EPA/600/R-09/116 September 2009| www.epa.gov/athens
?/EPA
United States
Environmental Protection
Agency
Modeling the Impacts of Hydromodification on Water
Quantity and Quality
-------
EPA/600/R-09/116
September 2009
Modeling the Impacts of Hydromodification on Water
Quantity and Quality
by
Yusuf M. Mohamoud, Ph.D., P.E.
U.S. EPA, Office of Research and Development
National Exposure Research Laboratory
Ecosystems Research Division
Athens, Georgia
Anne C. Sigleo, Ph.D. (Retired)
U.S. EPA, Office of Research and Development
National Health and Environmental Effects Laboratory
Western Ecology Division
Newport, Oregon
RajbirS.Parmar,Ph.D.
U.S. EPA, Office of Research and Development
National Exposure Research Laboratory
Ecosystems Research Division
Athens, Georgia
U.S. Environmental Protection Agency
Office of Research and Development
National Exposure Research Laboratory
Ecosystems Research Division
Athens, GA 30605
-------
NOTICE
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 an endorsement or recommendation for use.
11
-------
ACKNOWLEDGEMENT
We are grateful to Ben Scales and Lloyd VanGordon of the Oregon Water Resources
Department for providing streamflow data from the Chitwood gaging station, to K.
Ramage for the Nashville weather data, to the Dynamac team for help with sample
collections. We are also grateful to Bob Swank, Caroline Stevens, Michael Cyterski, and
Roger Burke for reviewing Part 2 of this report and Jim Carleton for reviewing the entire
report.
in
-------
Table of Contents
List of Figures vi
List of Tables vii
PART 1. METHOD DEVELOPMENT 1
MODELING THE IMPACTS OF HYDROMODIFICATION ON WATER QUANTITY AND
QUALITY 1
1 INTRODUCTION 3
2 HYDROMODIFICATION: SOURCE OF INTEGRATED STRESSORS 7
3 HYDROMODIFICATION CATEGORIES AND MODELING CHALLENGES 9
3.1 Urbanization 9
3.2 Water Withdrawals and Interbasin Transfers 10
3.3 Channel Modification and Streambank Erosion 11
3.4 Flow Regulation by Dams and Impoundments 12
3.5 Climate Change 13
4 THE NEED FOR HYDROMODIFICATION MODELING FRAMEWORK 15
4.1 Essential Components of a Hydromodification Modeling Framework 18
4.2 BASINS Modeling Framework: A Background 20
4.3 Hydromodification Decision-making Example 24
4.4 Linking Integrated Stressors to A quatic Ecosystem Impacts 2 7
4.5 Model Application Example 2 7
PART 2: APPLICATION EXAMPLE 29
5. FORECASTING URBANIZATION IMPACTS ON WATER QUANTITY AND QUALITY:
CASE STUDY OF THE YAQUINA WATERSHED, OREGON, USA 29
5.1 Introduction 31
5.1.2 Watershed Description 34
5.2 Methods 38
5.2.1 Model Setup and Input Data 39
5.2.2 Model Calibration and Validation 41
5.2.3 Model Performance Evaluation Criteria 42
5.2.4 Future Build-out Scenario Development 43
5.2.5 Forecasting Flow, Nitrate, and TSS Alterations 45
5.2.6. Linking Stressors and to Impacts 46
5.3 Results and Discussion 48
5.3.1 Model Calibration and Validation Results 48
IV
-------
5.3.2 Forecasting Stressor Levels for the Build-out Scenarios 54
5.4 Determining Ecologically Relevant Urbanization-Induced Stressor Indicators 60
6 CONCLUSION 65
7 REFERENCES 67
-------
List of Figures
Figure 1. Top 10 sources of impairment in assessed rivers and streams in the United
States (Source: USEPA, 2009) 8
Figure 2. Hydromodification activities and watershed management questions that
resource managers must answer before hydromodification projects are implemented.... 17
Figure 3. Components of a hydromodification modeling framework 19
Figure 4. Matching framework modeling capabilities with hydromodification drivers,
integrated stressors, and integrated effects 22
Figure 5.. Decision-making framework for watersheds receiving hydromodification
projects 26
Figure 6. Location of the study watershed 35
Figure 7. Distribution of long-term monthly precipitation and temperature for Yaquina
Watershed 36
Figure 8. Model calibration results: streamflow (a and b), TSS (c), and nitrate (d)
calibrations 53
Figure 9. Hydrologic changes simulated for baseline, scenariolS, scenario45, and
scenarioSS: (a) daily streamflows, (b) annual water balances (c) flow duration curves at
low flow conditions, and (d) flow duration curves at high flow conditions 56
Figure 10. HSPF simulated streamflow, TSS, and nitrate concentrations for baseline,
scenariolS, scenario45, and scenarioSS 59
Figure 11. HSPF simulated 30-year mean water temperature (a), water surface width (b),
and flow velocity (c) for baseline, scenariolS, scenario45, and scenarioSS 64
VI
-------
List of Tables
Table 1. Baseline and hypothetical future land use scenarios 37
Table 2. Model calibration and validation results 49
TableS. List of parameter values adjusted during calibration 50
Table 4. Indicators of hydrologic alteration (IHA) calculated from streamflow simulated
for the baseline and the three build-out scenarios for a 30-year period 61
vn
-------
PART 1. METHOD DEVELOPMENT
MODELING THE IMPACTS OF HYDROMODIFICATION ON
WATER QUANTITY AND QUALITY
Abstract: Hydromodification activities are driven by human population growth and
resource extraction and consumption including urbanization, agriculture, forestry,
mining, water withdrawal, climate change, and flow regulation by dams and
impoundments. These anthropogenic activities alter natural flow regimes and lead to
reduced downstream water quantity and degraded water quality. Recently, USEPA and
states recognized hydromodification as a stressor and a leading source of water quality
impairment in streams and rivers. Hydromodification-induced stressors include chemical
pollutants, pathogens, nutrients, suspended solids, and flow and habitat alteration. The
diverse and interacting nature of hydromodification-induced stressors has made Total
Maximum Daily Load (TMDL) development for impaired streams and rivers a major
regulatory challenge. Because hydromodification integrates stressors that have combined,
cumulative, and synergistic effects on water quantity and quality, TMDL modeling
approaches are not well-suited for simulating the impacts of hydromodification.
Modeling integrated stressors requires the development and application of predictive
models and innovative modeling approaches, such as the Better Assessment Science
Integrating Point and Nonpoint Sources (BASINS) modeling framework. Although
BASINS has been in use for the past 10 years, there has been limited modeling guidance
on its applications for complex environmental problems, such as modeling impacts of
hydromodification on water quantity and quality. This report consists of two parts: Part 1
presents the development of a B ASINS-based methodology that is applicable to modeling
1
-------
hydromodification. Part 2 is a case study of how the proposed modeling approach can
forecast the impacts of urbanization on water quantity and quality.
-------
1 INTRODUCTION
Rivers and river-fed lakes are valuable natural resources providing about 61 percent
of the nation's drinking water as well as serving riverine habitats to an estimated 40
percent of the fish species and about half of the birds in North America (Whiting, 2002).
In addition, rivers store flood waters for groundwater recharge and provide recreational
amenities, such as boating, fishing, and swimming (Whiting, 2002) along with
navigation, irrigation, power generation, and waste load transport and assimilation (Poff
et al., 1997). These valuable ecosystem services are threatened by hydromodification
projects such as land resources development for agriculture, energy, mining, forestry,
transportation, and residential housing; and water resources development for irrigation,
municipal water supply, and flood control. For instance, urban development and efforts to
use rivers for transportation, water supply, flood control, irrigation, and power generation
often alter flow regimes thus threatening the sustainability of the ecosystem services that
rivers and river-fed lakes provide (Poff et al., 1997).
Poff et al. (1997) emphasized the importance of managing the impacts of
anthropogenic watershed disturbances and urged scientists to develop management
protocols that accommodate economic uses while protecting ecosystem functions.
Naiman et al. (2002) stated that forecasting the impacts of changing water regimes is a
fundamental challenge for the scientific community. A way to address these challenges
and achieve sustainable management of land and water resources at the watershed level is
to develop integrative modeling approaches that consider stressors as a system with
positive and negative feedback loops, synergies, and interferences (Zimmerman et al.
2009).
-------
A number of factors have delayed the development of integrative modeling
approaches and their application to complex hydromodification problems. First, cross-
disciplinary, professional boundaries, and different views among hydrologists, engineers,
planners, ecologists, and biologists make it difficult to apply a holistic approach to
evaluating the impacts of hydromodification. Second, watershed boundaries, which are
basic environmental management units, do not usually coincide with local government
boundaries where hydromodification decisions are made. Third, water quantity and water
quality are mostly regulated separately by various federal and state agencies, even though
water quantity strongly influences water quality. Fourth, land use planning, which has
strong influence on water quantity and quality, is regulated at the city or county boundary
level even though the impacts of land use change on water quantity and quality transcend
local boundaries.
Recently, a number of studies have used integrative modeling approaches to assess
water allocation options (Letcher et al, 2004), develop hydrologic, agronomic, and
economic models for river basin management (Cai et. al. 2003), develop multi-objective
evolutionary algorithms for managing ecosystem services (Bekele and Nicklow, 2005),
and integrate water allocation and water quality models (Azevedo et al. 2000). Clearly,
managing the impacts of hydromodification on water quantity, habitat, and water quality
requires modeling approaches that forecast the stressor levels associated with alternative
future scenarios (Mohamoud, 2008). Two recent federal government reports on water
research and data across federal agencies have emphasized the need for comprehensive
watershed management approaches (NRC, 2004; General Accounting Office, 2004). In
-------
this report, integrative modeling approaches are defined as those capable of simulating
multiple stressors and their impacts on water quantity and quality.
The dominant hydromodification-induced stressors include flow alteration, water
quality degradation, and habitat alteration. At present, EPA's traditional water-quality
criteria and standards program do not address the effects of habitat alteration and flow
regulation on aquatic life (Jackson and Davis, 1994). Flow alteration due to
hydromodification alters water quantity which strongly influences water quality, yet
regulatory programs usually do not examine how water quantity changes affect water
quality. An area with strong relevance to water quantity and quality is the TMDL
program because pollutant load calculations are flow-dependent and are calculated as the
product of concentration and streamflow. Flow alterations due to hydromodification may
introduce errors in TMDL allocation estimates because relationships of water quantity
and quality are not usually examined when developing TMDL plans for impaired water
bodies for watersheds impacted by hydromodification.
The proposed modeling approach will allow resource managers to answer some key
resource development and management questions that have strong influence on
sustainable management of land and water resources. For example, how can managers
balance water availability and demand when watershed conditions are continually
changing? How can managers jointly forecast flow alteration and water quality
degradation due to hydromodification? How can resource managers identify allowable
levels of hydromodification using "what-if' scenarios? What ecological risks are
associated with different hydromodification categories and levels? What management
options are available to resource managers to mitigate the impacts of hydromodification?
-------
Answering these questions would require innovative modeling approaches that can
forecast hydromodification-induced stressors, such as flow alteration, water quality
degradation, and habitat alteration; and evaluate their impacts on the health of aquatic
ecosystems. Given the complex and the interacting nature of hydromodification-induced
stressors, a lack of holistic or integrative modeling approaches has been the reason for
our inability to manage land use, water quantity and quality, and health of aquatic
ecosystems jointly and sustainably. The objective of this study is to show how BASINS
can be used to simulate and forecast the impacts of hydromodification on water quantity
and quality and discuss ways to assess the cumulative impacts that hydromodification-
induced stressors may have on the health of aquatic ecosystems.
-------
2 HYDROMODIFICATION: SOURCE OF INTEGRATED
STRESSORS
Hydromodification describes land and water resources development activities that are
driven by human population growth and resource consumption. These activities often
produce direct or indirect changes to water quantity and quality. USEPA (1993) defines
hydromodification as the "alteration of the hydrologic characteristics of coastal and non-
coastal waters, which in turn could cause degradation of water resources." According to
USEPA (2007), hydromodification consists of channelization and channel modification,
construction of dams and impoundments, and streambank and shoreline erosion. In the
literature, hydromodification has also been narrowly defined as hydrograph modification.
USEPA (2007) presents hydromodification as a leading source of water quality
impairment for streams, lakes, estuaries, aquifers, and other water bodies in the United
States. The National Water Quality Inventory Report to Congress (2004) that was
released in 2009 identified agricultural nonpoint source (NFS) pollution as the primary
(48%) water quality impairment of assessed streams and rivers followed by
hydromodification (20%), and habitat alteration (14%) (USEPA, 2009). Figure 1 shows
the top ten sources of impairment in assessed rivers and streams in the United States.
They are closely linked to human activities that alter the physical structure or the natural
function of a water body. Water quality degradations caused by hydromodification
include increased sedimentation, higher water temperature, lower dissolved oxygen,
degradation of aquatic habitat structure, and loss offish and other aquatic populations.
-------
Miles
Agriculture
Hydromodification
Unknown
Habitat Alteration
Natural/Wildlife
Municipal Discharges/Sewage
Unspecified Nonpoint Source
Atmospheric Deposition
Resource Extraction
Urban Runoff/Storm water
i
5 10 15 20 25 30 35 40 45
Percent of Impaired Stream Miles Affected
Figure 1. Top 10 sources of impairment in assessed rivers and streams in the United States (Source:
USEPA, 2009)
Adding to USEPA's narrow definition of hydromodification, we define
hydromodification more broadly to include urbanization, climate change, water
withdrawals, and inter-basin transfers. Our intention is to use the term for a wide range of
anthropogenic watershed disturbances that alter natural flow regimes and degrade water
quality. Addressing the impacts of integrated stressors is more effective than addressing
stressors individually one at a time because integrated stressors have integrated effects
that are not independent. Furthermore, many watersheds are impaired by integrated
stressors and resource managers are unable to identify the cause of impairment. Despite
being a major source of impairment in assessed water bodies, modeling approaches that
consider the impacts of hydromodification on water quantity and quality are not
available. The following sections present selected hydromodification categories and
introduce our proposed approach to modeling each category. Note that water quantity,
flow, streamflow, and water availability are used interchangeably throughout this report.
-------
3 HYDROMODIFICATION CATEGORIES AND MODELING
CHALLENGES
3.1 Urbanization
As the total impervious area in a watershed increases, peak flow rates and flow
volumes increase (Arnold and Gibbons, 1996; Tang et al., 2005) and baseflow and
groundwater recharge decrease (Rose and Peters, 2001). Such alterations of natural flow
regimes affect the distribution of surface water and baseflow components of streamflow.
Hydrologic imbalances caused by urbanization have serious consequences for water
availability. In many parts of the world, incidence of water supply shortages due to land
use change have been reported for communities in water-rich areas (Okun, 2002). In the
United States, frequent, severe droughts and urbanization-induced water shortages have
been observed in some parts of New England, in the southeast (Atlanta), and in areas in
the west coast (Seattle and Portland) (Sehlke, 2004).
Urbanization not only alters natural flow regimes, it also degrades water quality. For
sub-basins in east of Melbourne, Australia, Hatt et al. (2004) found that water quality
loads were correlated with imperviousness and drainage connections. Other studies have
linked urbanization and associated imperviousness to increased sediment, bacteria, and
nutrient loads (Schueler, 1995; Gove et al., 2001; Mallin et al. 2001
Our approach to forecasting the impacts of urbanization on water quantity and quality
is presented in Part 2 of this report.
-------
3.2 Water Withdrawals and Interbasin Transfers
Water withdrawn directly from rivers and streams alters the natural flow regime.
Withdrawals and interbasin transfers are water management options that allow managers
to balance water demand and water availability by issuing water withdrawal permits.
However, without knowing the available water levels under future development
scenarios, issuing water withdrawal permits would not balance water demand and water
availability. It may, however, lead to violation of downstream environmental flow targets
and minimum flows required by the Endangered Species Act (ESA). Managing the
effects of water withdrawals and interbasin transfers not only requires the use of
hydrologic and water quality models to forecast scenario-specific water quantity and
quality changes, but also water allocation models to ensure that sufficient quantity and
quality of water is available to all water users. In general, water allocations are simulated
with state-specified models that have no hydrologic and water quality simulation
capabilities. These models include the Water Rights Analysis Package (WARP) used in
Texas (Wurbs, 2005) and MODSIM: River Basins Management Decision Support
System used in Colorado (Labadie, 2005).
Our approach to modeling the impacts of water withdrawals on streamflow uses
HSPF to simulate water quantity and quality projections for alternative future scenarios.
Under each scenario, the effects of water withdrawal levels on downstream flow targets
and water quality impaired streams are evaluated, and the scenario that most closely
matches water availability with water demand and minimizes the overall impacts of
hydromodification is selected.
10
-------
3.3 Channel Modification and Streambank Erosion
Streams and rivers, which are important habitats to many aquatic organisms, are
affected by channel modification. Channel modifications include direct channel
operations such as dredging, widening, and straightening, or indirect modifications
caused by flow alteration. Today, many streams and rivers are degraded by
hydromodification-induced stressors such as flow alteration, unsanitary discharge, and
channelization projects (Leblanc et al.,1997). A number of investigators have examined
the ecological impacts of hydromodification by assessing the biotic integrity of streams
using multi-metric indices, such as the indices of biotic integrity (IBI) (Karr 1991;
Fitzpatrick et al. (2004) or bioindicator approaches (Adams, 2005). To link urbanization-
induced stressors to stream habitat degradation, a number of investigators have examined
relationships between total impervious area (TIA) and biological integrity (Morse et al.,
2003; Booth et al. 2004). Other investigators have related hydrologic metrics (estimated
from simulated or observed streamflow data) to biological integrity (Claussen and Biggs,
1997; Booth et al., 2004; Konrad et al., 2005). For example, Richter et al. (1996)
identified flow magnitude, frequency, duration, timing, and rate of change as ecologically
relevant hydrological indicators.
Our approach to modeling channel modification and Streambank erosion consists of
three steps. First, modelers use HSPF to simulate changes in water quantity and quality
due to hydromodification. Second, modelers develop hydrologic indicators (e.g., Q2)
from simulated long-term streamflow time series. In general, Q2 is defined as the flow
that corresponds with the two-year recurrence interval. Third, modelers evaluate the
impacts of simulated stressors such as increased flow velocity, shear stress, and
11
-------
streambed and streambank erosion on stream habitat quality. Specifically, our approach
to modeling the impacts of channel modification on stream habitat is to use simulated
hydrologic, hydraulic, and water quality metrics that link hydromodification-induced
stressors to stream habitat quality. In addition, hydromodification impacts on stream
habitat can be evaluated by linking HSPF to a habitat suitability model such as
PHABSIM (Milhous et al. 1984).
3.4 Flow Regulation by Dams and Impoundments
Trends in urbanization, population growth, and increased water demand and usage
have led to extensive damming of rivers and streams. In the United States, more than
85% of the inland waterways are now artificially controlled (NRC, 1992), including
nearly 1 million km of rivers that are affected by impoundments (Echeverria et al. 1989).
Dams and impoundments control flooding, generate electric power, and provide
irrigation, navigation, recreation, and municipal water needs, but in some cases, their
benefits to society are outweighed by their adverse environmental impacts. Dams and
impoundments cause flow alteration, and inundate wetlands and riparian areas. They also
tend to reduce or eliminate downstream flooding, block fish migration routes, increase
turbidity and sedimentation during construction, and retain sediment after construction.
Flow regulation by dams and impoundments profoundly alters natural flow regimes,
which results in degraded river ecosystems (Ward and Stanford, 1995; Ligon et al. 1995;
Power et al., 1996). Fishery managers have long argued that maintaining the pre-
development natural flow regime is essential to the composition and structure of native
riverine ecosystems and associated biodiversity (Richter et al. 2000). To mitigate the
ecological impacts of flow regulation, several mitigation measures have been proposed.
12
-------
These measures include: establishing minimum flow releases (Colby, 1990), offering
controlled flushing flows (Collier et al 1997), or maintaining natural flow regimes to flow
levels observed before the regulation project (Stanford et al., 1996; Poff et al. 1997).
Our approach to modeling the impacts of flow regulation on water quantity and
quality is to set up a flow regulation scenario by placing a hypothetical dam and
impoundment at different locations in the watershed. The first step is to build an HSPF
hydraulic function table known as the FTable with the desired elevation-area-storage-
discharge relationships to represent the dam and the impoundment. To evaluate the effect
of flow regulation on natural flow regimes, modelers can compare pre-regulation and
post-regulation observed and simulated streamflows for each future scenario. Based on
these comparisons, they select scenarios that maintain the pre-development hydrologic
condition of the watershed. Resource managers can also examine levels of stressors
associated with different flow regulation scenarios to select a scenario that minimizes the
integrated effects of flow regulation on water quantity, quality, and the health of aquatic
ecosystems.
3.5 Climate Change
General circulation models (GCMs) are used to make future climate change
projections that account for increasing levels of CC>2 in the atmosphere. Mean global
surface temperature is expected to increase in the range of 1.5 to 5.8 °C by 2100
(Houghton et al., 2001). For the United States, mean temperature and precipitation
projections are about 4.5 °c increase in temperature and 7.5 % increase in precipitation. In
general, GCM model projections are used to estimate changes in precipitation and
temperature. Four commonly used GCM models are: The Goddard Institute of Space
13
-------
Studies (GISS), Geophysical Fluid Dynamics Laboratory (GFDL), the United Kingdom
Meteorological Office (UKMO), and the Oregon State University (OSU) models. Climate
change projections vary across regions, but most regions are projected to have increased
frequency of intense storms, increased soil erosion and sedimentation, and increased sea-
level. Consequently, climate change can be expected to have serious effects on water
quantity, water quality, and the health of aquatic ecosystems.
Our approach to modeling climate change impacts on water quantity and quality is to
use GCM temperature and precipitation projections. Model users can select scenarios that
are similar to mean GCM projections for the United States or scenarios that differ from
mean GCM projections. For example, a modeler could select a scenario that has 10%
increase in annual precipitation, with a 10% increase in the frequency of high
precipitation events, and increased return frequency of storms of particular magnitudes.
For temperature increases, the modeler could select a corresponding scenario with two
degree increases during the cool season months, and four degree increases during the
warm months.
Note that climate change is only one driver of hydromodification, and it can be
addressed separately or concurrently with urbanization, flow regulation, and channel
modification. For more discussion on generating climate change scenarios for the HSPF
model in BASINS, interested readers should refer to the Climate Assessment Tool (CAT)
manual, which can be accessed at
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=203460.
14
-------
4 THE NEED FOR HYDROMODIFICATION MODELING
FRAMEWORK
The proposed hydromodification modeling approach makes the watershed the unit of
management and regulatory focus. This is based on the concept that every land area
belongs to a watershed and that the integrated effects of all hydromodification activities
that occur in a watershed have a measurable impact at the watershed outlet or at some
downstream point of interest. Watershed-based approaches also address the complexities
in modeling hydromodification impacts and allow resource managers to forecast
scenario-specific hydromodification stressor levels, evaluate their impacts on water
quantity and quality, and develop management plans that mitigate the impacts of
hydromodification-induced stressors. The proposed approach is applicable to restoration
efforts, but is more suitable for protecting least developed watersheds or pristine
watersheds that require judicious management and development protocols. To guide
management decisions that protect or restore watersheds, we present some key
hydromodification related questions that resource managers must answer before
hydromodification project plans are implemented in a watershed (Figure 2).
The watershed-based concept presented in Figure 2 resembles the TMDL process in
the sense that TMDLs are intended to reduce pollutant loads to levels that can be
assimilated by a water body in order to meet EPA's water quality standards. In general,
the TMDL process has a very limited scope because it addresses water quality only, and
not water quantity. In addition, TMDLs often target a single stressor and a single water
body. Unlike TMDL approaches, the approach described in this report assesses a
watershed's capacity to assimilate not only pollutant loads, but also all
15
-------
hydromodification-related stresses. Managing hydromodification impacts on water
quantity and quality at the watershed level has many benefits but it requires cooperation
among stakeholders. Cooperation among stakeholders is essential because watershed
boundaries often cross multiple jurisdictions. The New Jersey Department of
Environmental Protection (NJDEP) recognized that pollutant loads, water withdrawals,
and land use required new approaches that could not be addressed by regulatory
programs alone (EPA, 2004). Regulatory programs such as TMDLs have limited scope
and cannot be used to address integrated stressors such as flow alteration, habitat
alteration, and water quality degradation.
The proposed approach was developed in recognition of the need for comprehensive
watershed management approaches that link economic development to ecosystem
sustainability. For instance, using the proposed modeling approach, resource managers
can identify where to institute land preservation, and where to place hydromodification
projects in a watershed (Figure 2).
16
-------
Can land use planners, resource managers, and
regulatory staff agree on allowable stressor levels before
implementing hydromodification project plans in a
watershed?
Water Withdrawal
Dam and Impoundment
What management
options are
available to
mitigate the
impacts of
hydromodification
projects at the
watershed scale?
Urbanization
Channel
Modification
What levels of flow alteration and water
quality degradation or loss of ecosystem
services due to hydromodification can be
assimilated by a watershed?
Figure 2. Hydromodification activities and watershed management questions that resource managers
must answer before hydromodification projects are implemented.
17
-------
4.1 Essential Components of a Hydromodification Modeling Framework
Figure 3 illustrates a conceptual hydromodification modeling framework that consists
of four components. The first component characterizes present and future
hydromodification levels using "what-if' scenarios that have different percentages of
urban land and percent impervious cover, different distributions of land use across a
watershed, number and location of dams and water impoundments, and water withdrawal
and inter-basin transfer amounts.
The second component identifies and quantifies scenario-specific levels of
hydromodification-induced stressors, i.e. flow and habitat alterations and water quality
degradation, and assesses how integrated stressors may affect the health of aquatic
ecosystems.
The third component identifies and selects best management practices (BMPs) that
mitigate the impacts of hydromodification on water quantity and quality. Mitigating the
impacts of hydromodification requires a watershed-scale hydromodification management
plan that may include of BMPs for controlling runoff at the source (i.e. those that
enhance infiltration). Local governments in coastal California and other western states
developed hydromodification management plans to control nonpoint source pollution
from urban watersheds.
The fourth component uses adaptive management to iteratively evaluate how model
simulation results agree with data obtained after future scenarios are implemented in the
watershed. By continuously updating the model and comparing scenario-specific model
forecasts with post-implementation observed data, resource managers can determine
18
-------
when stressor levels reach unsustainable levels and modelers can validate model
simulations and minimize predictive uncertainties..
1. Characterize present
condition (baseline) and project
future scenarios
4. Match model simulation
results and post-
development observed data
(adaptive management)
2.Quantify hydromodification-
induced stressors
3. Manage the impacts of
hydromodification-induced stressors
(hydromodifcation management plan)
Figure 3. Components of a hydromodification modeling framework
19
-------
4.2 BASINS Modeling Framework: A Background
The Clean Water Act was established to restore and maintain the chemical, physical,
and biological integrity of the nation's waters. In the 1970s and 1980s, EPA successfully
regulated point source pollution through the National Pollutant Discharge Elimination
System (NPDES) permit program. However, managing nonpoint source pollution from
terrestrial ecosystems has proven to be more difficult. To manage this problem
efficiently, EPA adopted an approach that makes the watershed the unit of regulatory
focus (USEPA, 1998; Whittemore and Beebe, 2000). Adoption of the watershed
approach as the management and regulatory unit has created a need for watershed models
and modeling approaches. Today, watershed models are widely used to manage nonpoint
source pollution, particularly through the development of TMDL plans for water quality
impaired water bodies. To manage land and water resources in a sustainable manner,
resource managers need modeling approaches that also forecast the impact of
hydromodification on water quantity and quality. In 2007, EPA released a guidance
document on managing nonpoint source pollution caused by hydromodification (USEPA,
2007). This document listed a number of models applicable to hydromodification
modeling, but did not present specific hydromodification modeling guidance. HSPF and
AQUATOX, which are part of the BASINS modeling framework, were among the
models listed.
In 1996, USEPA released BASINS ver. 1.0 modeling and decision support system
(EPA, 1996a). BASINS integrates Geographic Information System (GIS) tools, national
databases (elevation, hydrography, meteorological, land use, and soil), assessment tools
(target, assess, and data mining), data management and graphing programs (WDMUtil
20
-------
and GenScen), models (HSPF, SWAT, PLOAD, and AQUATOX), and analysis tools
including the Climate Analysis Tool (CAT).
Figure 4 matches BASINS' modeling capabilities with hydromodification drivers,
stressors, and impacts. As shown in Figure 4, using GIS tools and databases, BASINS
provides access to information about soils, topography, and land use and land cover of a
watershed. In addition, BASINS provides information on hydromodification projects
already present in the watershed and their distribution in the landscape. Resource
managers can use BASINS to identify priority areas for preservation and development.
As stated earlier, hydromodification activities are driven by population growth and the
accompanying need for resource extraction and consumption. The watershed
management goal is to minimize the impacts of economic development projects on water
quantity and quality by reducing hydromodification-induced stressors to levels that can
be assimilated by the watershed or mitigated through BMPs. Here, we briefly discuss
some of the models available in BASINS that apply to modeling hydromodification.
21
-------
Hydromodification
(Drivers)
Channel modification
Climate change
Resource extraction
Dam and impoundment
Urbanization
Water withdrawal
Integrated Stressors
(Consequences)
Flow alteration
Habitat alteration
(streambank erosion)
Water quality degradation
Integrated Effects
(Impacts)
Maintain healthy ecosystems to
achieve ecological sustainability
1 1 1
ASINS Modeling Framework _^ HSPF — + AQUATOX
)atabases, GIS tools, and Models) (CAT and BMPs)
Figure 4. Matching framework modeling capabilities with hydromodification drivers, integrated
stressors, and integrated effects.
HSPF is EPA's premier watershed hydrology and pollutant transport model
(Whittemore and Beebe, 2000) and is the core watershed model in BASINS. HSPF
simulates hydrologic processes and water quality for a range of types of user-defined
scenarios. Because of its extensive water quality simulation capabilities, HSPF is
frequently used for TMDL plan development for impaired water bodies. In addition to
water quality and hydrologic process simulations, HSPF also serves as a water allocation
model. It has been used to assess the effects of land use change on streamflow (Brun and
Band, 2000; McColl and Agett, 2007), effects of water withdrawal on streamflow
(Zariello and Reis, 2000), and effects of climate change on water quantity (Middlekoop et
al. 2000; Goncu and Albek, 2009). Furthermore, HSPF has been used for developing
hydrological and biological indicators of flow alteration in the Puget Sound low land
streams (Cassin et al. 2005). Although HSPF is the core model, BASINS has other
models and tools that are applicable to modeling hydromodification. Models and tools
that are relevant to the objectives of this study are AQUATOX, climate analysis tool
(CAT), and HSPF's BMP Toolkit. Whittemore and Beebe (2000) reviewed BASINS and
22
-------
noted that its success would depend on efforts to share technical experiences and
solutions to problems.
The BASINS Climate Analysis Tool generates climate change scenarios for HSPF.
CAT users need information on climate change projections for the region of interest
within the United States. Based on Global Circulation Models (GCMs) projections of the
region of interest, CAT users can develop climate change scenarios by changing
temperature, precipitation, and evapotarnspiration data. Note that CAT is not available in
early versions of BASINS, but is available in BASINS 4.
Representing best management practices and simulating their effectiveness in
controlling nonpoint source pollution is a desirable feature in regulatory models. In
general, models employed for TMDL plan development are not well-suited for plan
implementation because many models lack BMP representation and simulation
capabilities; Whittemore and Beebe (2002) emphasized these aspects of the HSPF model.
According to Endreny (2002), HSPF has no explicit simulation of storm sewer networks
and lacks the capability to simulate common water quality BMPs. A recently developed
HSPF BMP Toolkit represents some types of BMPs, and it can be used to evaluate their
effectiveness in controlling runoff and pollutant loads. HSPF BMP related limitations
have been addressed and HSPF is now one of the few publicly available watershed
models with capabilities to represent and simulate vegetative, storage, and infiltration
BMPs. Currently, the BMP Toolkit is a web-based tool and is not part of BASINS, but
modelers can easily work with the BMP toolkit to modify their model input data. .
AQUATOX is an ecological effects model that can be used to evaluate past, present,
and future direct and indirect effects from various stressors, including nutrients, organic
23
-------
wastes, sediments, toxic organic chemicals, flow and water temperature in aquatic
ecosystems (Park et al. 2008). AQUATOX uses HSPF output as input, including results
of simulations that include hydromodification-influenced stressors. Based on the
magnitudes of stressors generated by HSPF in different scenarios, AQUATOX can
simulate how hydromodification-induced stressors may affect biota in aquatic
ecosystems.
4.3 Hydromodification Decision-making Example
Figure 5 presents a flow chart describing how to apply the BASINS modeling
framework to watersheds that are likely to experience hydromodification. As an example,
we selected two hydromodification categories: urbanization and flow regulation by dams
and impoundments. For each, we present a list of hydromodification-induced stressors
and scenarios. As shown in Figure 5, the first step is to characterize the current
hydromodification levels, then determine if additional hydromodification projects are
planned for the watershed. First, modelers use HSPF to simulate streamflow and water
quality; modelers then calibrate and validate the model under present conditions. After
successful calibration and validation, model users then simulate flow and water quality
for the future scenarios. Comparisons of flow and water quality for pre-
hydromodification (current watershed condition) and post-hydromodification determines
flow alteration and water quality degradation levels associated with each scenario.
In the second step, after simulating flow and water quality for baseline and future
scenarios, model users set up HSPF as a water allocation model to allocate and track how
different hydromodification categories and scenarios affect water use and availability.
HSPF uses a schematic network of nodes and links to track changes in inflows, storages,
24
-------
and outflows over time. To minimize flow alteration impacts, resource managers may
seek to identify scenarios that closely match pre-hydromodification water availability
levels.
In the third step, model users simulate how alternative future may alter stream
habitats and water quality. Models in this case are used to evaluate how simulated
changes due to hydromodification are likely to affect existing TMDL allocations and/or
the health of aquatic ecosystems. As an example, modelers can employ the AQUATOX
model to explore whole-system stressor-response relationships. Modelers can also extract
hydrologic and water quality indicators or metrics from simulated streamflow and water
quality data, then use these metrics to explore system stressor-response relationships.
Based on simulated scenario comparisons, resource managers can select
hydromodification management plans that minimize undesirable impacts on aquatic
resources. Finally, to address model predictive uncertainties, resource managers may
employ an adaptive management approach to compare and update model simulations
based upon data collected after initiation of hydromodification projects.
25
-------
Characterize the current hydromodification level of the watershed
Maintain
current water
quantity an
quality
condition of the
watershed
Are there any
planned
hydromodification
projects?
Develop a baseline
model for the watershed
What type of hydromodification projects
are planned for the watershed?
Construction of Dams and Impoundments
Management Plan
Evaluation
Are flow alterations and
water quality
degradation controlled?
Are water availability
and demand balanced?
Are ecosystem services
maintained at prepoject
levels?
Future build-out scenanos
Set hypothetical
"what-if build-out
scenarios with
different total
impervious area
(TIA) levels
Future scenarios
Build a new dam/reservoir
Test different water withdrawal
levels
Test different reservoir
operations and water allocation
options
Hydromodifi
-cation
impacts are
minimized
Stressor forecasting (HSPF Model)
Forecast flow alteration
Forecast water quality degradation
Forecast habitat alteration
\/ Forecast water availability
\/ Forecast water demand
Link HSPF to AOUATOX Model
Assess the ecological risks associated
with alternative future scenarios
Hydromodification Management Plan
Simulate HSPF with/without BMPs,
select suitable best management practices
(BMPs) using HSPF BMP Toolkit
Adaptive Management
Select different scenarios and re-evaluate the
decision-making process
Figure 5. Decision-making framework for watersheds receiving hydromodification projects
26
-------
4.4 Linking Integrated Stressors to Aquatic Ecosystem Impacts
As stated earlier, AQUATOX is an ecological risk assessment model, which is part of
the BASINS modeling framework. AQUATOX simulates the effects of multiple
simultaneous Stressors that may include nutrients, organic toxicants, temperature,
suspended sediment, and flow, on the health of aquatic ecosystems. Potential applications
of the AQUATOX model include estimation offish recovery after pollutant loads are
reduced, ecosystem responses to invasive species, effects of mitigation measures, and
changes in ecosystem services. By forecasting hydromodification-induced Stressors using
HSPF as the load-generating model, and assessing the resulting ecological impacts of the
simulated Stressors with AQUATOX, resource managers can evaluate the biological
ramifications of hydromodification projects before the projects are implemented.
Although linking integrated Stressors to aquatic ecosystem impacts is beyond the scope of
the current study, our suggested approach to establishing stressor-response relationships
is to link HSPF to the AQUATOX model.
4.5 Model Application Example
In Part 2 of this report, we present a case study demonstrating how BASINS can be
used to forecast urbanization-induced Stressors, namely flow alteration and water quality
degradation under various urban development scenarios. We include in this example a
brief discussion of how urbanization-induced Stressors affect stream channel erosion. We
also attempt to link urbanization-induced Stressors to impacts on the health of aquatic
ecosystems using hydrological indicators or metrics. As a test watershed, we selected a
27
-------
medium-sized headwater watershed, which is a tributary of the Yaquina Watershed in
Oregon, USA.
28
-------
PART 2: APPLICATION EXAMPLE
5. FORECASTING URBANIZATION IMPACTS ON WATER
QUANTITY AND QUALITY: CASE STUDY OF THE YAQUINA
WATERSHED, OREGON, USA
Abstract: Protecting ecosystem services provided by headwater watersheds increasingly
is becoming an important land and water management objective. To allow resource
managers to minimize future watershed degradation and eliminate the need for
restoration, we present a method to forecast potential impacts of urbanization before
urban development plans are implemented in a watershed. The method establishes both a
baseline that represents the pre-development condition, and build-out scenarios that
represent future development. In this report, we arbitrarily selected three future build-out
scenarios that represent 15, 45, and 85 percent total impervious area (TIA), where
impervious cover was used as a landscape indicator to represent the effects of
development. We employed the Hydrological Simulation Program - FORTRAN (HSPF)
watershed model to simulate streamflow, total suspended solids (TSS), and nitrate in the
baseline scenario and for the three build-out scenarios. Comparisons of simulated results
in the baseline scenario to those forecasted for the three build-out scenarios show
increased peak flows and decreased baseflows for the build-out scenarios. Nitrate
simulation results show increased nitrate concentrations for the three build-out scenarios.
Suspended sediment concentrations were forecasted to increase with increasing TIA from
15 to 45 percent, but the 85 percent TIA scenario resulted in lower TSS concentrations.
Our forecasts also indicate that mean channel wetted width, flow depth, and flow velocity
decrease with increasing percent TIA. The proposed method serves as an exploratory
29
-------
approach in which resource managers and land use planners forecast urbanization-
induced stressors before urban development plans are implemented in the watershed. The
impacts of these stressors on aquatic ecosystems and services can then be simulated.
30
-------
5.1 Introduction
For nearby receiving water bodies, urbanization has been found to lead to altered
hydrologic regimes, deteriorated water quality, losses of habitat and biodiversity, beach
closings, fishery declines, and fish consumption advisories (Nixon, 1995; Richardson,
1997; Rabalais et al., 1996; Boesch at al., 2001; Elofson et al., 2003; Niemi et al., 2004).
Many of the adverse ecological impacts of urbanization are closely linked to increases in
impervious area (Paul and Meyer, 2001; Allan, 2004). The percent of total impervious
area (TIA) is that portion of a watershed area covered by built surfaces such as paved
roads, parking lots, sidewalks, driveways, and rooftops. Percent TIA in a watershed has
been used as a gross indicator of urbanization and associated impacts on streamflow (Ng
and Marsalek, 1989), water quality (Schueler, 1994; Arnold and Gibbons, 1996; Conway,
2007), and the health of aquatic ecosystems (Klein, 1979; Morse et al., 2003).
In general, as the percent TIA increases, the fraction of precipitation that is able to
infiltrate the soil and recharge groundwater decreases, and the fraction that becomes
overland runoff increases (Schueler, 1994). Without major mitigation measures,
urbanization causes reduced baseflow and declining water tables, and increased flood
flow magnitude and frequency (Arnold and Gibbons, 1996; Paul and Meyer, 2001), with
resulting changes in stream channel morphology and in-stream suspended sediment from
the increased scouring, and consequent stream habitat degradation (Booth, 1990; Paul
and Meyer, 2001).
Urbanization also causes increased nutrient (Creed and Band, 1998; Wernick et al., 1998;
Donner et al., 2004), suspended sediment (Nelson and Booth, 2002; Wotling and
Bouvier, 2002) and other pollutant inputs (e.g. metals, pesticides, road salts) from
-------
terrestrial areas to receiving water bodies, along with increased water temperature (Niemi
et al., 2004). Such urbanization-induced changes cause stresses to aquatic organisms that
impair the overall health of aquatic ecosystems. Increased nitrate export to coastal waters
leads to estuarine eutrophication, with changes in biotic community structure and
diversity (Turner and Rabalais, 1991; Vitousek et al., 1997; Boesch et al., 2001).
Compton et al. (2003) found positive correlation between nitrate concentrations and
broadleaf cover dominated by red alder (alnus rubra) in forested watersheds in the
Oregon Coast Range. Increased sediment inputs cause reduced light penetration and may
directly impede the reproductive processes of aquatic organisms (Wotling and Bouvier,
2002; Nelson and Booth, 2002).
Land use planners and decision-makers need guidance on how to forecast the impact
of urban development on hydrologic processes (e.g., increased flood frequency and loss
of groundwater recharge), on nutrient and sediment concentrations, and on ecological
integrity (e.g., loss of biodiversity and wildlife habitat). Much of what is known about the
effects of urbanization-induced stressors on aquatic ecosystems was obtained through
field observations (Morse et al., 2003; Konrad and Booth, 2005). Monitoring urban
streams alone cannot, however capture the full impact of urbanization because impacts
observed in urban streams are often moderated by existing best management practices
(BMPs). In addition, monitoring urban streams usually does not account for impacts that
have occurred during construction but before mitigation measures were fully
implemented. Stream monitoring is a reactive approach that offers limited insights to land
use planners and decision-makers. The ability to forecast future conditions reliably in
32
-------
urban streams before implementation of development plans obviously would be more
helpful.
Many investigators acknowledge the interdependence of structure and function of
stream ecosystems and natural flow variability (Poff and Allan, 1995, Power et al., 1996,
Richards et al., 1997). For example, flow alterations often result in changes to the
ecological organization of aquatic and riparian systems that lead to changes in physiology
and behavior of individuals, populations, community composition, and food web
structure (Poff et al., 1997; Bunn and Arthington, 2002; Poff et al., 2006). Building on
such established relationships, hydrologic metrics with strong ecological relevance have
been developed for natural flows (Poff and Ward, 1989; Poff et al., 1997) and for altered
flows (Richter et al., 1996; Olden and Poff, 2003). However, approaches that link
urbanization-induced stressors explicitly to specific ecological effects, such as loss of
aquatic life or physical habitat, are lacking.
A proactive approach is needed for forecasting urbanization-induced stressor levels
and establishing links between them and the resulting condition of aquatic ecosystems.
An alternative to monitoring streams to assess the ecological impacts of urbanization is to
forecast urbanization impacts with hypothetical build-out scenarios or actual
development plans. Using hypothetical build-out scenarios, resource managers employ
comprehensive watershed models to forecast urbanization-induced stressor levels before
implementation of urban development plans and any associated mitigation measures in a
watershed. We propose herein using the Hydrological Simulation Program - FORTRAN
(HSPF) (Bicknell et al., 2001) to simulate streamflow, nitrate, TSS, and channel
morphology changes.
33
-------
The objective of this study is to present a modeling approach to forecast potential
impacts of alternative urban development scenarios by using watershed total percent
impervious area as an indicator of urban development impacts on hydrology, flow width,
flow depth, flow velocity, nitrate concentration, and TSS concentration.
5.1.2 Watershed Description
The study watershed is a headwater tributary located upstream of the United States
Geological Survey's gaging station near Chitwood, Oregon (44° 39' 29" N, 123° 50' 15"
W) (Figure 6). The study watershed is characterized by wet winters, relatively dry
summers, and mild temperatures that are typical of the Coastal Range areas of Oregon.
The long-term average annual precipitation for Newport is 1767 mm. Precipitation
rainfall comes from moist air masses from the Pacific Ocean. A large portion of the
precipitation occurs in November, December, and January. Conversely, the warmest and
driest months are July, August, and September (Figure 7).
34
-------
USGSGage Elevation (m)
1 Yaqmna River MM High : 811
Streams
I Study Watershed Low : 13
Figure 6. Location of the study watershed.
35
-------
350
8
Q_
100 --
50 --
Average Monthly Precipitation
Average Monthly Temperature
n.n
16
14
12
10
o
8 !
o
Q.
E
o
6
4
2
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Month
Figure 7. Distribution of long-term monthly precipitation and temperature for Yaquina
Watershed
The watershed has a drainage area of 184 km2. Presently, the study watershed is 97
percent covered by evergreen forest, about two percent covered by agricultural land, and
about one percent covered by medium density urban residential land (Table 1). The
dominant land cover is coniferous forest because the original complex forest has been
replaced by single species silviculture and opportunistic pioneer species. In riparian areas
along streams, disturbed sites are frequently occupied by pioneer broad-leaf trees (e.g.,
red alder) (Ohmann and Gregory, 2002).
The dominant soil of the study watershed is Bohannon soil series found on the
western slopes of the Oregon Coast Range areas. The soils of the watershed are generally
36
-------
well drained with poorly developed horizons (Ohmann and Gregory, 2002 ). Geological
formations of the watershed contain massively to thinly-bedded tuffaceous siltstones,
sandstones, basalt breccias, and augite-rich tuff that provide a constant source of silica
and suspended sediment (Snavely and Wagner, 1963).
Table 1. Baseline and hypothetical future land use scenarios
Watershed Area (Hectares) by Land use Category Impervious Area at
Build-out (%)
Scenarios Forest (ha) Agriculture (ha) Urban (ha)
Baseline 18057 261
Scenario 15 12946 261
Scenario45 7443 261
ScenarioSS 106 261
25
5136
10639
17976
15
45
85
37
-------
5.2 Methods
The method proposed to forecast urbanization-induced stressors consists of the following
steps. More detailed discussion of each step is given in subsequent parts of the methods
section.
Step 1: Characterize the current land use, hydrology, and water quality of the watershed
(baseline analysis and characterization).
Step 2: Simulate flow, TSS, and nitrate under baseline conditions.
Step 3: Calibrate and validate flow, TSS concentration, and nitrate concentration
simulated by the model for the baseline condition using observed flow, TSS, and nitrate
data.
Step 4: Project future development scenarios using actual land use plans if available or
hypothetical build-out scenarios if plans are not available for the watershed.
Step 5: Simulate flow, TSS, and nitrate for selected future build-out scenarios using the
model parameters calibrated under the baseline condition.
Step 6: Develop urbanization-induced stressor indicators from simulated flow, TSS, and
nitrate data for each scenario. To establish cause and effect relationships between
forecasted urbanization-induced stressors and their ecological impacts, we used
hydrological alteration indicators. For each scenario, we estimated 7Q10 (7-day average
low-flow with a recurrence interval of 10 years), bankfull flow, baseflow index (total
volume of base flow divided by the total volume of runoff for a period) (Wahl and Wahl,
1995), and indicators of hydrological alteration (IHA) from scenario-specific simulated
streamflow. Indicators of hydrologic alteration (IHA) captures five flow characteristics
38
-------
that are related to biological integrity: magnitude, duration, frequency, timing, and rate of
change (Richter et al., 1996). Other flow-related forecasted physical habitat indicators
include shear stress, depth, and flow velocity. In addition, we estimated nitrate and TSS
concentrations for each scenario. Calculating indicators from forecasted streamflow, TSS,
and nitrate offers to land use planners an exploratory approach where potential ecological
impacts of urbanization can be predicted before urban development plans are
implemented in a watershed.
Step 7: Establish linkages between urbanization-induced stressors and ecological impacts
(hydrologic, hydrologic, and water quality indicators)
8. Link HSPF to AQUATOX model and assess the ecological impacts of build-out
scenarios
5.2.1 Model Setup and Input Data
HSPF is a calibrated-parameter model which uses observed data for its calibration
and validation. For this study, observed streamflow, TSS, and nitrate data were obtained
at the Chitwood gaging station (USGS 14306030). Daily streamflow data from 1972 to
1991 were obtained from a USGS website (USGS, 2007), supplemented by data from the
Oregon Water Resources Department that is available at
http://www.wrd.state.or.us/OWRD/SW/streamflow_midco.shtml. Daily observed
streamflow averages 7.1 m3/sec, although it can vary from 0.06 m3/sec during late
summer low flow conditions, to 186 m3/sec at peak flow conditions in late fall and winter
(October to March).
Nitrate data were collected weekly from December 1999 through December 2001, and
monthly from January 1999 to December 2002, at the USGS stream gaging station near
39
-------
Chitwood. Water samples collected for dissolved nutrient measurements were filtered
(Cole Farmer 142 mm nylon filter 0.45 jim membranes) and frozen for later analysis by
the Marine Science Institute (MSI), University of California, Santa Barbara, CA using a
Lachat QwickChem 8000 Autoanalyzer for simultaneous determination of nitrite and
nitrate + nitrite. Analytical information, including blank procedures, sample replicate
results, and other quality assurance details are available at www.msi.ucsb.edu/Analab.
Because nitrite generally comprised less than 2% of the nitrogen species at these sites, we
treated nitrate + nitrite as nitrate. Daily TSS was obtained from a USGS website (USGS,
2006); it was available for the period from October 1, 1972 through September 30, 1974.
The Hydrologic Simulation Program - FORTRAN (HSPF), part of U.S.EPA's Better
Assessment Science Integrating Point and Nonpoint Sources (BASINS) (USEPA, 2001),
was selected to simulate streamflow, TSS, and nitrate for a baseline or reference scenario
and build-out scenarios. HSPF is a comprehensive, conceptual, continuous simulation
model that simulates flow and water quality constituents originating from pervious and
impervious land surfaces, in streams and well-mixed impoundments (Bicknell, et al.,
2001). HSPF uses the PERLND and the IMPLND application modules to represent
processes that occur on pervious and impervious land surfaces, respectively, and it uses
the RCHRES module to represent processes that occur in water bodies. HSPF uses
BASINS to extract soil, land cover, and geomorphological data and parameter values
from geographic information databases, such as the Digital Elevation Model (DEM),
State Soil Geographic (STATSGO) database, and National Land Cover Data (NLCD)
databases, using the BASINS geographic information systems (GIS) analysis tools
(USEPA, 2001). The study watershed was delineated into eight subwatersheds; each
40
-------
subwatershed has its own land use, soil, and topography and is considered a
homogeneous hydrological response unit.
The model uses meteorological data such as precipitation, air temperature, potential
evapotranspiration, solar radiation, wind speed, dew point temperature, and cloud cover.
Our climate data were obtained from the Hatfield Marine Science Center in Newport,
Oregon (www.weather.hmsc.orst.edu), a weather station in Nashville, Oregon, and
BASINS databases. Potential evapotranspiration was estimated from meteorological data
using the Ham on (1961) method.
5.2.2 Model Calibration and Validation
After necessary input data were prepared, the model was calibrated using historical
streamflow, TSS, and nitrate data corresponding to the baseline. The parameter values
that gave the best calibration results for streamflow, TSS, and nitrate were retained. We
used those parameter values to forecast streamflow, nitrate, and TSS concentrations for
the three build-out scenarios. The hydrological calibration periods were 1973 through
1977 and 2000 through 2001; the nitrate calibration period was from October 2001
through September 2002; and the TSS calibration period was from October 1973 through
September 1974. We used one year for TSS calibration and the other for TSS validation.
Because both TSS and nitrate are affected by the hydrological calibration performance,
we performed two hydrologic calibrations. One coincided with the period when observed
TSS data were available and the other coincided with the period when observed nitrate
data were available. After successful hydrological, TSS, and nitrate calibrations, the
model was validated using observed streamflow data from periods that were not used for
41
-------
model calibration. We employed HSPF calibrations using observed and simulated daily
data, but we also did some monthly and annual calibration comparisons for the baseline
and the build-out scenarios.
5.2.3 Model Performance Evaluation Criteria
There are no widely accepted quantitative HSPF model calibration criteria to
determine if the model predictions are acceptable; there are, however, some generally
accepted HSPF calibration and validation guidelines. For most HSPF streamflow
calibrations, acceptable model calibration performance is achieved when the correlation
coefficient between monthly simulated and observed streamflow is greater than 0.85
(Donigian, 2002). Because sediment and nitrate calibrations are affected by the
hydrological calibration performance, the TSS and nitrate calibration criteria are far less
stringent than the hydrologic calibration criteria. Donigian (2002) states that HSPF
annual and monthly hydrology, sediment, and nutrient simulations are considered "good"
when the percentage differences between simulated and observed data are between 10
and 15 for flow, between 20 and 30 for sediments, and between 15 and 25 for nutrients.
HSPF calibration and validation procedures consist of matching simulated
streamflow, nitrate, and TSS concentrations with observed data. We used goodness-of-fit
measures such as coefficient of determination (R2), root mean square error (RMSE), and
coefficient of model-fit efficiency (E) (Nash and Sutcliffe 1970) to evaluate the model's
performance in predicting observed streamflow, TSS, and nitrate. The coefficient of
determination is written as:
42
-------
(i)
where Q0fo is the observed streamflow, Qsim is the corresponding simulated streamflow,
and Qobs and Qsimare average observed and average simulated streamflow, respectively,
and N is the number of data points used in the average calculation. The coefficient of
model fit efficiency can be written as:
N
/ , \(2obs ~ (2sim )
E=1.0--f - — — (2)
im ~\iobs>
where E is the Nash-Sutcliffe coefficient and other terms are as defined in (1). The root
mean square error (RMSE) is written as:
<2ofa-e_)2]05 (3)
where Q0fo and Qsim are defined as in (1) and N is the number of data points used for the
comparison period.
5.2.4 Future Build-out Scenario Development
The sequential nature of urban development projects introduces uncertainties in land
use projections, making it difficult to project changing land use across an entire
watershed. Land use change models that use economic and social drivers can project land
use change. These include the California Urban Futures Model (Landis, 1995) and the
Land Use Evolution and Impact Assessment Model (LEAM) (Deal, 2001). The LEAM
43
-------
model forecasts specific land use change projections using watershed-specific drivers,
such as economic, population, social, geography, transport, and open space.
Our approach can use information from a land use change model, but since land use
change information is not usually available at the watershed scale, we present a simple,
generic method that uses hypothetical build-out scenarios. In areas where information on
future land use change is available from existing zoning and master plans, watershed
managers can use actual land use plans instead of hypothetical scenarios. This allows
resource managers to set an allowable TIA level that corresponds to a watershed's
unknown future build-out. It assumes that a watershed will be developed eventually and
will have a threshold TIA at the end (build-out condition). Build-out is defined as an
estimate of the amount and location of potential development for an area. It is used by
land use planners who evaluate the potential impacts of urban development using build-
out analysis. Generally, build-out analysis, as currently used by land use planners, does
not employ watershed-scale models. The proposed method allows land use planners to set
different impervious area levels for build-out scenarios, and simulate how different
scenarios alter hydrology and water quality. Although there is no limit on the number of
build-out scenarios selected and impacts simulated, in this study we limited our analysis
to three build-out scenarios that correspond with 15, 45, and 85% total impervious levels.
Table 1 lists the area covered by each land use category and the percentages of TIA
used for the baseline (i.e., pre-development) and hypothetical future build-out scenarios.
The three build-out scenarios correspond to low density residential (scenariolS), high
density residential (scenario45), and commercial development (scenarioSS). Selection of
15, 45, and 85 percent TIA levels is arbitrary and these levels primarily demonstrate the
44
-------
model's sensitivity to different levels of imperviousness. ScenarioSS (in particular) was
derived only for comparative purposes since only a few urban watersheds ever reach such
high TIA levels. Impervious area may not reach 85% because some areas of the
watershed are unbuildable due to soil and slope characteristics or are zoned as committed
open spaces. The Ballona Watershed in Los Angles, CA is one of the few watersheds
with impervious cover that reaches 85%. Land use planners should select percent TIA
levels that realistically reflect their future land use plans so they can set threshold TIA
limits for a watershed.
For each scenario, we kept a minimum of 13 percent of the urban land area as
committed open spaces to include land occupied by detention basins, parks and other
recreational areas, wetlands, and riparian buffer zones. When developing scenarios with
this approach, watersheds can be subdivided, and different TIA levels can be assigned to
different subwatersheds. Questions to answer are what will be the total allowable
impervious levels at build-out for a watershed and what stressor levels are associated
with different build-out scenarios.
5.2.5 Forecasting Flow, Nitrate, and TSS Alterations
Modeling streamflow, nitrate, and TSS alterations under the three build-out scenarios
makes certain assumptions. For example, calibrated model parameter values obtained
from the baseline condition for each land use category were unchanged when modeling
the altered scenarios. In addition, to forecast streamflow, habitat, and water quality
conditions for the future scenarios, we converted existing pervious forest land to
impervious urban land — but retained the historical precipitation inputs for all simulations
since it is not possible to forecast future precipitation. The baseline and build-out scenario
45
-------
streamflow, TSS, and nitrate were all simulated for a 30-year (1973 to 2002) period. The
30-year streamflow, TSS, and nitrate projections for the three build-out scenarios reflect
the hydrologic and water quality degradations that would have been observed in the
watershed through 2002 if the watershed had reached 15, 45, or 85 percent urban TIA at
build-out in 1973.
As already noted, the 30-year simulations assumed that future precipitation and
climate would be similar to those observed over the previous 30 years. In addition, the
HSPF model treated all impervious areas as connected even though, in reality, not all of
them are. This latter assumption can lead to runoff-water and contaminant input over-
prediction because it ignores infiltration that occurs when runoff from an impervious area
passes over an adjacent pervious area.
5.2.6. Linking Stressors and to Impacts
To project the ecological impacts of urbanization, indicators of urbanization-
induced stressors were derived from forecasted flow data. For each scenario, we derived
indicators from long-term simulated flow, TSS, and nutrient data. For example, indicators
of flow alterations include the indicators of hydrological alterations (IHA) (Richter et al.,
1996), baseflow index, bankfull flow (Ch), and 7Q10. We estimated the baseflow index
using a recursive digital filter (Eckhardt, 2005); specifically, we examined indicators that
are closely related to peak flow and baseflow changes under the three build-out scenarios.
These indicators can be used to establish cause and effect relationships between
urbanization-induced stressors and changes in abundance, diversity, and fitness of aquatic
communities. To assess the ecological impacts of simulated urbanization-induced
46
-------
stressors, one can link HSPF to AQUATOX or to the Physical Habitat Simulation model
(PHABSIM), a major component of the Instream Flow Incremental Methodology (IFIM)
(Milhous et. al., 1984). Linking HSPF to aquatic ecosystem models and simulating
ecological impacts of urbanization-induced stressors with AQUATOX AND PHABSIM
is beyond the scope of this study.
47
-------
5.3 Results and Discussion
5.3.1 Model Calibration and Validation Results
Hydrologic Calibration and Validation. Statistical and graphical visualization
calibration tools available in BASINS were used to calibrate and validate HSPF for the
baseline scenario. Table 2 shows goodness-of-fit measures for the calibrations. For the
two hydrologic calibration periods, the results were within acceptable calibration range:
daily calibrations had coefficients of determination (R2) of 0.8 and 0.87, and monthly
calibrations had R2 values of 0.91 and 0.95. Validation performance was slightly lower
than that for the calibration. For the two hydrologic validation periods, the R2 values were
0.72 and 0.82 for the daily validations, and 0.83 and 0.89 for the monthly calibrations.
The RMSE and Nash-Sutcliffe Efficiencies (E) values are also provided in Table 2. As
shown in Figures 3a and 3b, simulated streamflow compared well to observed daily
streamflow for both hydrologic calibration periods. All the parameters adjusted as part of
the hydrologic, TSS, and nitrogen calibrations are listed in Table 3.
48
-------
Table 2. Model calibration and validation results
R2 RMSE E R2 RMSE E
Hydrologic Calibration and Validation
Calibration Period(1973 to 1977) Validation Period (1977 to 1979)
Daily 0.80 5.8 0.77 0.72 6.4 0.71
Monthly 0.95 2.2 0.94 0.89 3.0 0.87
Calibration Period (2000 to 2001) Validation Period (1999 to 2000)
Daily 0.87 3.6 0.82 0.69 7.8 0.62
Monthly 0.91 4.1 0.81 0.83 3.5 0.79
TSS Calibration and Validation
Calibration Period (1973 to 1974) Validation Period (1972 to 1973)
Daily 0.52 39.36 0.46 0.56 19.1 0.44
Monthly 0.89 11.78 0.82 0.78 7.7 0.59
Nitrate Calibration and Validation
Calibration Period (2000 to 2001) Validation Period (1999 to 2000)
Daily 0.75 — — 0.39 —
49
-------
Table 3. List of parameter values adjusted during calibration
Parameter
Hydrology
LZSN
UZSN
INFILT
INTFW
IRC
AGWRC
DEEPFR
BASETP
AGWETP
LZETP
CEPSC
Sediment Simulation
KRER
JRER
KSER
JSER
KGER
JGER
KSAND
EXPSND
W (cm/s)
M (kgAtil day)
TAUCD (kg^n2)
TAUCS (kg^n2)
Nitrogen Simulation
KTAM20 (per hour)
KTO220 (per hour)
TCNIT (per hour)
KNO320 (per hour)
TCDEN (none)
DENOXT (mg/1)
Description
Lower Zone Nominal Storage (cm)
Upper Zone nominal Storage (cm)
Infiltration parameter (cm/hr)
Interflow inflow parameter
Interflow Recession Parameter (per day)
Daily recession constant of groundwater flow (per day)
Fraction of groundwater inflow (inactive)
Fraction of E-T from baseflow
Fraction of E-T from active groundwater
Lower zone E-T parameter
Interception storage capacity (cm)
Coefficient-detachment equation
Exponent-detachment equation
Coefficient-detached washoff equation
Exponent-detached washoff equation
Coefficient-matrix scour equation
Exponent-matrix scour equation
Coefficient-sandload power function
Exponent-sandload power function
Fall velocity in still water
Erodibility coefficient
Critical bed shear for deposition
Critical bed shear for scour
Nitrification rates of ammonia
Nitrification rates of ammonia and nitrite
Temperature correction coefficient for nitrification
Nitrate denitrification rate at 20 degrees C
Temperature correction coefficients for denitrification
Dissolved oxygen concentration threshold for denitrification
Calibrated value
20.32
2.36
0.17
6.50
0.70
0.98
0.00
0.00
0.00
0.45
0.51
1.14
1.00
3.00
5.80
5.80
5.20
0.01
3.90
0.14
1.05
2.44
6.35
0.015
0.002
1.50
0.02
1.00
0.50
50
-------
Sediment Calibration and Validation. Suspended sediment concentration calibration
and validation performances generally are impacted negatively by the lack of available
long-term historical data with appropriate temporal resolution. In this study, only two
years of daily TSS concentration data were available for model calibration and validation.
As shown in Table 2, the coefficients of determination (R2) for the sediment calibration
were 0.52 for the daily and 0.89 for the monthly simulations. Sediment validations had
coefficients of determination (R2) of 0.56 for the daily and 0.78 for the monthly
simulations. Graphical comparisons of the observed versus simulated TSS concentrations
indicated that the model slightly under-predicted the higher TSS values, but overall
sediment simulations generally followed the runoff hydrograph trend (Figure 8c). Despite
under-predicting the higher TSS values, overall model calibration and validation
performances for TSS concentrations were acceptable.
Nitrate Calibration and Validation. Nitrate concentration simulation performance
depends on the quality and the quantity of observed data. Observed concentration data
had many missing data points, which created model calibration problems. To address this
problem, we compared observed and simulated nitrate only on days when observed
nitrate concentrations data were available. Despite using only limited observed versus
simulated comparisons, our nitrate concentration calibrations had coefficients of
determination (R2) of 0.75 for the daily simulations (Table 2). The nitrate validations had
lower coefficients of determination of 0.39 for daily simulations (Table 2). Lack of
continuous observed daily nitrate data at daily (or shorter) time intervals may have
contributed to the lower model validation performance. Both observed and simulated
51
-------
nitrate concentrations closely followed the seasonal hydrologic trend, except during the
first winter storm, suggesting that nitrate export from terrestrial areas is strongly
influenced by nitrate build-up and storm wash-out mechanisms (Figure 8d).
52
-------
Day
Rainfall (mm)
Observed
Predicted
Day
Day
Day
Figure 8. Model calibration results: streamflow (a and b), TSS (c), and nitrate (d) calibrations.
53
-------
5.3.2 Forecasting Stressor Levels for the Build-out Scenarios
Forecasting Flow Alterations. Flow alterations for build-out future scenarios show
significant changes as a function of rising percent of TIA levels (Figure 9). To document
these scenario-specific changes, we compared: daily streamflows simulated for the
baseline scenario to streamflows forecasted for scenariolS, scenario45, and scenarioSS
(Figure 9a); simulated annual water balances for all scenarios (Figure 9b); simulated flow
duration curves at low flow conditions, i.e., for flows with greater than 95% exceedances
(Figure 9c); and simulated flow duration curves at high flow conditions, i.e., for flows
with less than 5% exceedances (Figure 9d). Comparisons of the flow duration curves
show that low flows get smaller with increased percent TIA (Figure 9c) and high flows
get larger (Figure 9d). Comparisons of the annual average water balance components
under the baseline versus the build-out future scenarios for a 30-year simulation period,
using the same historical precipitation data, showed increased surface runoff and
decreased interflow, baseflow, and evapotranspiration for scenariolS, scenario45, and
scenarioSS (Figure 9b).
Peak Flow. The simulation results clearly show increased flow flashiness associated with
increases in percent TIA levels (Figures 4a and 4d); specifically, low intensity storm
events usually did not generate runoff for the baseline and scenariolS, but did so for
scenario45 and scenarioSS. Thus, the baseline and scenariolS were characterized by
smooth runoff hydrographs, whereas scenario45 and 85 exhibited high peak flow rates
(Figure 9a). The data suggest the existence of a threshold percent TIA above which peak
flows greatly increase with increased percent TIA level. In addition, comparisons of flow
54
-------
duration curves with exceedances less than 5% show small peak flow differences
between the baseline scenario and scenariolS, but relatively large peak flow differences
among the three build-out future scenarios (Figure 9d).
Baseflow and Groundwater Recharge. To examine the effect of increased
imperviousness on baseflow and groundwater recharge, we calculated baseflow index
(BFI) from observed streamflow (historical condition), streamflow simulated for the
baseline condition, and streamflow forecasted for the three future build-out scenarios.
Baseflow indices from the observed and simulated data for the baseline scenario were
0.62 and 0.64, respectively, whereas indices forecasted for scenariolS, scenario45,
scenarioSS were 0.59, 0.46, and 0.28, respectively. This decreasing BFI trend indicates
how baseflow and groundwater recharge decrease when TIA levels increase.
To further examine the relationship between increased imperviousness, baseflow
and groundwater recharge further, we compared flow duration curves for flows with
exceedance probability of greater than 95% for the baseline scenario and for the three
build-out scenarios over 30-years (Figure 9c). These comparisons reveal little difference
between the baseline condition and scenariolS simulations, but show drastic baseflow
reductions for the scenario45 and scenarioSS simulations (Figure 9c). In this report, we
also calculated 7Q10 flow values from streamflow simulated for the baseline, scenariolS,
scenario45, and scenarioSS as 0.16, 0.15, 0.10, and 0.03 nrVsec, respectively.
55
-------
Precipitation (mm)
Baseline
Scenraio15
Scenario45
ScenarioSS
«M T- «M
Day
0.35
0.3
o 0.25
0)
m
£ 0.2
5
5 0.15
ra
£ 0.1
0.05
•Baseline
-Scenario45
ScenariolS
ScenarioSS
in in u> u> u> u>
00 00 00 00 Of) Of) Of)
on of) of) of) of) of) of) of) of) of) of) of) of) of) of) of)
Percent of Time Flow Equaled or Exceeded
1600
1400
1200
1000
800
600
400
200
0
400
350
300
250
200
150
100
50
0
D Surface runoff (mm)
D Interflow runoff (mm)
• Baseflow runoff (mm)
ID Simulated Evapotranspiration (mm)
• Water Balance Error (mm)
(b)
Baseline
ScenariolS
Scenario45
ScenarioSS
• Baseline
•Scenario45
•so
•So
(c)
5 «
o o
fe
o
858
IO
$ 3 8 S S 2
ei ei ri ri ri •*
Percent of Time Flow Equaled or Exceeded
§
Figure 9. Hydrologic changes simulated for baseline, ScenariolS, scenario45,
and scenarioSS: (a) daily streamflows, (b) annual water balances (c) flow duration curves at low flow conditions, and (d) flow duration curves at high
flow conditions
56
-------
Forecasting TSS Concentrations. Suspended sediment concentrations forecasted for
scenariolS, scenario45, and scenarioSS over a 30-year simulation period did not yield a
continuously increasing function with percent TIA level. Indeed, our scenario projections
suggest that sediment concentrations are controlled by available sediment supply and by
transport capacity. Specifically, during some elevated flow periods, scenario45
projections had higher TSS concentrations than scenarioSS, even though scenarioSS had
higher simulated flows, thus higher sediment transport capacity (Figure 9b). One
explanation is that there is a TIA threshold value beyond which sediment loads
transported from land surfaces may not increase with increased impervious area. This
implies that when sediment supply is available, scenarioSS would have produced the
highest simulated TSS concentrations because it has the greatest sediment transport
capacity. However, as the landscape becomes increasingly impervious, sediment supply
becomes limited and increases in sediment transport capacity do not translate directly into
increased sediment export.
An analysis of forecasted daily TSS concentrations for the build-out scenarios
indicates a threshold percent TIA value somewhere between 45 and 85 TIA levels where
increased percent TIA does not translate to increased TSS concentrations. In this report,
we forecasted 30-year average annual sediment loads of 146, 211, 452, and 241 (kg/ha-
yr) for the baseline, scenariolS, scenario45, and scenarioSS, respectively.
Forecasting Nitrate Concentrations. For the build-out scenarios, the model forecasted
higher nitrate concentrations at low flows and lower concentrations at elevated flows
(Figure 9a). This is because nitrate simulations are strongly influenced by flow
57
-------
alterations, particularly baseflow reduction and increase in surface runoff. Conversely, at
baseline condition, higher nitrate concentrations were observed during elevated flow
conditions and lower concentrations at low flow conditions, suggesting that, at baseline,
nitrate accumulates in the upper soil layers during dry periods and is washed out during
wet periods (Figure 9). This observed nitrate build-up - wash-out cycle under the baseline
scenario was severely disrupted by urbanization, which alters the dominant runoff
generation mechanisms and flow pathways.
Model results simulated for the build-out scenarios show increased nitrate
concentrations during dry periods and lower concentrations at elevated flow periods.
Such interdependence between flow alteration and nitrate concentrations might also
impact soil denitrification processes. As stated previously, our nitrate projections for the
three build-out scenarios produced nitrate concentrations in the baseflow that increased
with increased percent TIA level. Even ignoring other nitrogen sources, such as lawn
fertilizers and atmospheric deposition, our results indicate that flow alteration alone can
increase nitrate concentrations that may cause eutrophication threats to coastal waters. In
summary, our forecasted 30-year average annual nitrate loads were 13, 14, 16, and 23 kg-
NO3-N/ha-yr for the baseline, scenariolS, scenario45, and scenarioSS, respectively.
58
-------
•Baseline
Scenario15
•Scenario45 — — ScenarioSS
.A*
^ J3
0& 0& 0& 0& 0& 0&
& \& A^ ^ ^ ^
> \^ ^ ^ ^ ^
** Day <* <* ^
O)
CO
P
•Baseline
ScenariolS
-Scenario45
-ScenarioSS
300
•Baseline
ScenariolS
•Scenario45
•ScenarioSS
Figure 10. HSPF simulated streamflow, TSS, and nitrate concentrations for baseline,
ScenariolS, scenario45, and ScenarioSS
59
-------
5.4 Determining Ecologically Relevant Urbanization-Induced Stressor
Indicators
Hydrologic Indicators. Flow alterations have an adverse impact on biological
integrity, but quantifying that impact has been a challenge. Richter et al. (1996) present
indicators of hydrologic alteration (IHA) calculated from observed historical streamflow
data. In this study, we estimated IHAs from simulated streamflow for the baseline and for
the build-out scenarios (Table 4). Simulated magnitude of average monthly flows vary
with the month of year. For some months, the monthly flows decrease with increase in
TIA levels whereas for other months monthly flows decrease with increase in TIA levels.
As the TIA levels increased, model results show a decrease in magnitude and duration of
annual extreme flows for build-out scenarios. Conversely, model results show increased
frequency and duration of average high and low flow pulses as well as increased average
rate and frequency of flow changes (Table 4). Simulated IHAs show that urbanization
does not alter the timing of average annual extreme flows.
Determining IHA values for build-out scenarios with different percent TIA levels,
and establishing quantitative links between urbanization-induced stressors and their
ecological impacts, are essential to urban ecosystem protection and restoration planning.
Because the amount of water available in a stream defines the suitability of a habitat to
aquatic organisms, flow alteration, especially low flows, creates unfavorable conditions
for native species (Poff et al., 2006). To ensure that sufficient water is available for
aquatic organisms, some states set 7Q10 flow values as an in-stream flow requirement
that must not be violated when issuing water withdrawal permits for irrigation and
municipal water supply.
60
-------
Table 4. Indicators of hydrologic alteration (IHA) calculated from streamflow simulated for the
baseline and the three build-out scenarios for a 30-year period.
IHA Parameter Group Baseline ScenariolS Scenario45 ScenarioSS
Group #1: Magnitude of average monthly flows (m3/sec)
October 0.47 0.51 0.41 0.19
November 4.29 6.16 6.93 5.49
December 11.35 11.04 11.20 6.12
January 9.91 10.54 10.16 5.40
February 9.12 9.88 8.00 3.70
March 8.45 8.52 8.48 5.75
April 6.20 6.13 4.95 2.35
May 3.36 3.27 2.14 0.76
June 1.98 1.85 1.27 0.40
July 1.14 0.98 0.64 0.19
August 0.67 0.58 0.38 0.11
September 0.47 0.46 0.29 0.09
Group #2: Magnitude and duration of annual extreme flows (m3/sec)
1-day minimum 0.32 0.27 0.18 0.05
3-day minimum 0.32 0.28 0.18 0.05
7-day minimum 0.33 0.31 0.20 0.06
30-day minimum 0.42 0.47 0.37 0.13
90-day minimum 0.76 0.90 0.98 0.87
1-day maximum 67.60 84.61 117.06 172.90
3-day maximum 50.60 57.54 73.07 94.16
7-day maximum 39.00 41.70 47.59 60.54
30-day maximum 21.65 22.59 24.86 29.65
90-day maximum 15.11 5.81 17.90 21.03
Days with zero flow 0.00 0.00 0.00 0.00
Base flow index 0.05 0.043 0.02 0.01
Group #3:Timing of annual extreme flows (Julian Day)
Date of minimum 277.00 277.00 275.00 275.00
Date of maximum 347.00 338.00 338.00 338.00
Group #4: Frequency and duration of average high and low pulses
Low pulse count 1 .00
Low pulse duration 62.25
High pulse count 8.50
High pulse duration 8.50
Low Pulse Threshold 30.65
High Pulse Threshold 288.00
Group #5: Rate and frequency of flow
Rise rate 30.75
Fall rate -2.53
Number of reversals 64.00
8.00
5.75
14.00
3.00
32.30
307.00
changes
98.75
-6.45
111.50
9.00
5.00
27.50
2.00
22.20
327.00
239.10
-6.42
119.00
8.00
5.50
35.00
2.00
7.50
310.00
445.00
-5.35
121.00
61
-------
Hydraulic and water quality indicators. High peak flow rates associated with high
percent TIA levels cause high flow velocities and associated shear stresses that move
sediments (Poff et al., 2006), and displace benthic invertebrates (Poff and Ward, 1991),
and small fish (Harvey, 1987). High shear stresses also increase the depth of scour of bed
sediments to induce higher mortality of benthic invertebrates (Palmer et al., 1992;
Townsend et al., 1997). To forecast the risks of increased stream channel erosion, we
used Ch, often referred to as the bankfull flow or channel forming flow, which
corresponds to a 2-year return period. Wolman and Miller (1960) reported that flows with
1 to 5 year return periods are important to stream geomorphology because these flows
move sufficient amounts of sediment. To forecast the impacts of urbanization on
streambank erosion, we examine how different build-out scenarios alter the frequency of
bankfull or channel forming flows (Ch). Using Ch as an indicator of increased channel
erosion, we forecasted the number of times the baseline scenario Ch flow value was
exceeded by Ch flows forecasted for the three build-out scenarios. We found 22, 91, and
274 days with flows greater than the baseline scenario Ch for the 30-year projections for
scenariolS, scenario45, and scenarioSS, respectively.
Establishing stressor-response relationships between changes in shear stress and Ch
values under future build-out scenarios and their impacts on aquatic organisms is key to
developing biologically relevant indicators. Although some stressor indicators may have
biological relevance, additional field research is needed to establish cause-and-effect
relationships between stressor indicators and ecological responses.
The HSPF model simulates many physical habitat variables closely related to channel
hydraulic geometry. To establish links between forecasted habitat variables and the health
62
-------
of aquatic ecosystems, we propose the use of habitat models such as the Physical Habitat
Simulation model (PHABSIM), a major component of the Instream Flow Incremental
Methodology (IFIM) (Milhous et. al., 1984), and ecological models such as AQUATOX.
We suggest importing HSPF forecasted habitat variables and stressors, such as scenario-
specific simulated hydraulic, temperature, nutrient, and contaminant data, as inputs into
the PHABSIM and AQUATOX models to compute the impact these stressors have on the
health of aquatic ecosystems.
Comparisons of simulated flow velocity for the baseline scenario to velocities
forecasted for the three build-out scenarios indicate significant baseflow velocity
reductions with increasing percent TIA level (Figure 10). For instance, scenarioSS, which
has the highest percent TIA, had the lowest baseflow velocity, followed by scenario45
and scenariolS. The baseline scenario, with the lowest percent TIA, had the highest
baseflow velocity. As presented earlier, the flow velocities fluctuated between baseflow
and peak flows within each scenario. However, these fluctuations increased with increase
in percent TIA; such that peak flows and baseflows had an inverse order relationship to
percent TIA; that is, scenarioSS had the highest peak flows followed in order by
scenario45, scenariolS, and baseline.
63
-------
Baseline
Day
• Baseline
Scenario16
Scenario46 —»—Scenario86
*•*•
I
I
36
30
26
20
16
10
• Baseline
• Scenario15
• Scenario45
• ScenarioSS
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
Day
Figure 11. HSPF simulated 30-year mean water temperature (a), water surface width
(b), and flow velocity (c) for baseline, scenariolS, scenario45, and scenarioSS.
64
-------
6 CONCLUSION
Controlling nonpoint source pollution requires the use of integrative modeling
approaches that simulate the interactions among integrated stressors. However, current
models and modeling approaches cannot represent integrated stressors and many
available models do not match the complexities of the system that is modeled. Only a
very few watershed models can represent the various hydromodification types and
scenarios and can simulate scenario-specific flow alterations and water quality
degradation. In this report, we present a modeling approach that is suitable for assessing
how hydromodification projects alter natural flow regimes and degrade water quality.
The proposed B ASINS-based modeling framework enables resource managers to select
alternative future scenarios, simulate scenario-specific stressors, and select scenarios that
minimize the impacts of hydromodification on water quantity and quality.
Urbanization has been closely linked to the degradation of aquatic ecosystems.
Methods to quantify urbanization-induced stressors from future development scenarios
have not been generally available. In this study, we present an application of a method to
forecast urbanization-induced stressors and discuss ways to develop indicators that link
stressors to their ecological impacts. The proposed approach will enable land use
planners and resource managers to forecast urbanization-induced stressors from
forecasted streamflow and water quality data. Forecasted urbanization-induced stressors
include flow alteration (e.g., altered peak flows and baseflows), water quality degradation
(e.g., increased nitrate and TSS concentrations), and geomorphic or habitat alterations
(e.g., changes in flow depth, wetted surface width, and flow velocity).
65
-------
The ultimate utility of the proposed forecasting approach depends on successfully
developing ecological response indicators that capture the responses of aquatic
ecosystems to the stressors of concern. Ecological response indicators can be derived
possibly from data on changes in native species richness or on the composition and
functional organization offish and invertebrates.
Future research should focus on developing ecological response indicators that
quantitatively reflect aquatic biota responses to forecasted stressors. Specifically, such
research should assess the effects of single or integrated stressors on the abundance,
diversity, and fitness of aquatic communities. Future modeling in this area ideally will
also simulate how different best management practices moderate the ecological effects of
urbanization to restore important ecological services and functions.
66
-------
7 REFERENCES
Adams, S.M., 2005. Assessing cause and effect of multiple stressors on marine systems.
Marine Pollution Bulletin, Vol. 52(8-12): 649-657.
Allan, J. D., 2004. Landscapes and riverscapes: the influence of land use on stream
ecosystems. Annual Review of Ecology Evolution and Systematics 35:257-284.
Arnold, C. L. Jr. and C. J. Gibbons, 1996. Impervious surface coverage: the emergence of
a key environmental indicator. J. Am. Plan. Assoc. 62(2): 243-258.
Azevedo, L.G.T., T.K.Gates, D.G.Fontane, J.W.Labadie, and R.L. Porto, 2000,
Integration of water quantity and quality in strategic river Basin planning, J. Water
Resour. Plann. Manage. ASCE 126, 85-97.
Bekele, E.G. and J.W. Nicklow. 2005. Multiobjective management of ecosystem services
by integrative watershed modeling and evolutionary algorithms. Water Resources
Research. 41(10): W10406-1-10.
Bicknell, B. R., J. C. Imhoff, J. L. Kittle Jr, A. S. Donigian, and R. Johanson, 2001.
Hydrologic Simulation Program- Fortran User's Manual for Release 12. Exposure
Research Laboratory, Office of Research and Development, US Environmental
Protection Agency, Athens, Georgia.
Boesch, D., E. Burreson, W. Dennison, E. Houde, M. Kemp, V. Kennedy, R. Newell, K.
Paynter, R. Orth, and R. Ulanowicz, 2001. Factors in the decline of coastal ecosystems.
Science 293:1589-1590.
Booth, D. B., 1990. Stream-channel incision following drainage-basin urbanization.
Journal of the American Water Resources Association 26(3):407-417.
Booth, D.B., J.R. Karr, S. Schauman, C.P. Konrad, S.A. Morley, M.G. Larson, and S.J.
Burges, 2004. Reviving urban streams: Land use, hydrology, biology, and human
behavior Journal of the American Water Resources Association 40(5): 1351-1364.
Brun, S. E. and L.E. Band. 2000. Simulatimng runoff behavior in an urbanizing
watershed, Computers, Environment, and Urban Systems 24(1): 5-22.
Bunn, S. E., and A. H. Arthington, 2002. Basic principles and ecological consequences of
altered flow regimes for aquatic biodiversity. Environmental Management 30:492-507.
Cai, X.M., D.C. McKinney, andL.S. Lasdon, 2003. Integrated hydrologic-agronomic-
economic model for river basin management. Journal of Water Resources Planning and
Management. 129 (1):4-17.
67
-------
Cassin, I, R. Fuerstenberg, L. Tear, K. Whiting, D. St. John, B. Murray, J. Burkey, 2005.
Development of hydrological and biological indicators of flow alteration in Puget Sound
lowland streams. King County Water and Land Resources Division. Seattle, Washington.
Clausen, B. and, B.J.F., Biggs, 1997. Relationships between benthic biota and
hydrological indices in New Zealand streams. Freshwater Biology 38, pp. 327-342.
Colby, B. G., 1990. Enhancing instream flow benefits in an era of water marketing.
Water Resources Research 26(6): 1113-1120.
Collier, M.P., R.H. Webb, and E.D. Andrews. 1997. Experimental flooding in the Grand
Canyon. Scientific American 276:82-89.
Compton, I.E., M.R., Church, S.T., Larned, W.E., Hogsett. 2003. Nitrogen export from
forested watershed in the Oregon Coast Range: the role of N2-fixing red
alder.Ecosystems 6:773-785
Conway, T.M., 2007. Impervious surface as an indicator of pH and specific conductance
in the urbanizing coastal zone of New Jersey, USA. Journal of Environmental
Management 85:308-316.
Creed, I. C., and L. E. Band, 1998. Export of nitrogen from catchments within a
temperate forest: Evidence for a unifying mechanism regulated by variable source area
dynamics. Water Resources Research 34(11):3105-3120.
Deal, B., 2001. Ecological urban dynamics: the convergence of spatial modelling and
sustainability. Building Research and Information. 29(5), 381-393.
Donigian, A. S., 2002. Watershed model calibration and validation: The HSPF
Experience, National TMDL Science and Policy Specialty Conference 2002. Water
Environment Federation, Phoenix, Arizona.
Donner, S. D., J. A. Foley, and C. J. Kucharik, 2004. Impact of changing land use
practices on nitrate export by the Mississippi River. Global Biogeochemical Cycles, 18,
GB1028, doi:10.1029/2003GB002093.
Echeverria, J.D., P. Barrow, and R. Roon-Collins. 1989. Rivers at risk. Washington,
D.C., Island press.
Eckhardt, K., 2005. How to construct recursive digital filters for baseflow separation.
Hydrological Processes 19, 507-515.
Elofson, K., H. Folmer, and I-M. Gren. 2003. Management of eutrophicated coastal
ecosystems. Ecological Economics 47:1-11.
68
-------
Endreny, T.A. (2002). Manipulating HSPF to simulate pollutant transport in suburban
systems. ASAE TMDL Meeting, Fort Worth, Texas March 11-13, 2002.
Fitzpatrick, F.A., H.A., Harris, T.L., Arnold, and K.D., Richards, 2004., Urbanization
influences on aquatic communities in northeastern Illinois streams: Journal of the
American Water Resources Association, v. 40, no. 2, p. 461-475.
General Accounting Office. 2004. Watershed Management: Better Coordination of Data
Collection Efforts Needed to Support Key Decisions. GAO Report Number GAO-04-
382, June 2004, Washington, D.C. (http://www.gao.gov/new.items/d04382.pdf)
Goncii, S. and E. Albek, 2009. Modeling Climate Change Effects on Streams and
Reservoirs with HSPF. Journal Water Resources Management.DOI 10.1007/sl 1269-009-
9466-6 (Online)
Gove, N.E., R.T. Edwards, and L.L. Conquest, 2001. Effects of scale on land use and
water quality relationships: A longitudinal basin-wide perspective. Journal of the
American Water Resources Association, 37(6): 1721-1734.
Hamon, W.R., 1961. Estimating potential evapotranspiration. Proceedings of the
American Society of Civil Engineers, Journal Hydraulics Division 87:107-120.
Harvey, B. C., 1987. Susceptibility of young-of-the-year fishes to downstream
displacement by flooding. Transactions of the American Fisheries Society 116:851-855.
Hatt, B.E., T.D., Fletcher, C.J., Walsh, and C.M. Taylor, 2004. The influence of urban
density and drainage infrastructure on the concentrations and loads of pollutants in small
streams. Environ. Manag. 34: 112-124.
Houghton, J. T., Y. Ding, D. J. Griggs, M. Noguer, P.J. van der Linden, and V. Xiaosu,
eds. 2001. Climate Change 2001 :The Scientific Basis. Intergovernmental Panel on
Climate Change: Working Group I. Cambridge University
Jackson, S. and W. Davis. 1994. Meeting the goal of biological integrity in water-
resource programs in the US Environmental Protection Agency. J. N. Am. Benthol. Soc.,
13(4):592-597.
Karr, J. R. 1991. Biological integrity: a long neglected aspect of water resources
management. Ecological Applications l(l):66-84.
Klein, R. D., 1979. Urbanization and stream quality impairment. Water Resources
Bulletin 15(4):948-963.
Konrad, C. P., and D.B. Booth, 2005. Hydrologic changes in urban streams and their
ecological significance, in effects of urbanization on stream ecosystems, edited by L. R.
Brown et al., Am. Fish. Soc., Bethesda, Md., In Press.
69
-------
Konrad, C. P., D. B. Booth, and S. J. Burges. 2005. Effects of urban development in the
Puget Lowland, Washington, on interannual streamflow patterns: Consequences for
channel form and streambed disturbance, Water Resour. Res., 41, W07009,
doi:10.1029/2005WR004097.
Labadie, J. 2005. MODSIM: River basin management decision support system. Chapter
23 in Watershed Models CRC Press, Boca Raton, FL.
Landis, J.D., 1995. Imagining land use futures: Applying the California urban futures
model. Journal of the American Planning Association. 61: 438-457.
Leblanc, R.T., R.D. Brown, and I.E. FitzGibbons, 1997. Modeling the effects of land use
change on the water temperature in unregulated urban streams. Journal of Environmental
Management 49:445-469.
Letcher, R.A., A.J., Jakeman, and B.F.W., Croke,. 2004. Model development for
integrated assessment of water allocation options. Water Resources Research
40, W05502, doi: 10.1029/2003WR002944.
Ligon, F.K., W.E. Dietrich, WJ. Trush, 1995. Downstream ecological effects of dams. A
geomorphic perspective Bioscience Vol. 45(3) 183-192.
Mallin, M.A., S.H. Ensign, M.R. McEver, G.C. Shank, andP.K. Fowler, 2001.
Demographic, landscape, and meteorological factors controlling the microbial pollution
of coastal watersheds. Hydrobiologia. 460:185-193.
McColl C. and G. Aggett, 2007. Land-use forecasting and hydrologic
model integration for improved land-use decision support, J. Environ.
Manage., 84,494-512.
Middelkoop, H., K. Daamen, D. Gellens, W. Grabs, J. C. J. Kwadijk, H. Lang, B. W. A.
H. Parmet, B. Scha'dler, J. Schulla, and K. Wilke (2001), Impact of climate change on
hydrological regimes and water resources management in the Rhine basin, Clim. Change,
49, 105-128.
Milhous, R.T., D. L. Wegner, and T. Waddle, 1984. User's guide to the physical habitat
simulation system. US Fish and Wildlife Service Biological Services Program,
FWS/OBS-81/43.
Mohamoud, Y.M. 2008. Impact of urbanization on water quantity and quality: the need
for an integrative watershed modeling approach. Water Resources Impact. 10(4):27-29.
Morse, C.C; A.D., Huryn, C. Cronan, 2003. Impervious surface area as a predictor of the
effects of urbanization on stream insect communities in Maine, USA Environmental
Monitoring and Assessment 89(1): 95-127.
70
-------
Naiman, R.J., S.E. Bunn, C. Nilsson, G.E. Petts, G. Pinay, L.C. Thompson,. 2002.
Legitimizing fluvial ecosystems as users of water: An overview. Environmental
Management, 30(4):455-467.
Nash, J. E. and J. V. Sutcliffe, 1970. River flow forecasting through conceptual models
part I — A discussion of principles. Journal of Hydrology 10(3):282-290.
National Research Council (1992) Restoration of Aquatic Ecosystems: Science,
Technology, and Public Policy. Washington, DC: National Academy Press.
National Research Council. 2004. Confronting the nations water problems: the research
role. The national Academies Press, Washington, D.C.
Nelson, E. J., and D.B. Booth, 2002. Sediment sources in an urbanizing, mixed land-use.
Journal of Hydrology 264(l-4):51-68.
Ng, H.Y.F., and J. Marsalek, 1989. Simulation of the effects of urbanization on basin
streamflow. Water Resource Bulletin 25:117-124.
Niemi, G., D. Wardrop, R. Brooks, S. Anderson, V. Brady, H. Paerl, C. Rakocinski, M.
Brouwer, B. Levinson, and M. McDonald, 2004. Rationale for a new generation of
indicators for coastal waters. Environmental Health Perspectives 112:979-986.
Nixon, S.W., 1995. Coastal marine eutrophication: A definition, social causes, and future
concerns. Ophelia 41:199-219.
Ohmann, J. L., and M.J. Gregory, 2002. Predictive mapping of forest composition and
structure with direct gradient analysis and nearest neighbor imputation in coastal Oregon,
USA. Canadian Journal of Forestry Research 32:725-741.
Okun, D.A., 2002. Water reuse introduces the need to integrate both water supply and
wastewater management at local and regulatory levels. Water Science and Technology,
46(6-7): 273-280.
Olden, J. D., and N. L. Poff, 2003. Redundancy and the choice of hydrologic indices for
characterizing streamflow regimes. River Research Applications 19, 101-121.
Palmer, M.A., A. E. Bely, and K.E. Berg, 1992. Response of invertebrates to lotic
disturbances - a test of the hyporheic refuge hypothesis. Oecologia 89:182-121.
Park, R. A., J. S. Clough and M. C. Wellman. 2008. AQUATOX: Modeling
environmental fate and ecological effects in aquatic ecosystems Ecological Modelling
213: 1-15.
Paul, M. J., and J. L. Meyer, 2001. Streams in the urban landscape. Annual Review of
Ecology and Systematics 32:333-365.
71
-------
Poff, N. L., B. P. Bledsoe, and C. O. Cuhaciyan, 2006. Hydrologic variations with land
use across contiguous United States: Geormorphic and ecological consequences for
stream ecosystems. Geomorphology 79:264-285
Poff, L.N., J.D. Allan, M.B. Bain, J.R. Karr, K.L. Prestegaard, B.D. Richter, R.E. Sparks
and J.C. Stromberg, 1997. The natural flow regime: A paradigm for river conservation
and restoration. BioScience 47: 769-784.
Poff, N.L., and J.D. Allan, 1995. Functional organization of stream fish assemblages in
relation to hydrological variability. Ecology 76:606-627.
Poff, N.L., and J.V.Ward, 1991. Drift responses of benthic invertebrates to experimental
hydrologic variation in a regulated stream. Canadian Journal of Fisheries and Aquatic
Sciences 48:1926-1936.
Poff, N. L., and J.V. Ward, 1989. Implications of streamflow variability and
predictability for lotic community structure, a regional analysis of streamflow patterns.
Canadian Journal of Fisheries and Aquatic Science 46(10): 1805-1818.
Power, M .E., W. E. Dietrich, and J. C. Finlay, 1996. Dams and downstream biodiversity:
potential food web consequences of hydrologic and geomorphic change. Environmental
Management 20(6): 887- 895.
Rabalais, N. N., R. E. Turner, D. Justic, A. Dortch, W. J. Wiseman Jr., and B. K. Gupta,
1996. Nutrient changes in the Mississippi River and system responses on the adjacent
continental shelf. Estuaries 19:386-407.
Richards, C., R. J. Haro, L. B. Johnson, and G. E. Host, 1997. Catchment and reach-scale
properties as indicators of macroinvertebrate species. Freshwater Biology 37:219-230.
Richardson, K., 1997. Harmful or exceptional phytoplankton blooms in the marine
ecosystem. Advances in Marine Biology 31:301-385.
Richter, B. D. and H.E. Richter. 2000. Prescribing flood regimes to sustain riparian
ecosystems along meandering rivers. Conservation Biology 14: 1467-1478.
Richter, B.D., J.V. Baumgartner, J. Powell, and D. Braun. 1996. A method for assessing
hydrologic alteration within ecosystems. Conservation Biology. (10) 4:1163-1174.
Rose, S. and N. E Peters,. 2001. Effects of urbanization on streamflow in the Atlanta area
(Georgia, USA): a comparative hydrological approach. Hydrological Processes 15(8):
1441-1457.
Schueler, T., 1995. Environmental land Panning Series: Site planning for urban streams
protection. Center for Watershed Protection Publication No. 95708. Metropolitan
Washington Council of Governments,Washington, DC.
72
-------
Schueler, T. R., 1994. The importance of imperviousness. Watershed Protection
Techniques 1:00-111
Sehlke, G. 2004. Basin management case studies: an overview of selected tools,
techniques, and methods for managing river basins: Water Resources Impact. Vol. (3): 5-
Snavely Jr, P.D. and H. C. Wagner, 1963. Tertiary geologic history of western Oregon
and Washington: Washington Division of Mines and Geology Report of Investigations
22, 25p.
Stanford, J.A., J.V. Ward, WJ. Liss, C.A. Frissell, R.N. Williams, J.A. Lichatowich, and
C.C. Coutant. 1996. A General Protocol for Restoration of Regulated Rivers.
RegulatedRivers: Research and Management 12:391-413.
Tang, Z., Engel, B.A., Pijanowski, B.C. and Lim, K.J., 2005. Forecasting land use change
and its environmental impact at a watershed scale. Journal of Environmental
Management, 76(1): 35-45.
Townsend, C. R., S. Doledec, and M. R. Scarsbrook, 1997. Species traits in relation to
temporal and spatial heterogeneity in streams: a test of habitat template theory.
Freshwater Biology 37:367-387.
Turner, R. E., and N.N. Rabalais, 1991. Changes in Mississippi River water quality this
century. Implications for coastal food webs. Bioscience 41(3): 140-148.
U.S. Geological Survey., 2007Water Resources for Oregon. Water Data Report OR-00 1,
498 pp., Portland Oregon USGS (U.S. Geological Survey).
U.S. Geological Survey, 2006. Suspended sediment database. Daily values of suspended
sediment and ancillary data, http://co.water.usgs.gov/sediment/seddatabase.cfm, accessed
July 2006.
USGS (U.S. Geological Survey). 2002. Concepts for national assessment of water
quantity and use. U.S. Geological Survey Circular 1223. Reston, Virginia.
USEPA. 2009. National Water Quality Inventory: 2004 Report to Congress. EPA 841-r-
08-001. Office of Water Washington, D.C.
USEPA. 2007. National Management Measures Guidance to Control Nonpoint Source
Pollution from Hydromodification. EPA 841-B-07-002. U.S. Environmental Protection
Agency, Washington, DC. (http://www.epa.gov/owow/nps/hydromod/index.htm).
USEPA. 2004. Protecting water resources with smart growth. EPA 231-R-04-002.
Washington, D.C.
73
-------
USEPA., 2001. Better Assessment Science Integrating Point and Non-point Sources
(BASINS Version 3.1). Office of Water. U.S. Environmental Protection Agency.
Washington, D.C.
USEPA. 1996a. Better Assessment Science Integrating Point and Nonpoint Sources.
EPA-823-R-96-001.USEPA. 1998. National Water Quality Inventory: 1996 Report to
Congress. EPA841-R-97-008. U.S. Environmental Protection Agency, Washington, DC.
http://www.epa.gov/305b/96report. Accessed May 2009.
USEPA. 1993. Guidance specifying management measures for sources of nonpoint
pollution in coastal waters. EPA 840-B-92-002B. U.S. Environmental Protection Agency,
Washington, DC.
Vitousek, P. M., H. A. Mooney, J. Lubchenko, and J. M. Mellilo, 1997. Human
domination of earth's ecosystem. Science 277(5325):494-499.
Wahl, K. L., and T. L. Wahl, 1995, Determining the flow of Comal Springs at New
Braunfels, Texas, Texas Water '95, American Society of Civil Engineers, August 16-17,
1995, San Antonio, Texas, pp. 77-86.
Ward ,J.V and J.A.Stanford ,1995. The serial discontinuity concept: extending the model
to floodplain rivers. Regulated Rivers 10: 159-168.
Wernick, B., G., K. E. Cook, and H. Schreier, 1998. Land use and streamwater nitrate-n
dynamics in an urban-rural fringe watershed. Journal of the American Water Resources
Association 34(3):639-650.
Whiting, P. J. 2002. Streamflow necessary for environmental maintenance, Annual
Review of Earth and Planetary Sciences, 30, 181-206.
Whittemore, R.C and J. Beebe, 2000. EPA's basins model: Good science or serendipitous
modeling? Journal of the American Water Resources Association 36(3): 493-499.
Wolman, M.G. and W.P. Miller, 1960. Magnitude and frequency offerees in geomorphic
processes. Journal of Geology, 68: 54-74.
Wotling, G., and C. H. Bouvier, 2002. Impact of urbanization on suspended sediment and
organic matter fluxes from small catchments in Tahiti. Hydrological Processes
16(9): 1745-756.
Wurbs, R. A. 2005. Modeling river/reservoir system management, water allocation, and
supply reliability Journal of Hydrology 300 (1-4): 100-113
Zarriello, P.J. and, K.G.Reis, 2000, A precipitation-runoff model for analysis of the
effects of water withdrawals on streamflow, Ipswich River basin, Massachusetts, U.S.
Geological Survey Water-Resources Investigations Report 00-4029.
74
-------
Zimmerman, J.B., J.R. Mihelic, J. Smith, 2008. Global stressors on water quality and
quantity. Environmental Science and Technology 42(4): 4247- 4254
75
-------
vvEPA
United States
Environmental Protection
Agency
Office of Research
and Development (8101R)
Washington, DC 20460
Official Business
Penalty for Private Use
$300
EPA/600/R-09/116
September 2009
Please make all necessary changes on the below label,
detach or copy, and return to the address in the upper
left-hand corner.
If you do not wish to receive these reports CHECK HERE
D; detach, or copy this cover, and return to the address in
the upper left-hand corner.
PRESORTED STANDARD
POSTAGE & FEES PAID
EPA
PERMIT No. G-35
Recycled/Recyclable
Printed with vegetable-based ink on
paper that contains a minimum of
50% post-consumer fiber content
processed chlorine free
------- |