Improving Air Quality Ecosystem Modeling Connections to EPA Ord’s Ecosystem Research Program 8 October 2010


WHITE PAPER:

IMPROVING AIR QUALITY-ECOSYSTEM MODELING
CONNECTIONS TO EPA/ORD'S ECOSYSTEM RESEARCH

PROGRAM

(8 October 2010)

Ecosystem & Hydrologic Linkage Team of the Modeling and Analysis Divison

Atmospheric Modeling and Analysis Division
National Exposure Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711

1

-------
Air Quality - Ecosystem Modeling Connection
1. Introduction

As part of the EPA Office of Research and Development (ORD), the National Exposure
Research Laboratory (NERL) is in the process of developing Integrated Transdisciplinary
Research (ITR) programs to better address Problems of Broad, National Significance (PoBN).
These high priority topics would benefit from more integrated collaborative research
implementation across ORD's Laboratories and Centers. Energy and Climate, which includes
environmental impacts on ecosystems, and Sustainable Water, which includes water quantity and
quality impacts, have been identified as PoBNS's. NERL also considers Ecosystem Services
research to be of broad national significance. Ecosystem responses are integral to all three
problem areas. Just as there has been a movement away from assessing human exposure to air
pollutants one chemical species at a time towards an integrated one-atmosphere approach, so too
should there be an integrated one-atmosphere approach to assessing ecosystems exposure to air
pollutants. With this in mind, we propose that now is the time to start advancing from simply a
one-atmosphere to a one-biosphere approach that includes integration across multiple media and
biogeochemical processes to more effectively address ecological interactions with the
atmosphere as well as human systems. This overall vision will facilitate the air-surface
intersection of atmospheric stressors-to-ecological receptor pathways. This overall vision is
consistent with the EPA/NERL exposure science framework, as well as its emphasis on cross-
disciplinary integration, and the vision is also consistent with the EPA/ORD research emphasis
on Integrated Transdisciplinary Research.

This white paper addresses progress that can be made towards realizing this longer-term one-
biosphere vision during the next 5 years by (1) describing critical issues, (2) describing the key
air quality-exposure science questions relevant to the integrated assessment of air quality and
ecosystem health, (3) discussing EPA/NERL's research program that will build on past linkages
to improve exposure assessments through collaborative measurement campaigns, long-term
monitoring and cross-disciplinary air-quality model refinement and decision tool development,
and (4) outlining the research outcomes and products to be produced by this research program.
The following sections of this white paper describe our early efforts to link air quality and
ecosystem models and discusses key challenges and ways in which scientific and technical
expertise within AMAD, across NERL and across the scientific community can be accessed to
address these challenges.

Environmental Problem

A first step towards implementing a one biosphere paradigm is to better link air quality and
ecosystem models. This is a critical initial research activity because atmospheric wet and dry
deposition play an important role in terrestrial, freshwater aquatic and marine ecosystem
functioning and degradation (Lovett and Tear, 2008; Driscoll et al., 2007; Vitousek, et al., 1997).
For example, atmospheric deposition of sulfur and nitrogen is the primary source of acidifying
chemicals that can impact aquatic and terrestrial ecosystems (Dennis et al., 2007). Acidification
causes a cascade of effects that alter both terrestrial (DeHayes et al., 1999) and aquatic
ecosystems (Driscoll et al., 2001). These effects include slower growth, the loss of soil fertility

2

-------
through acidification, the injury or death of forest vegetation and localized extinction of fish and
other aquatic species. Atmospheric deposition is also an important source of excess nitrogen as a
nutrient that can affect aquatic and terrestrial systems (Galloway et al., 2003). Excess nitrogen
nutrient alters freshwater and terrestrial biodiversity, increases susceptibility of vegetation to
insects and diseases, alters surface water quality and contaminates drinking water supplies
(Driscoll et al., 2003). In the western U.S. microbial communities, such as lichen, are altered
and diminished with increased nitrogen deposition (Fenn et al., 1998 and 2003). In the Rocky
Mountains excess nitrogen deposition causes shifts in biodiversity and replacement of native
plants (Baron, et al., 2000; Bowman et al., 2006). Excess nutrients alter estuarine systems
through increased phytoplankton and algal productivity leading to eutrophication, loss of habitat,
loss of dissolved oxygen, fish kills and decreased productivity (Valigura et al, 2001; Paerl et al.,
2002). Nitrogen stressors from the atmosphere have been increasing, posing an increasingly
serious problem (Galloway and Cowling, 2002). Atmospheric deposition is the primary source of
mercury that is bio-accumulated in aquatic systems, affecting insects, birds and humans (Driscoll
et al., 2007; Rimmer et al., 2005). Mercury methylation requires sulfur reducing bacteria, thus
deposition of both sulfur and mercury is involved in mercury bio-accumulation (Jeremiason et
al., 2006). Additionally, atmospheric re-emission and advection can be a key transport pathway
for some pesticides (Bossi et al., 2008).

An important aspect of the exposures discussed above is the exploration of changing exposure
risk in light of future climate projections facilitated by the connection between air quality and
ecosystem function. Atmospheric processes such as precipitation, solar radiation and
temperature that drive pollutant deposition are also central drivers of the biogeochemical
processes to which an ecosystem responds. The change in these driving variables in response to
changing levels of greenhouse gases in the atmosphere will impact ecosystem functions and
provision of services in conjunction with atmospheric deposition (C. Driscoll and K. Stolte,
personal communication).

Past interactions and research with the scientists in the ecological community have produced
ground breaking work linking air and ecosystem models. Atmospheric nitrogen deposition
linkage of the Community Multiscale Air Quality (CMAQ) model with the Chesapeake Bay
Watershed model (Linker et al., 2000) has gone through several advancements and
improvements over the past two decades, revealing temporal and spatial scale issues that arise
when trying to communicate between models. For example, airsheds were developed as part of
this work showing that sources of air deposition extended well beyond the boundaries of the
recipient watershed, illustrating the large difference in spatial scales of analysis (Dennis, 1997;
Paerl et al., 2002). Nitrogen deposition simulations were also performed for Tampa Bay in
support of the TMDL (Total Maximum Daily Load) analyses, to advance an understanding of
how to address management of air quality to best reduce deposition to the watershed and total
loadings to Tampa Bay. Air deposition for modeled base case and futures scenarios were linked
with aquatic models for a critical loads analysis for Shenandoah National Park to examine the
extent of atmospheric deposition reductions needed to support recovery (Sullivan et al., 2008).
General analyses using acidic deposition futures were carried out, including the Shenandoah
study, to assess the progress of acid rain controls, the progress of ecosystem indicator recovery
as a result of controls that have been implemented and estimation of the amount of additional
deposition reductions required to attain a prescribed degree of recovery (Driscoll, et al., 2001 and

-------
2003). These analyses uncovered inconsistencies between current deposition estimates using data
and future deposition estimates using air quality model projections and the need for multi-year
atmospheric deposition model predictions. Ongoing collaborative work between AMAD and the
Ecosystem Research Division (ERD) is addressing the linkage of meteorological model
predictions of precipitation with watershed models in headwater watersheds to uncover and
assess linkage issues at the level of hydrology (Golden, et al., accepted). Discussions are
continuing to develop and refine recommendations for priority indicators or measures for climate
change studies (C. Driscoll and K. Stolte, personal communication).

From these collaborations, valuable experience on connecting atmospheric model outputs of
deposition with ecosystem models has been gained. From our experience, we can characterize
the linkage of air and ecosystem models in the following manner: Linkage of atmospheric
models to aquatic/watershed and terrestrial models is necessary, but this connection has been
more or less ad hoc, often uses incomplete deposition budgets, and suffers some major problems
and gaps. However, our experience has been sufficiently rich to lead to the identification of
advances that will significantly address linkage issues and critical gaps that, if overcome, will
improve cross-communication of air and ecosystem models.

Overall Research Goals

Over the next 5-years, AMAD's goal for air-ecosystem linkage is to:

• Produce exposure estimates of deposition at the spatial and temporal scales responsible
for the receptor response

This will involve the marriage of the local ecosystem scale (tens of km2) with the regional
airshed scale (thousands to millions of km2). The aim is (1) to provide support for SOx-NOx
welfare standard and critical load assessments and determinations, (2) to support ecosystem and
ecosystem services assessments coupled with air management drivers, and (3) to support the
assessment of the sustainability of ecosystem health and services under future climate.

Beyond the 5-year time frame, AMAD's goal for air-ecosystem research is to:

• Link deposition exposure models to receptor response models in an integrated,
transdisciplinary, one-biosphere manner for current and future climate conditions

This longer term goal supports the vision of advancing towards a one-biosphere perspective.
2. Science and Policy Drivers

The National Research Council (NRC) and the Clean Air Act Advisory Committee (CAAAC)
have urged EPA to pay increased attention to ecosystem protection and to develop its capacity in
this direction (NRC, 2004; CAAAC, 2005). In addition, the NRC and CAAAC recommended
that EPA explore the use of critical loads in the development of secondary National Ambient Air
Quality Standards. The NRC further noted that "concentration-based standards are inappropriate

-------
for some resources at risk from air pollutants, including soils, ground waters, surface waters and
coastal ecosystems. For such resources, a deposition-based standard would be more
appropriate." In response to the 2004 NRC report, EPA is considering secondary standards to
ensure ecosystem protection and is emphasizing their development. Additionally, the most
recent 2005 National Acid Precipitation and Assessment Program (NAPAP) report states "...
scientific studies indicate that the emission reductions achieved by Title IV of the Clean Air Act
are not sufficient to allow recovery of acid-sensitive ecosystems. Estimates from the literature of
the scope of additional emission reductions that are necessary in order to protect acid-sensitive
ecosystems range from approximately 40-80% beyond full implementation of Title IV."

EPA is now addressing the secondary standards to protect ecosystems separate from the primary
national ambient air quality standards, the first time they have been separated. In the near-term
the development and revision of secondary standards is focused on ozone damage from exposure
to air concentrations, and adverse effects of acidification, from sulfur (S) and nitrogen (N)
deposition, and nutrient enrichment from atmospheric deposition of N. The near-term priority
for addressing adverse effects from deposition is development of a combined secondary standard
for SOx and NOx air concentrations to address acidification since the total deposition of S plus N
is acidifying. After acidification, the next priority is to develop secondary standards addressing
nitrogen nutrient enrichment of aquatic, estuarine and terrestrial ecosystems. Because the CAA
stipulated the National Ambient Air Quality Standards (NAAQS) were for ambient air
concentrations, not deposition, the secondary standards must be set in terms of air concentrations
and the air quality model serves a critical function to connect air quality and deposition. CMAQ
is used to develop an Atmospheric Deposition Transformation Function for the NOx-SOx
secondary standard development. The role of ammonia in acidic deposition and nutrient
enrichment is expected to significantly increase as NOx and SOx controls related to O3 and PM2.5
are implemented under the current Clean Air Act and under potentially new welfare standards for
NOx and SOx, requiring increased attention in the future. Following the setting of new
secondary standards, program assessments will be expected to document whether emission
control policies are working as intended to achieve ecosystem protection, necessitating reliable
estimates of atmospheric deposition and creating a further need to connect air and ecosystem
models in a more integrated manner for these assessments.

The management concept of critical loads supports the development of secondary standards.
Following the NRC suggestion of exploring the use of critical loads, a multi-agency group
working on critical loads (associated with atmospheric deposition) as an approach to ecosystem
management for land management agencies has been established. The agencies are the NPS,
USFS, USGS and EPA. Additional work on critical loads is also being pursued as part of the
U.S.-Canada Air Quality Agreement. As with the NOx and SOx welfare standard development,
the critical load studies are concentrating on acidification of fresh water and terrestrial systems
and excess nutrients in fresh water and terrestrial systems (the latter mostly in the western US,
e.g., Geiser et al., 2010). Atmospheric deposition is the source of the critical load being
considered at this time. Thus, atmospheric deposition is central to the development of critical
loads and estimation of critical load exceedances as a basis for ecosystem management and
protection. These deposition estimates also need to be state-of-the-science.

-------
Part of NERL's mission is to conduct ecological exposure research for the EPA, to carry out its
mission to protect human health and the environment (NERL, 2009). Within this mission,

NERL has a goal of fostering Integrated Multi-Disciplinary Research (IMDR) which is
consistent with the recently articulated ORD goal of developing Integrated Transdisciplinary
Research (ITR). These research approaches call for ORD researchers to develop sustainable
solutions to environmental problems by engaging partners from multiple disciplines who
transcend traditional scientific disciplines throughout the research process. The air-ecosystem
linkage research will bring together researchers from at least three NERL Divisions as well as
other agencies (as noted in the collaboration section below) forming an important contribution to
the NERL goal of creating an integrated, multidisciplinary research program across its Divisions.
The research can be a central core, from the atmospheric side, for the development of ITR
capabilities for ecosystem research to develop sustainable solutions and for being a
multidisciplinary learning laboratory for EPA ORD.

The climate change research program at EPA is equally concerned about ecosystem response as
well as air quality and human health. This requires that changes in climate be expressed in ways
that can be translated to changes in ecosystem functioning and provision of services. It is critical
for climate studies that model-generated meteorology, air deposition and hydrology be internally
consistent. Otherwise, the climate signal from meteorological models provided to ecosystem
researchers will have an unknown degree of contamination, leading to biased assessments. This
calls for air-ecosystem research to advance research efforts to create that consistency.

Air-ecosystem linkages are important to the EPA/ORD's Ecosystem Services Research Program
(ESRP). ESRP has selected nitrogen as a priority pollutant for its research regarding ecosystem
exposure, sustainability and the enhancement or degradation in quantity and quality of ecosystem
services. The atmosphere is an important source of nitrogen. The linkage of air and watershed
models can be an important contribution to the place-based (geographically focused) studies
within the ESRP, particularly the Future Midwest Landscape and the Albemarle-Pamlico
Watershed studies. State-of-the-science estimates of continental atmospheric deposition are
required for the national mapping of nitrogen and ecosystem services research project. The
ESRP has as a major long-term research objective answering questions regarding ecosystem
sustainability under climate and land use change (change in stress).

Bioaccumulation of deposited Hg and pesticide exposure are important issues associated with
toxic pollutants. The proposed mercury regulations, intended to reduce mercury emissions from
coal fired power plants, are being revised and there is continued interest in the effects of mercury
on humans and on ecosystems. The adequate representation of mercury bi-directional exchange
has important implications for assessments of source responsibility, providing important answers
to current management questions.

Other agencies are also interested in improved modeling of surface exchange. For example, NH3
and N2O surface exchange are important to USDA. Attention is being paid to regulating NH3
emissions at both USDA and EPA because of the fine particle effects and deposition of nitrogen
effects. N2O emissions are of concern because N2O is a greenhouse gas affecting climate and it
is the single most important ozone depleting substance in the stratosphere.

-------
3. Key Linkage Issues

To gain insight on priority research directions, we interviewed several ecosystem modelers with
experience in using atmospheric deposition data in their modeling and critical load analyses.
These modelers were asked to identify what they believed were the critical needs to improve the
linkage between air and ecosystem models and to overcome some of the critical gaps.

Combining their input with our experience and with perspectives from other modeling
communities has led us to identify the following list of research issues:

• The paradigm gap

• Spatial and temporal scale issues

• Improved and more complete total deposition estimates

• Precipitation and hydrology issues

Many of the research needs associated with these issues are discussed in Seigneur and Dennis (in
press) as part of the NARSTO assessment of multi-pollutant air quality management. Regional
air quality models would appear to provide the best, universal approach and spatial coverage for
linkage with ecosystem models. They will be more acceptable to the ecosystem modeling
scientists if many of these issues can be addressed or reduced; however, they most likely cannot
be eliminated. The paradigm gap is fundamental and provides an overarching context for setting
the stage to define and address the other three specific research issues.

The Paradigm Gap

There is a major paradigm gap that thwarts easy linkage of atmospheric deposition predicted by
regional atmospheric models to ecosystem models. This paradigm gap affects ecosystem studies
and especially affects climate change studies. This gap stems from a fundamental lack of
sufficient universally applicable ecosystem process information and too many degrees of
freedom regarding the behavior of the system to perform the modeling from a universal
applicability perspective. This limitation leads to a heavy reliance on calibrated ecosystem
models. While calibrated models perform well for the particular system and historical input data
(scenarios) for which they are calibrated, the models are not easily generalizable to other
ecosystem locations and other inputs and errors in input data may invalidate the model.
Atmospheric models, on the other hand, work from a perspective of universal applicability (first
principles) so that the same model parameterizations can apply everywhere and at anytime and
the models can be applied for multiple input data sets and they do not require calibration or long
spin-up times. However, the downside of this approach is that outputs of the atmospheric
models have error, at times significant error, relative to observations. Of particular concern are
errors in the location, amount and timing of precipitation and in the associated water balance and
surface water hydrology. These errors typically present challenges in linking with the ecosystem
models because the atmospheric modeling system predictions do not match the conditions used
to calibrate the ecosystem model. The paradigm gap is expected to introduce potentially serious
errors in climate change analyses. In these analyses, an inconsistency is created by mixing the
calibration base case with the futures prediction from the atmospheric model; the base case used
by the ecosystem models is not consistent with the hydrologic balance of the base or the futures

-------
case coming from the atmospheric model. As a consequence, an artificial degree of change, a
bias, can be introduced into the analysis of change which is particularly problematic for the
assessment of climate change impacts. The following three issues should be viewed in the
context of the paradigm gap because it structures our research response to them. The paradigm
gap particularly underlies the precipitation and hydrology issue. Although many aspects of the
following issues will be addressed over time, the paradigm gap will not soon disappear.

Spatial and temporal scale issues

Since the ecosystem responses to stressors are controlled by local conditions and local exposures,
ecosystem models require very fine spatial resolution to characterize the hydrology and
distinguish deposition to small catchments and areas with varying topography and geology.
However, the atmospheric models that are needed to provide a complete characterization of the
atmospheric emissions and chemistry operate at a coarser level. A challenge of air-ecosystem
linkage is the marriage of these two very disparate scales. Management strategies used in
ecological studies need to recognize that airsheds for the local deposition impacting ecosystems
are large and regional in extent (see Paerl et al., 2002). It is known that differences in dry
deposition arise as a function of different types of land cover. Deposition to specific vegetation
species or land cover types is needed for input to the ecosystem models while typical regional air
quality models provide one dry deposition value per grid cell. Differences in deposition to
different vegetation types can be addressed at the sub-grid level, however, the land cover
specifications need to be consistent. Land use specification is important to both ecological and
atmospheric models, but they have not historically been consistent. Therefore, it will be
important to update and harmonize the land use data in CMAQ with those routinely used by the
ecological community. Using this same finer scale base data and aggregating to regional data
will provide an important consistency between the models. The capability to vary land use to
study the effects of land use change is an investigatory area of great interest to the EPA
Ecosystem Services Research Program and the greater ecological modeling community. In areas
of complex terrain dry deposition can vary significantly due to differences in terrain influences
on meteorology and air concentrations. Empirical studies have examined the range of deposition
values in complex terrain that might be expected within a grid and underline the need for local,
spatial detail in deposition estimates for critical load studies (Weathers et al., 1995 and 2006).
Developing these data from regional model output presents a challenge. Terrain influences may
be able to be partially addressed by development of empirical surrogate relationships of dry
deposition variation with terrain features that are available from GIS information and maps.

Such semi-empirical sub-grid models of dry deposition variability could further reduce the dry
deposition uncertainty in complex terrain and would be of great interest to the ecosystem
modelers.

The temporal scales of interest in ecosystem models can vary from daily to annual time scales
and may be sensitive to short-lived episodic events (e.g., flooding events, intermittent toxic
releases, and pesticide applications), while air models operate hourly and are often aggregated up
to longer averaging times. Also of issue is the need for multiple years of atmospheric deposition
for establishment of the base case because the ecosystem models are typically calibrated using
several years of data. In addition, several ecosystem models require 150 years for a spin up

-------
period, starting with pre-industrial conditions and some climatic events of interest such as
droughts occur over prolonged time periods, e.g., months to decades.

Improved and more complete total deposition

Ecosystem analyses need to include total deposition (both wet and dry) for a more accurate
representation of the deposition budget. For greater credibility, wet deposition estimates should
be improved to address biases and known missing processes. A summertime under-prediction
bias is hypothesized to be associated with the omission in CMAQ of a parameterization for the
production of NOx aloft due to lightning. Two missing processes of concern to the ecological
community are cloud impaction and dissolved organic nitrogen (DON). Cloud-water impaction
at high elevation is an important source of deposition in complex terrain affecting critical load
calculations. DON is estimated to constitute approximately 25% of the total wet nitrogen
deposition. DON from the atmosphere is now thought to be an important source of nitrogen for
ecosystem exposure and is considered to be a missing process in air deposition models that
requires attention. Other chemical budgets important to ecosystem biogeochemical processes, in
particular base cations, are not represented in the atmospheric models for either wet or dry
deposition but should be addressed in future model development. Dry deposition is an important
part of total deposition yet this component is typically missing from many ecosystem analyses
because of the sparseness and incompleteness of measured dry deposition data. Dry deposition's
magnitude is often equal to or greater than that of wet deposition. Since atmospheric models
include a more complete characterization of the chemical species than most monitoring
networks, using the dry deposition from atmospheric models provides ecosystem models with a
more complete accounting of the pollutant budget. The importance of an air quality model for
dry deposition estimates creates a responsibility and need to reduce the uncertainty in the model
estimates.

Reducing uncertainty in the dry deposition algorithms especially involves keeping them state of
the science. This includes incorporation of bi-directional air-surface exchange where it is critical
and evaluating dry deposition algorithms with new measurements. It is well established that air-
surface exchange is bi-directional for NH3 (Walker et al., 2006) and Hg, two very important
species involved in ecosystem effects. Other species, for example semi-volatile pesticides, may
also need to be considered in the future. Formulations for bi-directional air-surface exchange
with terrestrial and aquatic surfaces are required in CMAQ for scientific credibility (keeping it
state of the science). Incorporation of bi-directional exchange is expected to significantly
influence the estimated range of influence of a source and the estimation of source attribution as
well as the wet and dry partitioning of total NH3 deposition (Dennis et al., in press). Second, the
evaluation of dry deposition algorithms has received limited attention due, in part, to lack a of
measurement methods capable of supporting field study investigations of dry deposition. The
importance of the sulfur, nitrogen and mercury species to deposition and ecosystem effects is
now well accepted. Advances in instrument sensitivity and sampling frequency for key species
have occurred that should create new opportunities for evaluation of deposition algorithms. The
implementation of these advances in collaborative field studies should be encouraged to develop
further credibility in CMAQ's dry deposition estimates.

-------
Precipitation and hydrology issues

Errors in the meteorological model predictions of precipitation influence estimates of wet
deposition, introducing errors in them. It is a real concern that the meteorological model
predictions of precipitation location, amount, and duration have levels of error such that the
predicted precipitation is inconsistent with the hydrologic response in the real world. Ecosystem
models are calibrated on observed precipitation and hydrology to develop a water balance.

Given the calibrated nature of the ecosystem models, observed precipitation and wet deposition
are the preferred inputs for the ecosystem models because of the large errors in the prognostic
model precipitation estimates. Unfortunately, this produces inconsistent total chemical budgets
across models, due to the paradigm gap, when an air model is used to provide deposition
estimates. It is important to remove/address this inconsistency (bridge the gap) to establish
successful ecological and air quality model linkage. This suggests a necessity to link hydrology
to the meteorological model's precipitation predictions to create an internally consistent data set
of precipitation and hydrology as one of the research directions. It is also critical to improve our
ability to simulate precipitation at the fine scale to improve the linkage with the hydrologic
models and to improve the accuracy of the wet deposition estimates. Additionally, simulated
precipitation and temperature improvements to account for orographic effects are needed in
complex terrain and mountainous settings which are home to many sensitive ecosystems in the
U.S. This is a challenging issue and the level of improvement needed may need to be
established by a risk assessment of the sensitivity of the receptor of interest to the variety of
precipitation-driven stressors (e.g., exposure risk to precipitation inputs), rather than metrics
associated with comparisons against observed precipitation data.

4. Key Science Questions

Science Questions

The issues in the previous section need to be addressed to achieve the research goals for air-
ecosystem linkage articulated above. Science questions associated with the issues are posed to
help formulate components of the research. The science questions are:

•	What are the sources of deposition bias and how can they be resolved?

•	How can air-ecosystem spatial and temporal scale mismatches be addressed or resolved?

•	What atmospheric processes are missing or not sufficiently state-of-the-science and what
is needed to incorporate or improve them to be state-of-the-science?

•	What are the sources of the paradigm gap between CMAQ, a universal, generalized
parameter model, and calibrated ecosystem/watershed models? What approaches will
bridge the paradigm gap?

•	How will climate and land use change impact atmospheric composition, biogeochemical
cycling and ecosystem health?

10

-------
Related Management Questions that Create Science Questions

•	Which sources of atmospheric emissions contribute to terrestrial and aquatic ecosystem
degradation?

o Where is the deposition coming from; where is it going to?

¦	How large and where are the airsheds? What is the range of influence?
o Who is the deposition coming from?

¦	Economic sectors

¦	Geographic regions

•	What is the accuracy of the deposition estimates?

•	What is the "policy-relevant" background for nitrogen deposition?

5. Research Directions and Approach

To answer these science questions requires a coordinated research strategy covering a variety of
activities and tasks to best use AMAD's limited resources. The strategy involves: (1)
Improvements in CMAQ's bi-directional and uni-directional air-surface exchange
characterizations, evaluation of these improvements, and quantification of the uncertainty of
these estimates; (2) Harmonization of land cover with ecosystem models and improvements in
the capability and flexibility of the WRF/CMAQ land surface characterizations and air-surface
exchange in complex terrain; (3) Incorporation of missing deposition pathways into CMAQ; (4)
Linkage of hydrologic models with WRF/CMAQ to provide a consistent connection between
precipitation and hydrology for a system-consistent linkage to ecosystem models and
development of improved WRF precipitation fields; (5) Research on close coupling of the
WRF/CMAQ/hydrology system with watershed models to test fidelity regarding biogeochemical
cycling; (6) Application of expanded model capability to simulate ambient air-quality and
ecosystem exposure to support assessments addressing sensitive ecosystems and ecosystem
services and assessments of scenarios depicting climatic and land use change, particularly
climate change. The applications may involve post-processing CMAQ deposition data for
critical loads assessments and error analysis.

(1) Further Develop CMAQ Air-Surface Exchange

Directions: Improve the parameterizations of the air-surface exchange of atmospheric pollutants,
both bi-directional and uni-directional for aquatic and terrestrial landscapes. Improvements will
target vegetated canopy atmospheric resistances and the relative partitioning between deposition
to physiologically mediated leaf tissue and vegetation and soil surfaces. Develop techniques to
account for deposition of pollutants at high elevations via impaction of cloud water.

Approach: CMAQ currently estimates bi-directional exchanges ofNH3 over terrestrial surfaces
and of Hg over terrestrial and aquatic land surfaces. This is a first for a 3-dimensional regional
air quality model. Deposition and emission fluxes are simultaneously estimated for NH3 based
on the two layer canopy compensation point resistance model of Nemitz et al. (2001) and Hg
following the two layer canopy compensation point resistance-capacitance model of Sutton et al.
(1998). The ambient concentration at which the net air surface exchange is zero, i.e., the
compensation point, in both the NH3 and Hg surface exchange models is dynamic and solved

11

-------
from bi-directional exchange estimates from air-soil, -vegetation surface, and -stomatal
interfaces. Quantification of the deposition to the canopy and soil components with different
physical and chemical resistances provides insights that can improve the net uni-directional
deposition estimate and the net bi-directional exchange flux. The prototype NH3 bi-directional
exchange model uses a priori specified emission potentials for soil, vegetation surfaces, and
stomatal exchange taken from the literature, while the Hg bi-directional exchange model
parameterizes dynamic compensation points using partitioning coefficients and first order
reduction of previously deposited divalent mercury in the surface media published in the
literature. The NH3 bidirectional exchange model will be refined to use dynamic fertilizer
application rate inputs to be able to derive in situ emission potentials from fertilized agricultural
soils. This will be accomplished by development of a Fertilizer Emissions Scenario Tool for
CMAQ (FEST-C) based on the USD A Environmental Policy Integrated Climate (EPIC) model
in collaboration with the USDA. This tool will become part of the CMAQ input processing
system. FEST-C will first treat chemical fertilizer application, but will then be extended to deal
with manure application, dependent on having the relevant driving information. The NH3
bidirectional exchange model will also be expanded to include bi-directional exchange over
aquatic ecosystems. The Hg bidirectional model currently assumes that vegetation is a sink for
divalent mercury species due to the lack of a published reduction mechanism for divalent
mercury on vegetation surfaces and uses a simple single layer water surface model to estimate
air-surface water exchange. Bi-directional Hg exchange model algorithms will be refined to
include the reduction of divalent mercury species on vegetation surfaces and a multiple
compartment parameterization of surface waters. Furthermore, other pollutants, most notably
persistent organic pollutants (POPs), exhibit bi-directional exchange that can significantly
enhance the transportation and ecosystem exposure. Future research and collaborations will
expand the bi-directional model to include POPs and other semi-volatile pollutants that exhibit
long range transport and are responsible for deleterious effects on ecosystem health.

The parameterization of bi-directional surface exchange requires more detailed
vegetation and soil parameters and field data to guide the parameterizations. AMAD model
development and evaluation teams will engage in collaborative development work with other
EPA laboratories, such as NRMRL, and federal and academic research institutions to advance
surface exchange parameterizations. AMAD will partner with process-oriented scientists, at EPA
and elsewhere, engaged in laboratory and in situ field measurements, via collaborations to gather
data needed to advance in-house air-surface exchange, bi- and uni-directional, model
development and evaluation. Top down and bottom up inverse modeling techniques using
satellite and special field observations will be used to assess the sensitivities of the newly
developed model air-surface exchange parameterizations and to identify sensitive model
variables and variables that are not routinely measured.

(2) Harmonize Land Surface Characterizations and Address Subgrid Variability of Air-Surface
Exchange

Directions: Advance and modernize the WRF/CMAQ land surface parameterizations and leaf
area index (LAI) estimates to improve connections or linkage to ecosystem models, to address
paradigm mismatches and develop community confidence in the system. Develop sub grid scale
land use specific deposition estimates, and improve meteorological and dry deposition
simulations in complex terrain at higher spatial resolutions. Develop parameterizations of sub
grid variation in dry deposition due to complex terrain.

12

-------
Approach: CMAQ, MM5, and WRF have historically used 1 km land use data from the 1992
National Land Cover Database (NLCD) with categories defined using the USGS 24 category
system. An updated version of the NLCD is now available which represents land cover from
2001 with a resolution of 30 m. This new data set is being incorporated into WRF and CMAQ
along with MODIS data for Mexico and Canada. The native NLCD and MODIS land use
categories will be retained as this is consistent with the categories used by ecosystem models and
will facilitate linkages between the air and ecosystem models. Additional capabilities will be
developed to allow modification of the land use designations to explore future scenarios of land
use change and their impact on air quality and air-surface exchange.

Currently, CMAQ output consists of grid-averaged deposition velocities and fluxes.
Deposition depends greatly on the characteristics of the underlying surface and therefore will
vary with land cover type. Ecological applications often require knowledge of the deposition to
individual land cover types within a grid rather than the grid-averaged value. CMAQ is being
modified to calculate and output sub-grid scale land-use specific deposition estimates (Mosaic
approach) which will provide the information needed for ecosystem assessments. Making these
changes within CMAQ rather than the meteorological model allows the use of previously
generated meteorological files rather than requiring the meteorological model to be rerun.

Leaf area index (LAI) is an important input to the land surface model in WRF and to the
deposition algorithms in CMAQ. The current method for obtaining LAI for WRF and CMAQ
uses the deep soil temperature for predicting leafout and crop growth. Fixed land use category
maximum and minimum values for LAI are provided in the models. Alternative sources such as
the EPIC model may provide similar input for alternative land use and climate scenarios.
Advances in remote sensing technology provide an opportunity for developing techniques for
using satellite data to obtain spatially and temporally explicit estimates of LAI for input to the
models. Collaboration with ESD's Landscape Characterization Branch will be pursued to further
this approach.

Deposition in complex terrain is poorly understood and not often measured.

Consequently, little air quality model development has occurred in this area. However,
numerous sensitive ecosystems occur in complex terrain and current deposition estimates to
these areas are likely inadequate to provide the information needed for use in, for example,
critical loads deposition assessments. While WRF and CMAQ model long term development
will work to improve meteorological and surface exchange capabilities in complex terrain at
higher spatial resolutions, there is a need to find alternative means to improve these estimates for
current deposition assessments. AMAD will need to partner with ecosystem scientists to collect
sub grid-scale deposition or air concentration data in complex terrain for sulfur, nitrogen and
mercury species to empirically tie within-grid dry deposition variability with GIS-available
metrics.

(3) Address Missing Deposition Processes

Directions: Incorporate processes or pathways that are currently known to be missing in CMAQ
that have an important effect on deposition or are highly desired by the ecosystem modeling
community. Interpret model evaluation results and error analysis in terms of missing processes
to further identify candidates related to missing processes, pathways or emissions, as distinct
from spatial issues of emissions accuracy.

Approach: Recent model evaluation analyses have identified the lack of a lightning NOx
generation processes in CMAQ as a source of a major summer under-prediction bias in wet

-------
nitrate deposition and NOx concentrations aloft in the free troposphere. Collaborative research
with NASA is culminating in a parameterization that can be driven by WRF to include lightning-
generated NOx in CMAQ.

Cloud impaction is an important source of wet deposition in mountainous, complex
terrain that is important to terrestrial critical load modelers, but it is not computed by CMAQ.
CMAQ can be modified to include a model of cloud impaction. The initial approach will be to
incorporate a simple model, such as that described in Katata, et al. (2008), to meet the needs of
the critical load modelers without creating a large computational burden in CMAQ.

Base cation deposition, mainly from soil emissions, is highly desirable for acidification
calculations for aquatic systems to help complete the chemical balance driving these systems, but
CMAQ does not track these emissions at this time. New work to help PM2.5 non-attainment
modeling is characterizing and identifying the species involved in soil emissions. When it is
completed under the EPA Air Program, this work will allow CMAQ to provide ecosystem
modelers with base cation deposition estimates.

Dissolved organic nitrogen (DON) deposition is an important source of nitrogen through
wet deposition that is currently not incorporated in current air deposition models. The difficulty
is that the chemical constituents are many and not well known, coupled with a lack of
understanding of where they are coming from. This means that we do not currently know how to
characterize emissions of DON species. Collaborations with EPA/NRMRL and with the
National Atmospheric Deposition Program will provide empirical data on the amounts of DON
depositing across the country under different mixes of land use. These data will be assessed to
provide more information to guide the next steps for developing interim estimates of DON
deposition and incorporating DON into CMAQ.

(4) Link Hydrological Processes and Models to WRF/CMAQ and Improve Precipitation Fields
Directions: Extend the WRF meteorological model to provide an internally consistent
representation of the hydrologic cycle, including precipitation, soil moisture, evapotranspiration
and overland flow. Focus on the elements of the hydrological cycle that impact ecosystem
loading and exposure of atmospheric pollutants. Provide an internally consistent representation
of precipitation and overland flow. Also, focus on the elements of the hydrological cycle that
affect the response of ecosystems to climate change. Develop a linkage between climate-model-
produced-precipitation and associated modeled hydrological processes to address paradigm
mismatches and support ecosystem exposure scenario estimates under climate change.

Improve the modeling of amount, location, and duration of precipitation by WRF to
reduce the error in wet deposition estimates from CMAQ. Extend the WRF precipitation
predictions to finer grid scales, such as 4 km, with a particular emphasis on regions with complex
terrain. Further develop techniques for accounting for occult deposition, particularly cloud-water
deposition in mountainous terrain.

Approach: Develop extensive collaboration with other research groups within EPA, other
federal agencies and academia to extend WRF/CMAQ capabilities to include hydrology
consistently linked to WRF. AMAD (ecosystem and climate teams) would develop the linkage
for retrospective WRF and down-scaled WRF climate simulations with the intent of providing
national coverage. Primary groups for collaboration are the NOAA Hydrology Laboratory of the
National Weather Service (NWS), the US Geological Survey (USGS) and Pacific Northwest
National Laboratory (PNNL). AMAD/EPA will sponsor a workshop co-sponsored with USGS
and NOAA to develop a sound research direction for this work and develop a conceptual plan to

-------
guide the research. A major consideration is how tightly the hydrology should be coupled with
the meteorological model's internal water budget in order to provide some degree of feedback or
whether the hydrology should be purely calculated off-line as a post-processing step. Attention
will also have to be paid to within-grid heterogeneity of the hydrologic processes affecting
evapotranspiration. Land surface models for WRF to be considered include the NOAH Land
Surface model, the Community Land Cover Model (CLM4) model and the Variable Infiltration
Capacity Model (VICM). Hydrology models to consider for adaptation in this effort could be the
NOAA/NWS Hydrology Laboratory Research Distributed Hydrologic Model (HLRDHM),
which is operational at 1 km grid size and the USGS Precipitation Runoff Modeling System
(PRMS). The connection would be targeted for WRF with a 12 km continental domain and 4 km
regional test domains. In conjunction with recommendations from the planning workshop,
AMAD would consult with NOAA and USGS on approaches to establish the degree of error
introduced by the use of larger WRF grid sizes. The NOAA Earth System Research Laboratory
Hydrometeorological Testbed (HMT) that is planned to start in 2010 in the Neuse Basin in North
Carolina will collect high temporal and spatial precipitation and hydrological data that will be a
potential source of test data for the WRF-hydrology linkage. The ecosystem and climate teams
within AMAD will coordinate closely on working with the down-scaled meteorology to produce
consistent hydrology for climate change studies. AMAD and ERD will coordinate closely on
evaluating the influence of the internally consistent hydrology predictions on watershed model
water balance and chemical biogeochemical response predictions. This research is a major
undertaking that will coordinate with the larger meteorological and hydrology community.

Improvement in WRF precipitation simulation for linkage of air quality to ecosystem and
watershed models will arise from improved/expanded data assimilation, improved boundary
condition definition and improved model physics, in addition to advances in land surface
modeling stemming from the above work to link hydrology with WRF. Advances in these areas
will come from a combination of in-house AMAD research and collaborative research with the
WRF community. In-house scientists have incorporated an "obs-grid" data assimilation
approach, to good effect, and are expanding assimilation to include soil moisture (e.g., Pleim and
Gilliam, 2009) which will improve surface heat and moisture flux simulation from soil and
vegetation. Long-term research by UNC scientists is yielding promising results in expanding the
application of 3-D variational techniques to include, for example, radar data assimilation and
scientists at the University of Alabama are exploring the assimilation of satellite data to improve
cloud characterization.

Recent WRF simulations make use of advanced data products such as the 12-km North
American Model (NAM) analysis to provide model boundary conditions (Appel, et al, 2009), but
these data are not available prior to 2005. In-house research is exploring the use of alternative,
higher resolution reanalysis data such as the North American Regional Reanalysis (NARR). The
precipitation errors in complex terrain at different grid resolutions will be quantified using the
Parameter-elevation Regressions on Independent Slopes Model (PRISM) data set. AMAD will
continue to explore the value of the radar-based precipitation estimates in complex terrain for
establishing ground truth. Tests will compare performance at 12 km and 4 km grid sizes.
Collaboration will be developed with researchers interested in improving WRF predictions in
complex terrain, including precipitation predictions, such as researchers at the University of
Washington. The University of Washington research is also examining WRF simulations at 4 km
grid sizes. Improvements derived from these enhancements will be evaluated with scientists at
ERD and Syracuse University in light of improved hydrologic model performance.

-------
Research regarding the incorporation of improved model physics will follow the same
successful in-house/external collaboration design. For example, in-house scientists have
successfully implemented the Asymmetric Convective Model 2 (ACM2; Pleim, 2007a, b), but
further refinements are being explored through collaboration with scientists at Texas A & M
regarding the use of Kalman filtering techniques to refine ACM2 parameter values.

(5) Explore Close Coupling of WRF/CMAQ/Hydrolosv and Watershed Models and Identify One-
Biosphere Linkage Issues

Directions: Assess the performance of watershed models using internally consistent
precipitation and hydrology input from the WRF/CMAQ system as compared to using observed
precipitation and hydrology. Assess the fidelity or robustness of ecosystem biogeochemical
response when internally consistent precipitation and hydrology is provided to bridge the
paradigm gap between air and watershed models. Do these data bridge the gap or do they
introduce a lack of robustness of response to stressor change? Address the fidelity and
robustness of assessing ecosystem response to climate change with linked air-watershed
biogeochemical models with exposure inputs based on a down-scaled WRF/CMAQ/Hydrology
system.

Approach: Conduct collaborative, multi-model assessments of (1) the impact of precipitation
errors on watershed water balance estimates, (2) the impact of the WRF vs observed
precipitation differences on predicted biogeochemical cycling of nitrogen, (3) the impact of the
internally consistent WRF/Hydrology data on watershed water balance modeling, and (4) the
impact of the internally consistent WRF/CMAQ/Hydrology data on biogeochemical cycling of
nitrogen, particularly with respect to assessing the effects of climate change. AMAD will
evaluate the internally consistent meteorology/hydrology/chemistry from the standpoint of the
robustness of assessments of ecosystem exposure scenarios using calibrated biogeochemical
models.

AMAD will collaborate with NERL/ERD and academia to compare hydrologic routing
and depiction of the watershed water balance based on input of WRF precipitation with the
routing predicted by calibrated watershed and ecosystem models, such as the Grid Based
Mercury Model (GBMM) of ERD and the PnET-BGC model of Syracuse University, an
integrated biogeochemical model developed to simulate forest and aquatic ecosystems, calibrated
on observed precipitation and hydrology. This will eventually be compared to the hydrologic
routing produced by the WRF/Hydrology linked model set when those data are available.

AMAD will also collaborate with NERL/ESD to test and compare the hydrology predictions
from the WRF/Hydrology system and GIS-based models for small catchment regions and
examine the ability to represent regional hydrology with available GIS data layers. AMAD will
evaluate biases between observed and modeled hydrological processes and their impact on the
ecosystem model predictions. We would be looking for internal consistency and ability of the
WRF/Hydrology model set to support calibrated ecosystem models and test whether a
"calibration" based on the WRF/Hydrology linked model set is distorted or is basically
equivalent relative to a calibration based on observed data.

AMAD will collaborate with NERL/ERD and academia to evaluate the response surface
of the processing/cycling of nitrogen predicted by biogeochemical models under changing
environmental conditions. The response surface and its position in parameter space will be
evaluated for a suite of inputs with the hydrology and chemistry based on several different input
data sets: observation-based inputs, WRF-precipitation-based input, WRF/Hydrology system

16

-------
input, and WRF/CMAQ/Hydrology system input. The goal is to establish whether the nitrogen
response surface is significantly moved in parameter space due to the use of the internally
consistent WRF/CMAQ/Hydrology system set compared to a set based on observations that is
consistent with the model calibrations. This research will provide guidance on how best to link
the WRF-consistent hydrology with ecosystem models for management and climate change
studies. AMAD will also develop guidance on how best to provide hydrology coupled to
precipitation for critical load models such as the Model of Acidification of Groundwater in
Catchments (MAGIC) that do not have internal routing capability. The ecosystem and climate
teams within AMAD will coordinate closely on working with the down-scaled meteorology to
produce consistent hydrology for climate change studies based on the guidance developed above.
The hydrology will be characterized for a climate study period to establish a consistent baseline
for climate change studies involving ecosystem response to changes in precipitation that avoid or
at least mitigate the paradigm gap.

(6) Application of Expanded Model Capabilities

Directions: Assess the ability of CMAQ and WRF to address current and future ecosystem
exposure, and the development of future policy through modeling scenarios with policy and
climate change forcing. Target critical load studies and climate change assessments.
Communicate information on uncertainties to client stake-holders.

Approach: Develop targeted model applications to support EPA, federal agency and academic
partners. Potential application studies are expected to be EPA SOx-NOx welfare standards
assessments, Chesapeake Bay TMDL studies, multi-agency (NPS, USFS and EPA) critical loads
assessments (deposition) and Ecosystem Services Research Program (ESRP) scenario
assessments involving land use change and climate change. AMAD will provide CMAQ
deposition and air quality estimates for the ESRP research involving nitrogen at the national and
regional level. AMAD will develop post-processing procedures to address precipitation error
and potential emission errors, leading to estimates of uncertainty of relevance to the stake-holder
communities. AMAD will engage in providing improved and post-processed deposition
estimates for critical load studies conducted by EPA and the other federal agencies, learning to
enhance the use of CMAQ deposition estimates for this purpose. AMAD will engage in critical
load inter-comparisons with Canadian models under the US-Canada agreement between EPA
and Environment Canada. Key variables driving ecosystem response to climate change will be
identified with academic and federal partners, and guidance will be developed on metrics for
assessing the impacts of climate change. The AMAD ecosystem team will develop studies to
examine the potential impact of climate change on ecosystems and ecosystem services based on
the AMAD down-scaled meteorology in coordination with the AMAD climate team. Results
from these application studies will further guide in-house model development and application.
AMAD will use CMAQ as a laboratory to study the interactions with deposition and the
hydrological cycle to spur the development of methods and metrics for the application of
regional air quality models to ecosystem exposure studies under climate change scenarios.

A conceptual diagram of how the component parts of the above research integrate into the
overall modeling system to provide the capabilities needed to address air-ecosystem linkage and
climate change research needs, and how to answer the science questions is shown below.

-------
Linking Air Quality to Ecosystems and Watersheds to assess Land Use, Land

Cover and Climate Change Effects

Input Data Reflecting U.S. Policy
Options, Climate Change, and
Land Use Change Options

State-of-the-S cience
Environmental Process,
Transport and Loss models

Best Estimate of Climate, Land

Use, and Policy Change on
Human and Ecosystem Health

Implications for Model Development. Evaluation and Application Planning

Extend and refine bi-directional exchange algorithms in CMAQ

• Collaborative development of advanced bi-directional exchange (emission and
deposition) algorithms for Hg, NH3 and other bi-directionally mobile species for water
and terrestrial (soil and vegetation) surfaces. Include improved modeling of NO and the
addition of N2O.

• Collaborative development and evaluation (with USD A) of the fertilizer tool for creating
NH3 emissions from soils due to fertilizer application to support NH3 bi-directional
exchange calculations in CMAQ.

• Collaborative development and evaluation (with USD A) of the bi-directional flux and
regional transport of pesticides in the WRF/CMAQ system.

• Develop adjoint model for NH3 bi-directional exchange version of CMAQ

Harmonize air-ecosystem and WRF/CMAQ land surface characterizations

• Convert WRF/CMAQ to use NLCD land use and the CMAQ Mosaic approach to
estimate dry deposition output by land-use category within a grid.

• Establish consistent treatment of variables that are common, e.g., land cover, soil across
CMAQ, WRF, BEIS/MEGAN, EPIC, etc.

-------
•	Enhance national-scale vegetation distribution information (e.g., LAI) and expand to
include additional remotely sensed and ground-based survey (e.g., USDA Cropland Data
Layer, USGS LANDFIRE) data sources (ESD).

Address missing or inaccurate process descriptions

•	Enhance WRF and CMAQ capabilities to simulate complex terrain using a smaller grid
size (e.g., 4 km) .

•	Address missing process descriptions

o Lightning NOx

o Cloud impaction in high elevation terrain

o Base cation deposition

o Approach to address dissolved organic nitrogen (DON) deposition

Extend tool capabilities related to agricultural and land use management

•	Enhance WRF and CMAQ to respond to more specific land cover information e.g., corn,
peanuts as opposed to "agricultural land", wetlands, irrigated lands, etc.

•	Integrate the capability to model the effects of agricultural management and non-point
source pollution releases beyond fertilizer application into CMAQ to improve current
land surface parameterization.

•	Upgrade land-surface model to account for water budgets and denitrification of water-
column N and potential production of N2O (greenhouse gas).

•	Link CMAQ flux algorithms (e.g., early morning release of ammonia and agri-chemicals
that have accumulated in the vegetation canopy overnight) to local-scale human
exposure.

Enhance ability to address within-grid deposition variability

•	Develop new CMAQ output options e.g., MOSAIC to estimate dry deposition output by
land-use category and the Watershed Deposition Tool to better communicate CMAQ
simulation outcomes to our hydrologic and ecosystem exposure clients.

•	Enhance land-surface model to account for fine scale processes and complex terrain,
particularly as it relates to sub-grid scale dry-deposition variability.

•	Develop a GIS-based approach with CMAQ that uses sub grid information on terrain and
other key variables to account for sub grid variation in dry deposition due to variation in
terrain.

Develop link between WRF and hydrology models. Improve WRF precipitation.

•	Develop and evaluate a post-WRF/CMAQ processing tool or system able to estimate
hydrology metrics such as runoff that are consistent with modeled (e.g., WRF-generated)
precipitation.

•	Enhance CMAQ-hydrology linkage to subgrid hydrology and biogeochemical models
such as the Pnet-BGC systems for aquatic systems response to assess critical loads and
climate change.

•	Improve WRF precipitation simulations, including extremes to support linkage to
calibrated hydrological models, to support estimation of ecosystem services and
ecosystem exposure.

•	Improve and evaluate downscaled climate precipitation scenarios against historical
climate means and variability to establish credibility in the ecological and hydrological

19

-------
communities. Exploration of alternative methods of using existing precipitation
scenarios (e.g., ensemble approaches).

Advance model evaluation and model intercomparison techniques

• Evaluate CMAQ atmospheric deposition predictions and define biases, input errors, and
missing input components, such as base cations, and those relating to the nitrogen budget
(e.g., lightning NOx).

• Develop additional CMAQ tools in-house and through collaborations for addressing
source apportionment management questions (e.g., DDM-3D).

• Conduct an inverse modeling study to examine differences introduced by the NH3 bi-
directional parameterization.

• Conduct model intercomparisons with Canada on critical load calculations.

• Design approaches to post process CMAQ deposition to reduce error (e.g., precipitation
adjustment of wet deposition or data fusion) to obtain most accurate deposition estimate.

• Develop new diagnostic output options, e.g. flux components as opposed to net flux,
additional chemical species to make better use of collaborative field study results for
model development and evaluation.

Advance model application capability

• Assess barriers and potential approaches to development of long term multi-year CMAQ
deposition values needed by biogeochemical/watershed models

• Conduct targeted research applications

o Critical load/Secondary Standards/TMDL Studies

o Ecosystem Services Research Studies for national atlas, and place-based research

oriented towards nitrogen
o Water quality/water quantity studies
o Climate change studies with down-scaled meteorology

Anticipated Major End Points
1-2 Years

CMAQ with advanced bi-directional air-surface exchange algorithms incorporated for NH3 and
Hg and updates to uni-directional dry deposition algorithms.

Lightning NOx included in CMAQ (addressing a missing pathway)

FEST-C available to the community for retrospective and near-term assessments.

CMAQ with the Mosaic option implemented to output subgrid dry deposition by land use type
and fraction.

Land use fully harmonized between WRF, CMAQ and ecosystem models (NLCD land use).

Method developed to post-process WRF precipitation and CMAQ wet deposition predictions to
reduce wet deposition error for use in, for example, critical load studies.

-------
CMAQ sulfur and nitrogen deposition results with bi-directional NH3 available for multi-agency
studies of critical loads in collaboration with the Office of Air Programs/Clean Air Markets
Division.

Atmospheric deposition scenarios for Ecosystem Services Research Program national and place-
based study assessments.

2-4	Years

Bi-directional NH3 adjoint version of CMAQ operational

Continued CMAQ air-surface exchange evaluations against new flux data available through
collaborative research efforts in coordination with Clean Air Markets Division

Cloud water deposition and tracking of base cations included in CMAQ (addressing two missing
pathways)

WRF improvements implemented based on 3-D variational assimilation techniques and WRF
physics options and data assimilation approaches defined that produce the most accurate
simulation at a 4 km grid size.

Harmonized soils between WRF, CMAQ, FEST-C or EPIC, ecosystem and hydrology models.

FEST-C available to the community for land use change and climate change assessments.

Development of linkage between WRF and a hydrology model accomplished.

NH3 inversion modeling initiated with new NH3 adjoint version of CMAQ to replace previous
investigations

Model intercomparison with Canadian models of critical load calculations for US-Canada
Accord (follow-on the AQMEII model intercomparison)

Application study support for the next round of NOx-SOx welfare standard setting process and
key TMDL assessments in collaboration with NCEA and OAQPS

Atmospheric deposition scenarios for Ecosystem Services Research Program national and place-
based study assessments and for critical load assessments for alternative scenarios.

Preliminary investigations of the effects of land use and climate change on ecosystem health and
water quality using new WRF/CMAQ/Hydrology systems

3-6	Years

CMAQ with bi-directional air-suface exchange of select pesticides developed in collaboration
with USD A

21

-------
Preliminary approach developed to address dissolved organic nitrogen (DON) in CMAQ
(address a missing pathway)

Continued CMAQ air-surface exchange evaluations against new flux data available through new
mobile flux platform deployed at selected CASTNet sites in coordination with Clean Air
Markets Division

WRF/CMAQ with advanced hydrology surface layer developed with an uncertainty
characterization

Initial advanced subgrid variability estimates for dry deposition developed for application
studies.

FEST-C advanced to consider nitrogen fate and N20 production for climate change assessments.

Guidance developed for linking WRF/CMAQ with ecosystem models to address the paradigm
gap. Guidance developed for linking WRF/CMAQ with coupled hydrology to ecosystem
models. Guidance developed for creating climate baseline meteorology and coupled hydrology

Application study support for the next round of NOx-SOx welfare standard setting process and
critical load and key TMDL assessments in collaboration with NCEA and OAQPS

Atmospheric deposition studies and scenarios for Ecosystem Services Research Program national
nitrogen, wetlands and place-based assessments for future conditions that include land use and
climate change.

Integrated studies with atmospheric and ecosystem models assessing the impact of climate
change on critical loads and ecosystem health.

22

-------
References

Appel, K.W., Roselle, S.J., Gilliam, R.C. and Pleim, J.E., 2009. Sensitivity of the Community
Multiscale Air Quality (CMAQ) Model v4.7 results for the eastern United States to MM5 and
WRF meteorological drivers. Geoscience Model Development Discussion, 2:1081-1114.

Baron, Jill S.; Rueth, Heather M.; Wolfe, Alezander M.; Nydick, Koren R.; Allstott, Eric J.;
Minear, J. Toby, and Moraska, Brenda, 2000. Ecosystem Responses to Nitrogen Deposition in
the Colorado Front Range, Ecosystems, 3, 352-368.

Bossi, R., Skov, H., Vorkamp, K., Christensen, J., Rastorgi, S.C., Egel0v, A., Petersen, D., 2008
Atmospheric concentrations of organochlorine pesticides, polybrominated diphenyl ethers and
polychloronaphthalenes in Nuuk, South-West Greenland. Atmospheric Environment 42, 7293-
7303.

Bowman, William D., Julia R. Gartner, Keri Holland, and Magdalena Wiedemann. 2006.
Nitrogen Critical Loads for Alpine Vegetation and Terrestrial Ecosystem Response: Are We
There Yet? Ecological Applications, 16, 1183-1192.

Clean Air Act Advisory Committee, 1005. Air Quality Management Group Report to the Clean
Air Act Advisory Committee, 190 pp, http://www.epa.gov/air/caaac/aqm/reportl-17-05.pdf.

DeHays, D.H., Schaberg, P.G., Hawley, G.J., Strimbeci, G.R., 1999. Acid rain impacts on
calcium nutrition and forest health, Bioscience, 49, pp 789-800.

Dennis, R.L., Mathur, R., Pleim, J.E. Walker, J.T., in press. Fate of ammonia emissions at the
local to regional scale as simulated by the Community Multiscale Air Quality Model,
Atmospheric Pollution Research, in press.

Dennis, R., Haeuber, R., Blett, T., Cosby, J., Driscoll, C., Sickles, J. and Johnston, J., 2007.
Sulfur and nitrogen deposition on ecosystems in the United States. Environmental Manager,
December 2007: 12-17.

Dennis, R.L., 1997. Using the Regional Acid Deposition Model to Determine the Nitrogen
Deposition Airshed of the Chesapeake Bay Watershed, in Baker, J.E. editor, Atmospheric
Deposition to the Great Lakes and Coastal Waters, Society of Environmental Toxicology and
Chemistry, Pennsacola, FL, pp 393-413.

Driscoll, Charles T., Gregory B. Lawrence, Arthur J. Bulger, Thomas J. Butler, Christopher S.
Cronan, Christopher Eagar, Kathleen F. Lambert, Gene E. Likens, John L. Stoddard, and
Kathleen C. Weathers. 2001. Acidic Deposition in the Northeastern United States: Sources and
Inputs, Ecosystem Effects, and Management Strategies. Bioscience, 51, 180-198.

Driscoll, Charles T., David Whitall, John Aber, Elizabeth Boyer, Mark Castro, Christopher
Cronan, Christine L. Goodale, Peter Groffman, Charles Hopkinson, Kathleen Lambert, Gregory

23

-------
Lawrence, and Scott Ollinger. 2003. Nitrogen Pollution in the Northeastern United States:
Sources, Effects, and Management Options. Bioscience, 53, 357-374.

Driscoll, CharlesT., Young-Ji Han, Celia Y. Chen, David C. Evers, Kathleen F. Lambert,

Thomas M. Holsen, Neil C. Kamman and Ronald K. Munson. 2007. Mercury contamination in
forest and freshwater ecosystems in the Northeastern United States. Bioscience, 57, 17-28.

Fenn, Mark E., Mark A. Poth, John D. Aber, Jill S. Baron, Bernard T. Bormann, Dale W.
Johnson, A. Dennis Lemly, Steven G. McNulty, Douglas F. Ryan, and Robert Stottlemyer. 1998.
Nitrogen Excess in North American Ecosystems: Predisposing Factors, Ecosystem Responses,
and Management Strategies. Ecological Applications, 8, 706-733.

Fenn, Mark E., Jill S. Baron, Edith B. Allen, Heather M. Rueth, Koren R. Nydick, Linda Geiser,
William D. Bowman, James O. Sickman, Thomas Meixner, Dale W. Johnson, and Peter Neitlich.
2003. Ecological Effects of Nitrogen Deposition in the Western United States. Bioscience, 53,
404-420.

Galloway, J.N., Cowling, E.B., 2002. Reactive Nitrogen and the World: 200 years of change,
Ambio, vol. 31 no. 2, pp64-71.

Galloway, J.N., Aber, J.D., Erisman, J.W., Seitzinger, S.P., Howarth, R.W., Cowling, E.B.,
Cosby, B.J., 2003. The nitrogen cascade, Bioscience, 53, pp 341-356.

Geiser, Linda H., Jovan, Sarah E., Glavich, Doug A., Porter, Matthew K., 2010. Lichen-based
critical loads for atmospheric nitrogen deposition in Western Oregon and Washington Forests,
USA, Environmental Pollution, 158, 2412-2421.

Golden, H.E., Knightes, C.D., Cooter, E.J., Dennis, R.L., Gilliam, R.C., in press. Linking air
quality and watershed models for environmental assessments: Analysis of the effects of model-
specific precipitation estimates on calculated water flux, Environmental Modeling & Software
(2010), doi:10.1016/j.envsoft.2010.04.015.

Jeremiason, J.D., Engstron, D.R., Swain, E.B., Nater, E.A., Johnson, B.M., Almendinger, J.E.,
Monson, B.A., Kolka, R.K., 2006. Sulfate addition increases methylmercury production in an
experimental wetland, Environmental Science and Technology, 40, pp 3800-3806.

Katata, G., Nagai, H., Wrzesinsky, T., Klemm, O., Eugster, W., and Burkard, R, 2008.
Development of a Land surface Modeling Including Cloud Water Deposition on Vegetation,
Journal of Applied Meteorology and Climatology, 47, 2129-2146.

Linker, L.C, G.W. Shenk, R.L. Dennis, J.S. Sweeney, 2000. Cross-Media Models of the
Chesapeake Bay Watershed and Airshed, Water Quality and Ecosystem Modeling, 1, 91-122.

Lovett, G.M. and T.H. Tear, 2008. Threats from Above: Air Pollution Impacts on Ecosystems
and Biological diversity in the Eastern United States, The Nature Conservancy and the Cary
Institute of Ecosystem Studies, Bethesda, MD and Millbrook, NY. 28 pp.

24

-------
(www.ecostudies.org/reprints/Threats from Above.pdf). (Available online at
www.nature.org/wherewework/northamerica/states/maryland/)

National Exposure Research Laboratory, 2009. A conceptual Framework for U.S. EPA's
National Exposure Research Laboratory, Report No. EPA/600/R-09/003, Office of Research and
Development, Washington, DC 20460, 42 pp.

National Research Council (NRC), 2004. Air Quality Management in the United States. Report
by the Committee on Air Quality Management in the United States, The National Academy
Press, Washington, D.C.

Nemitz, E., Milford, C., Sutton, M.A., 2001. A two-layer canopy compensation point model for
describing bi-directional biosphere-atmosphere exchange of ammonia. Q. J. R. Meterol. Soc.,
727, 815-833

Paerl, H.W., Dennis, R.L., Whitall, D.R., 2002. Atmospheric Deposition of Nitrogen:
Implications for Nutrient Over-enrichment of Coastal Waters, Estuaries, 25, No. 4B, 677-693.

Pleim, J.E. and R.C. Gilliam, 2009. An Indirect Data Assimilation Scheme for Deep Soil
Temperature in the Pleim-Xiu Land Surface Model. Journal of Applied Meteorology and
Climatology, 48(7): 1362-1376.

Pleim, J.E., 2007a. A combined local and nonlocal closure model for the atmospheric boundary
layer. Part I: model description and testing, Journal of Applied Meteorology and Climatology,
46, 1383-1395.

Pleim, J.E., 2007b. A combined local and nonlocal closure model for the atmospheric boundary
layer, Part II: application and evaluation in a mesoscale meteorological model, Journal of
Applied Meteorology and Climatology, 46: 1396-1409.

Rimmer, C.C., Mcfarland, K.P., Evers, D.C., Miller, E.K., Aubry, Y., Busby, D., Taylor, R.J.,
2005. Mercury concentrations in Bicknell's thrush and other insectivorous passerines in
Montane forsts of northeastern North America, Ecotoxicology, 14, pp 223-240.

Seigneur, C. and Dennis, R.L., in press. Chapter 8: Atmospheric Modeling, in Technical
Challenges of Risk- and Results-Based Multipollutant Air Quality management, Hidy, G., J.R.
Brook, K.L. Demeijian, L.T. Molina, W.T. Pennell, and R. Scheffe (editors), Springer,
Dordrecht, 2010.

Sullivan, T.J., Cosby, B.J., Webb, J.R., Dennis, R.L., Bulger, A.J. and Deviney, F.A., 2008.
Streamwater acid-base chemistry and critical loads of atmospheric sulfur deposition in
Shenandoah National Park, Virginia. Environmental Monitoring and Assessment, 137(1-3): 85-
99.

25

-------
Sutton, M.A., J.K. Burkhardt, D. Guerin, E. Nemitz, D. Fowler, 1998. Development of resistance
models to describe measurements of bi-directional ammonia surface-atmosphere exchange.
Atmos. Environ. 32, 473-480

Valigura, R.A., Alexander, R.B., Castro, M.S., Meyers, T.P., Paerl, H.W., Stacey, P.E., Turner,
R E. (editors), 2001. Nitrogen Loading in Coastal Water Bodies: An Atmospheric
Perspective, Coastal and Estuarine Studies No. 57, American Geophysical Union, Washington
DC, 254 pp.

Vitousek, M., J.D. Aber, R.W. Howarth, G.E. Likens, P.A. Matson, D.W. Schindler, W.H.
Schlesinger, D.G. Tilman, 1997. Technical Report: Human Alteration of the Global Nitrogen
Cycle: Sources and consequences, Ecological Applications, vol. 7, no. 3, pp 737-750.

Walker, J.T., Robarge, W.P., Wu, Y., Meyers, T., 2006. Measurement of bi-directional ammonia
fluxes over soybean using the modified Bowen-ratio technique, Agricultural and Forest
Meteorology, 138, 54-68.

Weathers, K.C., Lovett, G.M., Likens, G.E., 1995. Cloud deposition to a spruce forest edge,
Atmospheric Environment, 29, 665-672.

Weathers, K.C., Simkin, S.M., Lovett, G.M., Lindberg, S.E., 2006. Empirical modeling of
atmospheric deposition in mountainous landscapes, Ecological Applications, 16, 1590-1607.

26

-------