Protection Agency, Clean Air Markets Division.
http://nv.water.usgs.gov/proiects/NAPAP/
Nilsson. J: Grennfelt. P. (1988). Critical loads for sulphur and nitrogen: Report from a
workshop held at Skokloster, Sweden, 19-24 March 1988. Skokloster, Sweden: Nordic
Council of Ministers: Copenhagen.
NRC (National Research Council). (2000). Clean coastal waters: Understanding and reducing
the effects of nutrient pollution. In Clean Coastal Waters: Understanding and Reducing the
Effects of Nutrient Pollution. Washington, DC: The National Academies Press.
http://dx.doi.org/10.17226/9812
O'Dea. CB; Anderson. S: Sullivan. T; Landers. D; Casey. CF. (2017). Impacts to ecosystem
services from aquatic acidification: using FEGS-CS to understand the impacts of air
pollution. Ecosphere 8. http://dx.doi.org/10.1002/ecs2.1807
Padgett. PE; Parry. SD; Bytnerowicz. A: Heath. RL. (2009). Image analysis of epicuticular
damage to foliage caused by dry deposition of the air pollutant nitric acid. J Environ Monit
11: 63-74. http://dx.doi.org/10.1039/b804875d
Pardo. LH; Fenn. ME: Goodale. CL; Geiser. LH; Driscoll. CT; Allen. EB; Baron. JS:
Bobbink. R; Bowman. WD: Clark. CM: Emmett. B; Gilliam. FS: Greaver. TL; Hall. SJ:
Lilleskov. EA; Liu. L; Lynch. JA: Nadelhoffer. KJ; Perakis. SS: Robin-Abbott. MJ;
Stoddard. JL; Weathers. KC: Dennis. RL. (201 la). Effects of nitrogen deposition and
empirical nitrogen critical loads for ecoregions of the United States. Ecol Appl 21: 3049-
3082. http://dx.doi.Org/10.1890/10-2341.l
Pardo. LH: Robin-Abbott. MJ: Driscoll. CT. (201 lb). Assessment of nitrogen deposition
effects and empirical critical loads of nitrogen for ecoregions of the United States. In
Assessment of nitrogen deposition effects and empirical critical loads of Nitrogen for
ecoregions of the United States. (NRS-80). Newtown Square, PA: U.S. Department of
Agriculture, Forest Service, Northern Research Station.
http://www.nrs.fs.fed.us/pubs/38109
Petraitis, PS: Latham, RE. (1999). The importance of scale in testing the origins of alternative
community states. Ecology 80: 429-442. http://dx.doi.org/10.1890/0Q12-
9658(1999)080r0429:TIOSIT12.0.CQ:2
Phelan. J: Belvazid. S: Jones. P; Caika. J: Buckley. J: Clark. C. (2016). Assessing the effects
of climate change and air pollution on soil properties and plant diversity in sugar maple-
beech-yellow birch hardwood forests in the northeastern United States: Model simulations
from 1900 to 2100. Water Air Soil Pollut 227: 84. http://dx.doi.org/10.1007/sll270-Q16-
2762-x
126
-------
Phelan. J; Belvazid. S; Kurz. D; Guthrie. S; Caika. J; Sverdrup. H; Waite. R. (2014).
Estimation of soil base cation weathering rates with the PROFILE model to determine
critical loads of acidity for forested ecosystems in Pennsylvania, USA: Pilot application of
a potential national methodology. Water Air Soil Pollut 225: 2109-2128.
http ://dx.doi .org/10.1007/s 11270-014-2109-4
Pinder. RW: Davidson. EA; Goodale. CL; Greaver. TL; Herrick. JD; Liu. L. (2012). Climate
change impacts of US reactive nitrogen. ProcNatl Acad Sci USA 109: 7671-7675.
http://dx.doi.org/10.1073/pnas. 1114243109
Povak. NA; Hessburg. PF; McDonnell. TC: Reynolds. KM: Sullivan. TJ; Salter. RB; Cosby.
BJ. (2014). Machine learning and linear regression models to predict catchment- level base
cation weathering rates across the southern Appalachian Mountain region, USA. Water
ResourRes 50: 2798-2814. http://dx.doi.org/10.1002/2013WRO14203
Reich. PB; Tilman. D; Isbell. F; Mueller. K; Hobbie. SE; Flynn. DF; Eisenhauer. N. (2012).
Impacts of biodiversity loss escalate through time as redundancy fades. Science 336: 589-
592. http://dx.doi.org/10.1126/science.1217909
Reinds. G: de Vries. W. (2010). Uncertainties in critical loads and target loads of sulphur and
nitrogen for European forests: Analysis and quantification. Sci Total Environ 408: 1960-
1970. http://dx.doi.Org/10.1016/i.scitotenv.2009.12.001
Rhodes. C: Bingham. A: Heard. AM: Hewitt. J: Lynch. J: Waite. R; Bell. MP. (2017).
Diatoms to human uses: linking nitrogen deposition, aquatic eutrophication, and ecosystem
services : Supplementary materials. Ecosphere 8.
Rockstrom. J: Steffen. W: Noone. K; Persson. A: Chapin. FS. Ill: Lambin. EF; Lenton. TM;
Scheffer. M; Folke. C: Schellnhuber. HJ; Nvkvist. B; de Wit. CA; Hughes. T; van Per
Leeuw. S: Rodhe. H; Sorlin. S: Snyder. PK; Costanza. R; Svedin. U; Falkenmark. M;
Karlberg. L; Corell. RW: Fabry. VJ; Hansen. J: Walker. B; Liverman. D; Richardson. K;
Crutzen, P; Foley, JA. (2009). A safe operating space for humanity. Nature 461: 472-475.
http://dx.doi.org/10.1038/461472a
Root. HT; Geiser. LH; Jovan. S: Neitlich. P. (2015). Epiphytic macrolichen indication of air
quality and climate in interior forested mountains of the Pacific Northwest, USA. Ecol
Indicat 53: 95-105. http://dx.doi.Org/10.1016/i.ecolind.2015.01.029
Scheffe. RD; Lynch. JA: Reff. A: Kelly. JT; Hubbell. B; Greaver. TL: Smith. JT. (2014). The
aquatic acidification index: A new regulatory metric linking atmospheric and
biogeochemical models to assess potential aquatic ecosystem recovery. Water Air Soil
Pollut 225: 1838. http://dx.doi.org/10.1007/sll270-013-1838-0
Schwede. DB; Dennis. RL; Bitz. M. (2009). The watershed deposition tool: A tool for
incorporating atmospheric deposition in water-quality analyses(l). J Am Water Resour
Assoc 45: 973-985. http://dx.doi.Org/10.l 111/i. 1752-1688.2009.00340.X
Schwede. DB: Lear. GG. (2014). A novel hybrid approach for estimating total deposition in
the United States - Supplemental information [Supplemental Data], Atmos Environ 92:
207-220. http://dx.doi.Org/10.1016/i.atmosenv.2014.04.008
127
-------
Simkin, SM; Allen. EB; Bowman. WD; Clark. CM; Belnap, J; Brooks. ML; Cade. BS;
Collins. SL; Geiser. LH; Gilliam. FS; Jovan. SE; Pardo. LH; Schulz. BK; Stevens. CJ;
Suding, KN; Throop. HL; Waller. DM. (2016). Conditional vulnerability of plant diversity
to atmospheric nitrogen deposition across the United States. Proc Natl Acad Sci USA 113:
4086-4091. http://dx.doi.org/10.1073/pnas.1515241113
Stoddard. JL. (1994). Long-term changes in watershed retention of nitrogen: Its causes and
aquatic consequences. In LA Baker (Ed.), Environmental Chemistry of Lakes and
Reservoirs (Advances in chemistry series no. 237 ed., pp. 223-284). Washington, DC:
American Chemical Society, http://dx.doi.org/10.1021/ba-1994-0237.ch008
Sullivan. TJ; Cosby. BJ; Driscoll. CT; McDonnell. TC; Herlihy. AT. (201 la). Target loads of
atmospheric sulfur deposition protect terrestrial resources in the Adirondack Mountains,
New York against biological impacts caused by soil acidification. Journal of
Environmental Studies and Sciences 1: 301-314. http://dx.doi.org/10.10Q7/sl3412-011-
0062-8
Sullivan. TJ; Cosby. BJ; Driscoll. CT; McDonnell. TC; Herlihy. AT; Burns. DA. (2012).
Target loads of atmospheric sulfur and nitrogen deposition for protection of acid sensitive
aquatic resources in the Adirondack Mountains, New York. Water Resour Res 48:
W01547. http://dx.doi.org/10.1029/201 lwrOl 1171
Sullivan. TJ; Cosby. BJ; Jackson. WA. (201 lb). Target loads of atmospheric sulfur deposition
for the protection and recovery of acid-sensitive streams in the Southern Blue Ridge
Province. J Environ Manage 92: 2953-2960.
http://dx.doi.Org/10.1016/i.ienvman.2011.07.014
Suter. GW; Norton. SB; Fairbrother. A. (2005). Individuals versus organisms versus
populations in the definition of ecological assessment endpoints. Integr Environ Assess
Manag 1: 397-400. http://dx.doi.org/10.1002/ieam.56300104Q9
Tilman, D. (2000). Causes, consequences and ethics of biodiversity. Nature 405: 208-211.
http://dx.doi.org/10.1038/35012217
Tilman. D; Downing. JA. (1994). Biodiversity and stability in grasslands. Nature 367: 363-
365. http://dx.doi.org/10.1038/367363a0
Tilman. D; Lehman. CL; Bristow. CE. (1998). Diversity-stability relationships: Statistical
inevitability or ecological consequence? Am Nat 151: 277-282.
Tilman, D; Reich, PB; Isbell, F. (2012). Biodiversity impacts ecosystem productivity as much
as resources, disturbance, or herbivory. Proc Natl Acad Sci USA 109: 10394-10397.
http://dx.doi.org/10.1073/pnas.12082401Q9
Tominaga, K; Aherne, J; Watmough, SA; Alveteg, M; Cosby, BJ; Driscoll, CT; Posch, M.
(2009). Voyage without constellation: Evaluating the performance of three uncalibrated
process-oriented models. 40: 261-272. http://dx.doi.org/10.2166/nh.2009.085
Turnev, CSM; Palmer, J; Maslin, MA; Hogg, A; Fogwill, CJ; Southon, J; Fenwick, P; Helle,
G; Wilmshurst, JM; McGlone, M; Bronk Ramsey, C; Thomas, Z; Lipson, M; Beaven, B;
Jones, RT; Andrews, O; Hua, 0. (2018). Global peak in atmospheric radiocarbon provides
a potential definition for the onset of the anthropocene epoch in 1965. Sci Rep 8: 3293.
http://dx.doi.org/10.1038/s41598-018-2097Q-5
128
-------
U.S. EPA (U.S. Environmental Protection Agency). (1993). Air quality criteria for oxides of
nitrogen (final, 1993) [EPA Report], (EPA/600/8-9l/049aF-cF). Washington, DC: U.S.
Environmental Protection Agency, Environmental Criteria and Assessment Office.
http://cfpub.epa. gov/ncea/cfm/recordisplav.cfm?deid=40179
U.S. EPA (U.S. Environmental Protection Agency). (1998). Guidelines for ecological risk
assessment [EPA Report], (EPA/630/R-95/002F). Washington, DC: U.S. Environmental
Protection Agency, Risk Assessment Forum, https://www.epa.gov/risk/guidelines-
ecological-risk-assessment
U.S. EPA (U.S. Environmental Protection Agency). (2004). Air quality criteria for particulate
matter [EPA Report], (EPA/600/P-99/002aF-bF). Research Triangle Park, NC: U.S.
Environmental Protection Agency, Office of Research and Development, National Center
for Environemntal Assessment- RTP Office.
http://cfpub.epa. gov/ncea/cfm/recordisplav.cfm?deid=87903
U.S. EPA (U.S. Environmental Protection Agency). (2006a). Multi-agency critical loads
workshop: sulfur & nitrogen deposition effects on freshwater and terrestrial ecosystems
Final report [EPA Report], (EPA 68-W-03-02). Washington, DC.
U.S. EPA (U.S. Environmental Protection Agency). (2006b). Wadeable streams assessment:
A collaborative survey of the nation's streams [EPA Report], (EPA 841-B-06-002).
Washington, DC: U.S. Environmental Protection Agency, Office of Research and
Development, Office of Water, http ://www.epa. gov/owow/streamsurvev
U.S. EPA (U.S. Environmental Protection Agency). (2008). Integrated science assessment for
oxides of nitrogen and sulfur: Ecological criteria [EPA Report], (EPA/600/R-08/082F).
Research Triangle Park, NC: U.S. Environmental Protection Agency, Office of Research
and Development, National Center for Environmental Assessment- RTP Division.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=201485
U.S. EPA (U.S. Environmental Protection Agency). (2009a). Integrated science assessment
for particulate matter [EPA Report], (EPA/600/R-08/139F). Research Triangle Park, NC:
U.S. Environmental Protection Agency, Office of Research and Development, National
Center for Environmental Assessment- RTP Division.
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=216546
U.S. EPA (U.S. Environmental Protection Agency). (2009b). National lakes assessment: A
collaborative survey of the nations lakes. (EPA/841/R-09/001). Washington, DC: U.S.
Environmental Protection Agency, Office of Research and Development, Office of Water.
https://www.epa.gov/sites/production/files/2013-l 1/documents/nla newlowres fullrpt.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2009c). Risk and exposure assessment
for review of the secondary National Ambient Air Quality Standards for oxides of nitrogen
and oxides of sulfur [EPA Report], (EPA/452/R-09/008A). Research Triangle Park, NC:
U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards,
Health and Environmental Impacts Division.
U.S. EPA (U.S. Environmental Protection Agency). (2011). Policy assessment for the review
of the secondary national ambient air quality standards for oxides of nitrogen and oxides of
sulfur [EPA Report], (EPA-452/R-1 l-004a). Washington, DC.
http://www3.epa.gov/ttn/naaqs/standards/no2so2sec/data/2011Q114pamain.pdf
129
-------
U.S. EPA (U.S. Environmental Protection Agency). (2013). Integrated science assessment for
lead [EPA Report], (EPA/600/R-10/075F). Research Triangle Park, NC: U.S.
Environmental Protection Agency, National Center for Environmental Assessment.
http: //cfpub. epa. gov/ncea/ cfm/recordi spl ay. cfm? dei d=255721
U.S. EPA (U.S. Environmental Protection Agency). (2015). Preamble to the integrated
science assessments [EPA Report], (EPA/600/R-15/067). Research Triangle Park, NC:
U.S. Environmental Protection Agency, Office of Research and Development, National
Center for Environmental Assessment, RTP Division.
https://cfpub.epa.gov/ncea/isa/recordisplav.cfm?deid=310244
U.S. EPA (U.S. Environmental Protection Agency). (2016a). Integrated review plan for the
national ambient air quality standards for particulate matter [EPA Report], (EPA-452/R-
16-005). Research Triangle Park, NC.
https://vosemite.epa.gOv/sab/sabproduct.nsf/0/eb862b233fbd0cde85257dda004fcb8clQpen
Document&TableRow=2.0
U.S. EPA (U.S. Environmental Protection Agency). (2016b). Integrated science assessment
(ISA) for sulfur oxides health criteria (Second external review draft) [EPA Report],
(EPA/600/R-16/351). Washington, DC.
https://cfpub.epa.gov/ncea/isa/recordisplav.cfm?deid=326450
U.S. EPA (U.S. Environmental Protection Agency). (2016c). Integrated science assessment
for oxides of nitrogen-health criteria (final report) [EPA Report], (EPA/600/R-15/068).
Research Triangle Park, NC: U.S. Environmental Protection Agency, Office of Research
and Development, National Center for Environmental Assessment.
http://ofmpub.epa.gov/eims/eimscomm.getfile?p download id=526855
U.S. EPA (U.S. Environmental Protection Agency). (2016d). National coastal condition
assessment 2010. (EPA 841-R-15-006). Washington, DC: U.S. Environmental Protection
Agency, Office of Water, Office of Research and Development.
https://www.epa.gov/national-aquatic-resource-survevs/ncca
U.S. EPA (U.S. Environmental Protection Agency). (2016e). National lakes assessment 2012:
A collaborative survey of lakes in the United States [EPA Report], (EPA 841-R-16-113).
Washington, D.C. https://nationallakesassessment.epa.gov/
U.S. EPA (U.S. Environmental Protection Agency). (2016f). National rivers and streams
assessment 2008-2009: A collaborative survey. (EPA/841/R-16/007). Washington, DC:
U.S. Environmental Protection Agency, Office of Water, Office of Research and
Development, https://www.epa.gov/national-aquatic-resource-survevs/national-rivers-and-
streams-assessment-2008-2009-report
U.S. EPA (U.S. Environmental Protection Agency). (2016g). National wetland condition
assessment 2011: A collaborative survey of the nation's wetlands (pp. 1-105). (EPA-843-
R-15-005). Washington, DC. https://www.epa.gov/national-aquatic-resource-
survevs/national-wetland-condition-assessment-2011 -report
U.S. EPA (U.S. Environmental Protection Agency). (2017a). Integrated review plan for the
secondary national ambient air quality standards for ecological effects of oxides of
nitrogen, oxides of sulfur and particulate matter [EPA Report], (EPA-452/R-17-002).
Research Triangle Park, NC.
130
-------
U.S. EPA (U.S. Environmental Protection Agency). (2017b). Northern Gulf of Mexico
hypoxic zone [Website], Washington, DC. Retrieved from https://www.epa.gov/ms-
htf/northern- gulf-mexi co-hypoxi c-zone
U.S. EPA (U.S. Environmental Protection Agency). (2019). U.S. EPA Integrated Science
Assessment for particulate matter [EPA Report], (EPA/600/R-19/188).
https://cfpub.epa.gOv/ncea/i sa/recordisplav.cfm?deid=347534
U.S. EPA (U.S. Environmental Protection Agency). (2020). 2017 National Emissions
Inventory complete release: technical support document, https://www.epa.gov/air-
emissions-inventories/2017-national-emissions-inventorv-nei-technical-support-document-
tsd
Walker. WE: Harremoes, P; Rotmans, J: Van Per Sluiis, JP; Van Asselt MBA: Janssen, P;
Kraver Von Krauss. MP. (2003). Defining uncertainty: a conceptual basis for uncertainty
management in model-based decision support. Integrated Assessment 4: 5-17.
http://dx.doi.Org/10.1076/iaii.4.l.5.16466
Whitfield. CJ: Phelan. JN; Buckley. J: Clark. CM: Guthrie. S: Lynch. JA. (2018). Estimating
base cation weathering rates in the USA: challenges of uncertain soil mineralogy and
specific surface area with applications of the profile model. Water Air Soil Pollut 229: 61.
http://dx.doi.org/10.1007/sll270-Q18-3691-7
Williams. JJ; Chung. SH; Johansen. AM: Lamb. BK; Vaughan. JK; Beutel. M. (2017a).
Evaluation of atmospheric nitrogen deposition model performance in the context of US
critical load assessments. Atmos Environ 150: 244-255.
http://dx.doi.Org/10.1016/i.atmosenv.2016.ll.051
Williams. JJ: Lynch. JA: Saros. JE; Labou. SG. (2017b). Critical loads of atmospheric N
deposition for phytoplankton nutrient limitation shifts in western U.S. mountain lakes.
Ecosphere 8: e01955. http://dx.doi.org/10.1002/ecs2.1955
Williams. MR: Filoso. S: Longstaff. BJ; Dennison. WC. (2010). Long-term trends of water
quality and biotic metrics in Chesapeake Bay: 1986 to 2008. Estuaries Coasts 33: 1279-
1299. http://dx.doi.org/10.1007/s 12237-010-9333-v
Worm. B: Barbier. EB: Beaumont. N: Duffy. JE: Folke. C: Halpern. BS: Jackson. JBC: Lotze.
HK: Micheli. F; Palumbi. SR: Sala. E: Selkoe. KA: Stachowicz. JJ: Watson. R. (2006).
Impacts of biodiversity loss on ocean ecosystem services. Science 314: 787-790.
http://dx.doi.org/10.1126/science.l 132294
Zavaleta. ES: Pasari. JR: Hulvev. KB: Tilman. GD. (2010). Sustaining multiple ecosystem
functions in grassland communities requires higher biodiversity. Proc Natl Acad Sci USA
107: 1443-1446. http://dx.doi.org/10.1073/pnas.09068291Q7
Zhang, Y; Chen, HYH; Reich, PB. (2012). Forest productivity increases with evenness,
species richness and trait variation: a global meta-analysis. J Ecol 100: 742-749.
http://dx.doi.org/10.1111/i. 1365-2745.2011.01944.x
131
-------