&EPA
www.epa.gov/research
EPA/600/F-14/315
science in ACTION
INNOVATIVE RESEARCH FOR A SUSTAINABLE FUTURE
Opportunities and Challenges for Geographically Expanding N-Sink
Introduction
Nitrogen is increasingly being
identified as a pollutant of concern
in both coastal and inland waters. In
some areas, the majority of the
nitrogen loading comes from
wastewater treatment plants and/or
combined sewer overflows.
However, in less urbanized
catchments nonpoint source runoff
and nitrogen from septic systems are
the primary vehicles of nitrogen
delivery. In these areas, catchment
land use has a direct relationship
with both sources and sinks of
nitrogen.
The N-Sink tool was created to
provide a useful and accessible
means for local land use managers to
explore the relationship of land use
in their towns and counties to
nitrogen pollution of their waters. N-
Sink focuses on three types of
landscape N sinks: wetlands,
lakes/ponds/reservoirs, and stream
reaches. N-Sink uses the best
available science on landscape-
nitrogen interactions, plus widely
available basic datasets for
hydrography, soils and land cover, to
highlight major sources and sinks of
nitrogen within a catchment context.
N-Sink is available as an
extension to ArcMap, and as a web-
based tool,
(www. edc .uri. edu/nsinkv2/). The
geographic extent of N-Sink is
currently from just west of
Narragansett Bay, extending west
nearly to the Connecticut River
(Figure 1). N-Sink uses HUC12
boundaries as an organizing unit,
fc* f O
Figure 1. The current geographic extent of N-Sink, from just east of the Connecticut River
to just west of Narragansett Bay (figure from the web version).
and as such, currently has analysis
capabilities along coastal HUC12s,
and those abutting the coastal
HUC12s.
The focus of this Science in Action
brief is to explore the technical pros
and cons of applying N-Sink to other
geographic areas.
Considerations for Geographic
Expansion of N-Sink
Data Sources:
NHDPlusV2
N-Sink uses the National
Hydrography Dataset, NHDPlusV2,
(McKay et al., 2012) to acquire
hydrologic data necessary for
flowpath generation and N removal
estimates for stream reaches and
lakes/ponds/reservoirs along a
specified flowpath. N-Sink flowpath
generation makes use of
NHDPlusV2 flowlines. N removal
estimates use NHDPlusV2 estimates
of cumulative drainage areas, and
gauge adjusted mean annual flow
and velocity. Using this national
dataset has capitalized on intensive
efforts by USGS to develop and
refine NHDPlusV2.
However, there were some
technical issues associated with
NHDPlusV2 that required manual
editing before those data could be
used in N-Sink. This is a major
consideration for geographic
expansion of N-Sink. The manual
tasks included: 1) multiple flowlines
through lakes/ponds had to be
merged, and the associated lengths
and drainage areas summed, in order
to obtain a single lake/pond feature
(see Figure 2); 2) upstream and
downstream segment IDs needed to
be manually changed to ensure
U.S. Environmental Protection Agency
Office of Research and Development
-------
hydroconnectivity and allow the
model to iterate properly through the
merged segments, and; 3) some table
joins and calculations were needed
to deal with stream reaches with
multiple line segments, all with the
same reach code.
Pros: Nationally available data, so
variations in climate are accounted
for across regions.
Cons: Apparent need for manual
editing (labor intensive).
Prospects/Next Steps: Investigate
methods to automate one or all of
the manual tasks listed above. This
may include converting the vector
analysis that estimates N removal
along a flowpath to raster. Based on
discussions with GIS practitioners, it
would be worthwhile to convert one
HUC12 from vector to raster to
better understand the full range of
opportunities and challenges and to
compare results from the two
methods. Conduct expansion of N-
Sink to a regional or larger (HUC-8)
watershed to gauge feasibility of
manual method. Talk to USGS about
future updates to NHDPlusV2 that
would ensure that every unique
stream reach is composed of only
one line segment, and that multiple
reaches within water bodies can be
merged such that each lake/pond can
be treated as a single feature.
Figure 2. Detail of N-Sink flowpath for the Chipuxet River tributary of the
Pawcatuck River in RI. The three NHD flow lines (two in dotted blue and the one
chosen by the flow path) shown have to be merged by hand in order for N-Sink to
treat the lake as one N sink feature (inset, color coded by N removal potential).
U.S. Environmental Protection Agency
Office of Research and Development
NLCD
The National Land Cover Dataset
(http://www.epa.gov/mrlc/nlcd-
2006.htmr) provides land cover
raster data at a cell size of 30 m.
Pros: Nationally available data.
Cons: The cell size limits our ability
to reliably identify narrow strips of
riparian wetlands (discussed below
under SSURGO).
Prospects/Next Steps: None
necessary. N-Sink seems to produce
useful results using 30m data.
Eventually, it would be good to
build in the ability to allow the user
to add higher resolution local data if
available, but this is not critical.
SSURGO
N-Sink identifies wetlands by
extracting hydric soils (classified as
"poorly drained" and "very poorly
drained") from the Soil Survey,
SSURGO (Soil Survey Staff, 2012)
and overlaying NLCD data to
include only those hydric soils that
are "undeveloped".
Pros: Nationally available data.
Resolution ranges from 1:12,000 to
1:63,360. Vector data, with high
resolution and accuracy.
Cons: None identified.
Prospects/Next Steps: None needed.
Model Assumptions:
Hydrologic Flowpaths
N-Sink characterizes the flowpath
from any given point to a catchment
outlet as having a terrestrial
component (from source to surface
water), followed by a surface water
component. Once a flowpath enters
surface water, it remains as surface
water. This means that even if a
stream reach is surrounded by
wetlands, removal is based on
movement through the stream reach,
not on movement through wetland
soils. This approach results in a
conservative estimate of cumulative
N removal.
Each of these components
(terrestrial and surface water) is
usually made up of a number of
"reaches". The terrestrial component
assumes that surface flowpaths
-------
represent a close approximation of
subsurface flowpaths. A major
consideration for geographic
expansion of the N-Sink tool is those
situations where this assumption will
likely be violated. These include
dense urban and other areas with
engineered drainage, tile drained
agricultural areas, areas with karst
geology or where significant flow
occurs through bedrock fissures.
Pros: It is possible to estimate a
complete flowpath from source to
outlet.
Cons: This approach cannot be used
in areas where surface flowpaths do
not approximate subsurface
flowpaths.
Prospects/Next Steps: This appears
to be a major consideration for the
use of N-Sink as decision support
for the areas of a watershed with
urban drainage. While this does not
negate N-Sink's usefulness as a N
delivery tool for many areas of a
given watershed, it does require
more discussion by the team in how
to deal with this issue.
N Removal Estimates
Estimated rates of removal from
landscape sinks are based on data
from a wide geographic range (see
Arnold et al., 2013 for methods and
citations). In order to evaluate the
validity of these N removal
estimates in other regions, it will be
necessary to engage biogeochemists
who work in these regions and are
familiar with the mechanisms of N
movement through soils and surface
water in their regions. It is quite
likely that estimates will translate
well to other glaciated landscapes
similar to the Northeastern US.
Pros: Current methods estimate N
removal based on relationships that
depend on retention time as a key
factor. Because retention times are
characterized using data from
NHDPlusV2, regional differences in
climate and other hydrologic drivers
are fairly well accounted for.
Cons: Some accuracy is likely
sacrificed by using this broadbrush
approach.
Prospects/Next Steps : We are
exploring ways to assess this.
Colleagues in other areas of the
country need to be consulted.
Summary
There is great interest in expanding
the geographic extent of N-Sink in
the region, within the Narragansett
Bay and Long Island Sound
watersheds. N-Sink is designed to
use nationally available data sets,
which will facilitate expanding the
tool to other regions. One large
hurdle is the apparent need to
modify the data from NHDPlusV2.
If we stay with vector-based
flowpath calculations, this will be
potentially prohibitive unless an
automated process can be found. A
possible alternative is to migrate to
raster-based operations. Another
hurdle is to determine how to use N-
Sink in watersheds that have
extensive urban, tiled agriculture, or
karst areas.
Recommended Next Steps:
1) Explore the feasibility of
converting N-Sink calculations
to raster operations using a pilot
HUC12 from the current extent
of N-Sink, allowing us to
compare vector-based
calculations with raster-based
calculations.
2) If conversion to raster allows for
relatively efficient expansion of
the NHDPlusV2 data, then the
team would identify regions
where surface and subsurface
flowpaths are expected to be
significantly different and
eliminate these areas from
consideration.
3) Identify the next region for
expansion and work with
biogeochemists from that region
to adjust N removal
calculations, if necessary.
References
Arnold, C., D.Q. Kellogg, K. Forshay, C.
Damon, E.H. Wilson, A. Gold, E.A.
Wentz, and M.M. Shimizu. 2013. The
"N-Sink" Web Tool and Two Case
Studies. EPA Technical Report
EPA/600/R-13/230.
Kellogg, D. Q., Arthur J. Gold, Suzanne
Cox, Kelly Addy and Peter V. August.
2010. A geospatial approach for
assessing denitrification sinks within
lower-order catchments. Ecological
Engineering (36): 1596-1606.
Kellogg. D.Q., K. Forshay and C.
Arnold. Development of a Tool to
Identify Nitrogen Sources and Sinks
within a Watershed Framework. EPA
Science Brief EPA/600/F-13/223.
McKay, L., T. Bondelid, T. Dewald, et
al. 2012. NHDPlus Version 2: User
Guide. Available online at:
ftp://ftp.horizon-
svstems.com/NHDPlus/NEroPlusV21/
Documentation/NHDPlusV2_User_Gui
de.pdf. Accessed 06/05/14.
Soil Survey Staff, Natural Resources
Conservation Service, United States
Department of Agriculture. Web Soil
Survey. Available online at
http://websoilsurvev.nrcs.usda.gov/.
Accessed 06/05/14.
Contacts:
Dr. Q Kellogg
University of Rhode Island
(401)874-4866
q@edc.uri.edu
Chet Arnold
University of Connecticut
(860)345-5230
Chester. arnold@uconn. edu
Dr. Ken J. Forshay
Project Officer
US EPA Office of Research and Development
(580)4368912
Forshay.Ken@EPA.gov
Dr. David Burden
Tech Support Center Director
US EPA Office of Research and Development
Burden.David@EPA.gov
U.S. Environmental Protection Agency
Office of Research and Development
------- |