This is the accessible text file for GAO report number GAO-12-335 
entitled 'Nonpoint Source Water Pollution: Greater Oversight and 
Additional Data Needed for Key EPA Water Program' which was released 
on July 2, 2012. 

This text file was formatted by the U.S. Government Accountability 
Office (GAO) to be accessible to users with visual impairments, as 
part of a longer term project to improve GAO products' accessibility. 
Every attempt has been made to maintain the structural and data 
integrity of the original printed product. Accessibility features, 
such as text descriptions of tables, consecutively numbered footnotes 
placed at the end of the file, and the text of agency comment letters, 
are provided but may not exactly duplicate the presentation or format 
of the printed version. The portable document format (PDF) file is an 
exact electronic replica of the printed version. We welcome your 
feedback. Please E-mail your comments regarding the contents or 
accessibility features of this document to Webmaster@gao.gov. 

This is a work of the U.S. government and is not subject to copyright 
protection in the United States. It may be reproduced and distributed 
in its entirety without further permission from GAO. Because this work 
may contain copyrighted images or other material, permission from the 
copyright holder may be necessary if you wish to reproduce this 
material separately. 

United States Government Accountability Office: 
GAO: 

Report to Congressional Requesters: 

May 2012: 

Nonpoint Source Water Pollution: 

Greater Oversight and Additional Data Needed for Key EPA Water Program: 

GAO-12-335: 

GAO Highlights: 

Highlights of GAO-12-335, a report to congressional requesters. 

Why GAO Did This Study: 

Pollution from nonpoint sources—such as runoff from farms or 
construction sites—remains the leading cause of impairment to the 
nation’s waters. Under section 319 of the Clean Water Act, each year 
EPA provides grants to states to implement programs and fund projects 
that address nonpoint source pollution; the program received $165 
million in fiscal year 2012. Section 319 includes minimum conditions 
that states must meet to receive grants. By regulation, EPA’s 10 
regional offices oversee state programs and are to ensure that states’ 
projects can be feasibly implemented. USDA also has programs to 
protect water resources. 

GAO examined (1) states’ experiences in funding projects that address 
nonpoint source pollution, (2) the extent to which EPA oversees the 
section 319 program and measures its effectiveness, and (3) the extent 
to which key agricultural programs complement EPA efforts to control 
such pollution. GAO surveyed project managers, reviewed information 
from EPA’s 10 regional offices on oversight of state programs, and 
analyzed USDA data. 

What GAO Found: 

Under section 319 of the Clean Water Act, state-selected projects to 
reduce nonpoint source pollution have helped restore more than 350 
impaired water bodies since 2000, but other projects have encountered 
significant challenges. According to GAO survey results, 28 percent of 
projects did not achieve all objectives originally identified in the 
project proposal (e.g., implementing the desired number of pollution 
reduction practices), while many that did so still faced considerable 
challenges. About half such challenges were beyond staff control 
(e.g., bad weather or staff turnover), but the other half were 
challenges that generally could have been identified and mitigated 
before projects were proposed and selected for funding, such as 
gaining access to desired properties. In one state, for example, 
$285,000 in section 319 funds was to subsidize the cost to homeowners 
of repairing damaged septic systems. Once the grant was awarded, 
however, one homeowner signed up to participate. 

The Environmental Protection Agency’s (EPA) oversight and measures of 
effectiveness of states’ programs have not consistently ensured the 
selection of projects likely to yield measurable water quality 
outcomes. EPA’s 10 regional offices varied widely in their review of 
states’ work plans, which describe projects states plan to undertake 
in the upcoming year, and project selection criteria, which identify 
eligibility parameters for receiving section 319 funds. For example, 
three regional offices reported reviewing annual work plans in depth 
and actively influencing the types of projects selected, while three 
others reported limited to no involvement in such reviews, instead 
deferring to states’ judgment on project feasibility and selection. 
EPA, however, has not provided its 10 regions with guidance on how to 
oversee the state programs. Also, EPA’s primary measures of program 
effectiveness may not fully demonstrate program achievements. Section 
319 requires states to report to EPA on two measures, including 
reductions in key pollutants. It does not limit EPA to these two 
measures, but the agency has chosen to use them as barometers of 
success for the section 319 program. States can demonstrate their 
achievements in additional ways—ways that may provide a more accurate 
picture of the overall health of targeted water bodies, such as the 
number and kind of living organisms in the water. 

USDA’s Environmental Quality Incentives Program is the key 
agricultural conservation program that can complement EPA efforts to 
reduce nonpoint source pollution, and its conservation practices have 
significantly reduced pollutants coming from agricultural land across 
the country. Notwithstanding its achievements, certain conservation 
practices can adversely affect water quality if not properly 
implemented—for example, by transporting polluted runoff from nutrient-
laden fields into nearby water bodies. The agency’s Natural Resources 
Conservation Service (NRCS) has procedures in place intended to ensure 
that its practices do not inadvertently harm water quality. During its 
field work, GAO identified a few instances where these procedures may 
not have been followed (including in watersheds where EPA’s section 
319 funds had been used), and therefore sought NRCS data to determine 
if they were isolated instances or indicative of a more prevalent 
issue. NRCS’ national level data, however, are not sufficiently 
detailed to identify whether appropriate measures are always in place 
to mitigate potential water quality impacts. According to NRCS, such 
data are instead located in its field offices and are not analyzed by 
the agency. 

What GAO Recommends: 

GAO recommends, among other things, that EPA provide section 319 
oversight guidance to its regional offices and that USDA analyze data 
to determine if measures were taken to mitigate water quality impacts 
in section 319 project areas. EPA agreed with the recommendations, 
while USDA was silent on them. Both agencies commented on specific 
findings, which are addressed within the report. 

View [hyperlink, http://www.gao.gov/products/GAO-12-335]. For more 
information, contact David C. Trimble at (202) 512-3841 or 
trimbled@gao.gov. To view the e-supplement online, click on GAO-12-
377SP. 

[End of section] 

Contents: 

Letter: 

Background: 

States Have Funded Many Successful Projects but Also Some That 
Encountered Preventable Challenges: 

EPA's Oversight and Measures of Effectiveness of States' Programs Have 
Not Consistently Ensured Projects Likely to Yield Measurable Water 
Quality Outcomes: 

A Key USDA Program Largely Complements EPA's Section 319 Program, 
although It Is Unclear How Closely Its Procedures to Protect Water 
Quality Are Followed: 

Conclusions: 

Recommendations for Executive Action: 

Agency Comments and Our Evaluation: 

Appendix I: Objectives, Scope, and Methodology: 

Appendix II: Comments from the Environmental Protection Agency: 

Appendix III: Comments from the U.S. Department of Agriculture: 

Appendix IV: GAO Contact and Staff Acknowledgments: 

Figures: 

Figure 1: A Watershed: 

Figure 2: Percentage of Section 319-Funded Projects Addressing 
Different Categories of Nonpoint Source Pollution, Fiscal Years 2004-
2010: 

Figure 3: Percentage of Section 319-Funded Projects Addressing 
Different Categories of Pollution That Were Designed Primarily to 
Implement Direct Pollution Reduction Approaches, Fiscal Years 2004-
2010: 

Figure 4: Focus of Projects Funded under Section 319 Designed 
Primarily to Implement Indirect Pollution Reduction Approaches, Fiscal 
Years 2004-2010: 

Figure 5: An Eastern Washington Stream Channeled to Maximize Crop 
Yield: 

Figure 6: Restoration of the Headwaters of the Big Darby River in Ohio: 

Abbreviations: 

CRP: Conservation Reserve Program: 

CREP: Conservation Reserve Enhancement Program: 

EPA: Environmental Protection Agency: 

EQIP: Environmental Quality Incentives Program: 

NRCS: Natural Resources Conservation Service: 

TMDL: Total Maximum Daily Load: 

USDA: U.S. Department of Agriculture: 

[End of section] 

United States Government Accountability Office: Washington, DC 20548: 

May 31, 2012: 

The Honorable Timothy Bishop: 
Ranking Member: 
Subcommittee on Water Resources and Environment: 
Committee on Transportation and Infrastructure: 
House of Representatives: 

The Honorable Eddie Bernice Johnson: 
House of Representatives: 

Forty years after the 1972 Clean Water Act recognized the problem of 
water pollution from diffuse, or nonpoint, sources--such as runoff 
from farms or construction sites--nonpoint source pollution remains 
the leading cause of impairment of the nation's waters. The 
Environmental Protection Agency (EPA) reports that more than 33,000 
water bodies nationwide are impaired primarily by such pollution; that 
is, they do not meet water quality standards to provide for, among 
other things, propagation and protection of aquatic wildlife and human 
use and recreation. Runoff from nonpoint sources, including many 
farms, managed forests, and urban areas, often carries harmful 
pollutants such as fertilizers, pesticides, and sediment from fields 
and logging roads; metals and toxins from abandoned mines; and oils 
from roads and highways. Nonpoint source pollution may also alter 
natural water flow patterns, water temperatures, and ecosystems in 
rivers, streams, lakes, estuaries, and oceans, to the detriment of 
aquatic life. 

Recognizing such issues, Congress in 1987 amended the Clean Water Act, 
adding section 319 to explicitly address nonpoint source pollution 
through a cooperative, grant-based program with states, which fund 
projects to reduce nonpoint source pollution and restore impaired 
water bodies. Section 319 authorizes EPA to award federal funds to 
states to implement nonpoint source management programs. Since fiscal 
year 2000, section 319 funding has averaged about $200 million 
annually for projects designed to reduce nonpoint source water 
pollution.[Footnote 1] Guidelines issued by EPA in 2003 for 
implementing section 319 draw particular attention to the potential 
benefits of working closely with the U.S. Department of Agriculture 
(USDA), which administers agricultural conservation programs--a source 
of federal funding that can complement EPA's water quality improvement 
efforts. USDA's conservation programs provide billions of dollars in 
assistance to farmers. In particular, through its Environmental 
Quality Incentives Program, which is implemented by the Natural 
Resources Conservation Service (NRCS), USDA develops contracts with 
agricultural producers to implement conservation practices on working 
agricultural land to, for example, reduce soil erosion and nonpoint 
source water pollution. 

In 1990, when we last reported on EPA's section 319 program,[Footnote 
2] we found that inherent conflicts existed between some federal 
agencies' policies and states' water quality goals. Pollution trends 
since that time suggest that such inherent conflicts and their 
downstream consequences remain today. In 2011, the National Oceanic 
and Atmospheric Administration reported that the "dead zone" at the 
mouth of the Mississippi River--an area in the Gulf of Mexico where 
low oxygen levels can stress or kill resident organisms--was nearly 
7,000 square miles, about the size of the state of New Jersey. In 
2009, the U.S. Geological Survey reported that nine states in the 
Mississippi River basin contribute over 70 percent of the nitrogen and 
phosphorus that lead to such dead zones, largely through fertilizers 
for corn, whose production is supported by USDA's commodity programs. 

Concerns have been raised by federal and state agencies, scientists, 
and nongovernmental organizations about the persistence and severity 
of nonpoint source pollution. In this context, you asked us to review 
the effectiveness of EPA's section 319 program in addressing nonpoint 
source pollution of the nation's waters. Accordingly, our objectives 
were to examine (1) states' experiences in funding projects that 
effectively address nonpoint source pollution problems, (2) the extent 
to which EPA oversees the section 319 program and measures program 
effectiveness in reducing the adverse impacts of nonpoint source 
pollution on water quality, and (3) the extent to which key 
agricultural conservation programs complement EPA's efforts to reduce 
nonpoint source pollution. 

To conduct this work, we reviewed relevant laws, regulations, and 
agency research and guidelines. We visited 8 states, in four EPA 
regions, which we chose on the basis of their varied types of nonpoint 
source pollution and proximity to some of the nation's premier 
watersheds, such as the Chesapeake Bay, Great Lakes, and Mississippi 
River.[Footnote 3] 

* To address the first objective, we surveyed a random sample of 524 
managers of projects that have been implemented with section 319 funds 
to obtain information on the factors influencing project selection, 
implementation, and effectiveness on the ground. Our response rate was 
57 percent. Survey results are generalizable to the 1,273 projects 
that were completed from January 1, 2004, through June 9, 2011; that 
expended section 319 funds; and that involved direct implementation of 
conservation practices or remediation techniques. We also examined 
nearly 2,000 summaries of completed projects in EPA's Grants Reporting 
and Tracking System database, including those that provided funds for 
projects conducted under USDA conservation programs, such as the 
Conservation Reserve Program and the Conservation Reserve Enhancement 
Program.[Footnote 4] Further, we obtained information from nonpoint 
source program officials in all 50 states, including information on 
states' project selection criteria, types of organizations that have 
received funding, and program oversight practices.[Footnote 5] 

* To address the second objective, we obtained from EPA's 10 regional 
offices information on the nature and extent of their oversight of 
state nonpoint source management programs, including the extent to 
which they examine states' project selection processes, annual plans, 
program objectives, and the criteria they use to annually award funds. 
In addition, we examined section 319's statutorily required reporting 
requirements--which EPA uses as national measures of program 
effectiveness--and collected information from state and local staff 
about whether EPA's national measures accurately reflect the water 
quality improvements from nonpoint source pollution control projects. 

* To address the third objective, we analyzed data on conservation 
practices, funded under USDA's Environmental Quality Incentives 
Program, that have been implemented in watersheds where states have 
funded section 319 projects.[Footnote 6] We focused on the 
Environmental Quality Incentives Program in this report because 
improving water quality is among its specific purposes, and EPA 
guidelines encourage states to seek opportunities to leverage 
resources with this program. We reviewed USDA reports on the 
effectiveness of conservation practices, and we also interviewed USDA 
officials familiar with the program. Appendix I describes our scope 
and methodology in more detail. 

We conducted this performance audit from December 2010 through May 
2012, in accordance with generally accepted government auditing 
standards. Those standards require that we plan and perform the audit 
to obtain sufficient, appropriate evidence to provide a reasonable 
basis for our findings and conclusions based on our audit objectives. 
We believe that the evidence obtained provides a reasonable basis for 
our findings and conclusions based on our audit objectives. 

Background: 

Land use substantially affects the type and extent of nonpoint source 
pollution of water bodies. For example, soil erodes naturally from 
undisturbed land, but the amount of erosion can increase manyfold when 
trees are cut or the land is farmed. In addition, when land is used 
for housing or urban development, erosion from land clearing and 
excavation during construction can increase tremendously. Moreover, 
land use activities can also produce toxic pollution. For example, 
pesticide use in farming has resulted in toxic runoff, and mining has 
produced leaching of heavy metals and acid mine drainage.[Footnote 7] 

Nonpoint source pollution can have long-lasting impacts. For instance, 
a heavy rain can wash tons of soil from a field, and the material can 
either scour a streambed or settle out and cover gravel that fish 
spawn in. Long after the water itself clears, populations of fish and 
other aquatic life may still not have recovered. Similarly, when trees 
and bushes are cut next to stream banks, debris falling into the 
stream or washing into the water may initially degrade the water, but 
a longer-term problem may be caused by persistent elevated water 
temperatures resulting from the removal of shade. In time, altered 
water temperatures can make the stream a less sustainable habitat for 
fish and other animals or may make it totally uninhabitable. 

Clean Water Act's Approach to Addressing Nonpoint Source Pollution: 

For nearly 40 years, the Clean Water Act has played a critical role in 
reducing water pollution and improving the health of the nation's 
waters, including rivers, lakes, and streams.[Footnote 8] The purpose 
of the law is to restore and maintain the chemical, physical, and 
biological integrity of the nation's waters. Passed by Congress in 
1972, the act marked a shift in clean water policy, establishing a 
significant federal role in controlling point sources of water 
pollution. Through the 1980s, EPA issued a series of reports, 
including a key 1984 report to Congress, finding that in the decade 
following passage of the act, control of point sources had resulted in 
significant achievements toward water quality goals but that these 
point source reductions had illuminated the nonpoint source 
contribution to water quality problems.[Footnote 9] For example, at 
that time, a majority of EPA regions identified nonpoint sources as 
the principal remaining cause of water quality problems. 

In 1987, Congress amended the Clean Water Act, adding section 319 and 
creating the nonpoint source management program. Section 319 provides 
for annual grants to be administered by EPA and incentives for states 
to develop and implement nonpoint source management programs. Section 
319 includes various minimum conditions that states must meet to 
receive grants, including the development of nonpoint source 
management programs--which EPA must approve--and annual reports on 
states' progress in achieving the goals of their management programs. 
States must also obtain a determination from EPA that they made 
"satisfactory progress" in meeting their goals from the prior year. 
For its part, EPA has discretion by statute to add terms and 
conditions to grants; to require additional information on 
applications; and to request additional information, data, and reports 
it considers necessary to determine continuing state eligibility for 
grants.[Footnote 10] 

EPA Guidance to States in Developing Their Section 319 Programs: 

The 1987 amendments to the Clean Water Act created the nonpoint source 
program, but EPA did not receive an appropriation to implement the 
program until fiscal year 1990. In the early 1990s, EPA focused on 
developing technical support for states' use in developing and 
implementing their programs. During these initial stages, some states 
and EPA regions focused their nonpoint source programs narrowly on 
demonstrations of particular pollution control technologies. In 
response to EPA program guidelines issued in 1996, states upgraded 
their programs, and in so doing, several states incorporated watershed-
based approaches as a significant and sometimes central organizing 
theme. These states focused their pollution reduction efforts in 
specific watersheds--areas of land through which all the rainfall and 
streams flow downhill toward a main river channel (see figure 1). 
According to EPA documents, state nonpoint source programs that 
adopted this approach improved their capacity to solve nonpoint source 
pollution problems. 

Figure 1: A Watershed: 

[Refer to PDF for image: illustration] 

Depicted in the illustration: 

Precipitation; 
Snowpack; 
Tributaries; 
Lake; 
Watershed divide; 
Subbasin; 
Ridge; 
Riparian zone; 
Wetland; 
Percolation; 
Groundwater (aquifer). 

Source: GAO and EPA. 

[End of figure] 

In the early 2000s, EPA encouraged states to sharpen the focus of 
their nonpoint source management programs toward impaired water 
bodies. Specifically, EPA and the states determined they needed to 
target their efforts to reduce nonpoint source pollution within 
defined geographic areas representing the most severe water quality 
problems. According to EPA documents, the two key steps states need to 
solve nonpoint source problems at the watershed level are the 
development of a watershed-based plan that addresses water quality 
needs within a watershed and the actual implementation of the plan. 

In 2003, EPA issued Federal Register guidelines for the section 319 
program, which remain in use today.[Footnote 11] These guidelines 
followed a substantial increase in appropriated funds. Key features of 
the guidelines include the following: 

* The guidelines direct states to use about half their section 319 
grants to develop and implement watershed-based plans for impaired 
watersheds. Beyond targeting funds to geographic areas in need of 
restoration, the guidelines also allow states to fund activities that 
generally support section 319 goals, such as technical assistance, 
staffing, projects that demonstrate innovative approaches to pollution 
reduction, and education programs that promote awareness and changes 
in behavior. 

* EPA's 10 regional offices, which provide oversight of the section 
319 program, are to place special emphasis on reviewing states' 
progress in developing and implementing watershed-based plans 
according to the guidelines. Regional offices are to review and 
discuss with states the projects states select for section 319 funding 
to ensure the plans' effective implementation. It is through this 
review process that regional offices have the opportunity to influence 
the types of projects states select if they believe that the projects 
selected by a state are not adequate to effectively reduce nonpoint 
source pollution.[Footnote 12] 

* States are encouraged to leverage section 319 funds with projects 
from other federal programs that have water quality objectives, 
including USDA's Environmental Quality Incentives Program (EQIP). 
[Footnote 13] This program is designed to fund conservation practices 
on working agricultural land to achieve national priorities, including 
reducing soil erosion and nonpoint source water pollution. The 
guidelines state that section 319 funding is especially suitable for 
supporting activities that either are ineligible for or typically do 
not receive significant USDA funding, including developing watershed-
based plans in impaired watersheds, monitoring water quality, and 
funding staff to work with local communities to help assist and 
promote the development and implementation of watershed-based plans. 

Under EPA's section 319 program, states retain the primary role for 
addressing nonpoint source water pollution. EPA's 10 regional offices 
annually distribute program funds to the states using a formula that 
is weighted heavily toward state population and the number of acres in 
agricultural crop production. States develop their own project 
selection processes and the criteria that their nonpoint source 
management programs will consider when determining what projects to 
fund. Annually, each state submits its list of selected projects to 
the EPA regional office for incorporation into the state's work plan, 
which describes what projects will be funded through section 319. 
Organizations that apply for section 319 funds--often including 
conservation districts, local governments, and nonprofit 
organizations--submit project proposals to states' nonpoint source 
management programs and, if selected, are responsible for implementing 
their proposed nonpoint source pollution projects under an agreement 
with the state. Section 319 is a nonregulatory program, and many 
states therefore rely primarily on voluntary approaches to address 
nonpoint source pollution.[Footnote 14] The programs' nonregulatory 
status, combined with private ownership of much of the nation's land, 
means that securing voluntary landowner participation is a key aspect 
of nonpoint source pollution control and, according to EPA, can 
introduce significant uncertainty in how and when projects are 
implemented. 

States Have Funded Many Successful Projects but Also Some That 
Encountered Preventable Challenges: 

Under EPA's section 319 program, states have funded many projects that 
have helped successfully address nonpoint source pollution and restore 
and protect water bodies across the country, but states have also 
funded projects that have encountered significant challenges--
including many that could have been prevented. This section discusses 
(1) the types of projects states have selected to address various 
categories of nonpoint source pollution, (2) the successes some 
section 319-funded projects have achieved in restoring impaired water 
bodies, and (3) projects states have funded that encountered 
preventable challenges. 

States Have Funded Projects Addressing Varied Categories of Nonpoint 
Source Pollution: 

States' nonpoint management programs have used their annual section 
319 grants from EPA to fund projects that address different categories 
of nonpoint source pollution. The scope of individual section 319 
projects varies considerably; common activities include direct 
implementation of conservation practices, education and outreach 
efforts, water quality monitoring, and funding of state nonpoint 
source management program staff. 

According to data from EPA's Grants Reporting and Tracking System, 
from fiscal year 2004 through fiscal year 2010, states awarded more 
than $1.2 billion in section 319 funds to more than 5,800 projects. 
Projects have been funded in all 50 states and by many tribes and have 
been targeted to seven different categories of nonpoint source 
pollution, mainly agricultural, urban and stormwater runoff, and 
hydromodification (see figure 2). 

Figure 2: Percentage of Section 319-Funded Projects Addressing 
Different Categories of Nonpoint Source Pollution, Fiscal Years 2004-
2010: 

[Refer to PDF for image: horizontal bar graph] 

Category of pollution: Agricultural[A]; 
Percentage of projects addressing the category: 41%. 

Category of pollution: Urban and Stormwater Runoff; 
Percentage of projects addressing the category: 30%. 

Category of pollution: Hydromodification; 
Percentage of projects addressing the category: 12%. 

Category of pollution: Resource Extraction; 
Percentage of projects addressing the category: 4%. 

Category of pollution: Silviculture; 
Percentage of projects addressing the category: 4%. 

Category of pollution: Construction; 
Percentage of projects addressing the category: 4%. 

Category of pollution: Land Disposal[B]; 
Percentage of projects addressing the category: 4%. 

Category of pollution: All Sources[C]; 
Percentage of projects addressing the category: 25%. 

Category of pollution: Other[D]; 
Percentage of projects addressing the category: 18%. 

Source: GAO analysis of EPA data. 

Notes: Because many projects were listed in EPA's Grants Reporting and 
Tracking System as addressing multiple categories of pollution, the 
cumulative percentage of projects in the figure totals more than 100 
percent. Also, some projects included in the tracking system did not 
identify the categories of pollution they were meant to address. The 
percentages displayed in the figure (and discussed in the report) are 
based only on projects whose categories of nonpoint source pollution 
were identified. 

[A] The "agricultural" category includes projects listed in the 
database as addressing, at least in part, pollution from either 
agriculture or animal-feeding operations. 

[B] The "land disposal" category includes projects listed in the 
database as addressing pollution from "land disposal/storage/ 
treatment." 

[C] The "all sources" category is used for projects that address all 
categories of nonpoint source pollution. 

[D] The "other" category includes projects identified in the database 
as addressing pollution from any of the following categories: other 
nonpoint source pollution, historical pollutants, marinas and 
recreational boating, or turf management. 

[End of figure] 

States funded section 319 projects that supported a mix of direct and 
indirect approaches to help restore, protect, or prevent further 
degradation of water quality.[Footnote 15] Direct approaches generally 
involved projects implementing conservation practices or other 
corrective actions to directly reduce or eliminate pollutants entering 
a water body. Common direct approaches included, but were not limited 
to, agricultural conservation practices, such as the installation of 
fences to exclude cattle from shorelines or stream banks; erosion 
control projects; and stormwater discharge projects such as the 
installation of surfaces in parking lots that absorb rainfall rather 
than allow it to run off into urban streams. Indirect approaches 
typically involved activities to help build state and local capacity 
to address nonpoint source pollution, raise public awareness, and 
assess water quality in particular places of concern and commonly 
involved methods such as education and outreach, watershed planning, 
and staffing. 

According to our analysis of EPA data, approximately 45 percent of 
projects that states funded in fiscal years 2004 through 2010 under 
section 319 involved direct approaches and were designed primarily to 
implement activities to directly restore, protect, or prevent further 
degradation of water quality. The categories of pollution that were 
the most common focus of direct restoration approaches--such as 
implementing agricultural conservation practices, stabilizing stream 
banks, or restoring a stream's natural channel configuration--include 
agricultural pollution and pollution resulting from urban and 
stormwater runoff (see figure 3). 

Figure 3: Percentage of Section 319-Funded Projects Addressing 
Different Categories of Pollution That Were Designed Primarily to 
Implement Direct Pollution Reduction Approaches, Fiscal Years 2004-
2010: 

[Refer to PDF for image: vertical bar graph] 

Category of pollution: Agricultural[A]; 
Percentage of projects addressing the category: 52%. 

Category of pollution: Urban and|Stormwater Runoff; 
Percentage of projects addressing the category: 36%. 

Category of pollution: Hydromodification; 
Percentage of projects addressing the category: 17%. 

Category of pollution: Resource Extraction; 
Percentage of projects addressing the category: 6%. 

Category of pollution: Silviculture; 
Percentage of projects addressing the category: 4%. 

Category of pollution: Construction; 
Percentage of projects addressing the category: 4%. 

Category of pollution: Land Disposal[B]; 
Percentage of projects addressing the category: 3%. 

Category of pollution: All Sources[C]; 
Percentage of projects addressing the category: 4%. 

Category of pollution: Other[D]; 
Percentage of projects addressing the category: 19%. 

Source: GAO analysis of EPA data. 

Notes: The percentages displayed in the figure are based on projects 
that listed a direct approach to restoring, protecting, or preventing 
further degradation of water quality as their primary functional 
category of activity and listed the categories of nonpoint source 
pollution they were designed to address. Because many projects were 
listed in EPA's Grants Reporting and Tracking System as addressing 
multiple categories of pollution, the cumulative percentage of 
projects in the figure totals more than 100 percent. 

[A] The "agricultural" category includes projects listed in the 
database as addressing, at least in part, pollution from either 
agriculture or animal-feeding operations. 

[B] The "land disposal" category includes projects listed in the 
database as addressing pollution from "land disposal/storage/ 
treatment." 

[C] The "all sources" category is used for projects that address all 
categories of nonpoint source pollution. 

[D] The "other" category includes projects identified in the database 
as addressing pollution from any of the following categories: other 
nonpoint source pollution, historical pollutants, marinas and 
recreational boating, or turf management. 

[End of figure] 

According to our analysis of EPA data, approximately 55 percent of 
projects that states funded in fiscal years 2004 through 2010 under 
section 319 were designed primarily to implement activities that 
indirectly help restore, protect, or prevent further degradation of 
water quality.[Footnote 16] We classified such projects into the 
following six broad groups based on the type of indirect activities 
they were primarily designed to support (see figure 4):[Footnote 17] 

* Planning activities include the development of various planning 
documents designed to help identify and address nonpoint source 
pollution, such as watershed-based plans and total maximum daily loads 
(TMDL) for pollutants.[Footnote 18] 

* Education activities include statewide and local education and 
information projects, such as educating local officials about the 
causes and effects of nonpoint source pollution or developing nonpoint 
source-related educational curriculums for use in schools. 

* Water quality monitoring and assessment activities include 
biological monitoring and assessments to determine water body health 
and monitoring the effectiveness of conservation practices. 

* Management and staff activities include funding the administrative 
and personnel costs associated with state nonpoint source management 
programs, as well as supporting other program management efforts, such 
as funding watershed coordinators at the local level. 

* Technical assistance activities include activities such as 
engineering assistance related to implementing conservation practices 
provided to state or local entities. 

* Other activities include a variety of regulatory and enforcement 
activities, as well as activities related to groundwater and soil 
analyses. 

Figure 4: Focus of Projects Funded under Section 319 Designed 
Primarily to Implement Indirect Pollution Reduction Approaches, Fiscal 
Years 2004-2010: 

[Refer to PDF for image: pie-chart] 

Planning activities: 25%; 
Education activities: 24%; 
Monitoring and assessment activities: 20%; 
Management and staff activities: 17%; 
Technical assistance activities: 8%; 
Other activities: 6%. 

Source: GAO analysis of EPA data. 

Note: The percentages displayed in the figure are based only on 
projects that listed an indirect approach to restoring, protecting, or 
preventing further degradation of water quality as their primary 
purpose. 

[End of figure] 

States Have Used Section 319 Funds to Achieve Water Quality Benefits 
and Restore Hundreds of Impaired Water Bodies: 

States have used section 319 funds to improve the condition of water 
bodies impaired by nonpoint source pollution. EPA reported that as of 
December 2011, restoration efforts supported by section 319 funds had 
helped 49 states partially or fully restore 356 water bodies that were 
listed under section 303(d) of the Clean Water Act as impaired by 
nonpoint source pollution.[Footnote 19] In addition, many other 
section 319-funded projects continue to restore portions of water 
bodies and may help them attain their water quality standards or 
designated aquatic uses in the near future. According to our survey 
results, project managers for 72 percent of the projects reported that 
their projects accomplished all objectives originally identified in 
the project proposal.[Footnote 20] In addition, water quality 
improvements resulting from section 319-funded projects have been 
demonstrated across a variety of categories of nonpoint source 
pollution. For example: 

* Agricultural runoff: Since 2007, Pennsylvania has awarded more than 
$700,000 in section 319 funds to three projects to help implement the 
Hungry Run Watershed Implementation Plan. Two of these projects focus 
on installing suites of complementary agricultural conservation 
practices on farms located in the watershed, such as stream bank 
fencing, riparian buffers (undisturbed or planted areas along stream 
banks), and cover crops to address agricultural runoff. State and 
local officials we spoke with said that water quality monitoring data 
suggest that as a result of these efforts, Hungry Run may be removed 
from Pennsylvania's list of impaired waters in the future if its water 
quality continues to improve as a result of the department's efforts. 
These projects also illustrate the value of putting multiple 
conservation practices in place at the same location to increase the 
overall effectiveness of water quality restoration efforts, according 
to the project manager. 

* Urban and stormwater runoff: In Michigan, urban development and its 
associated runoff led to the impairment of Malletts Creek, a tributary 
to the Huron River. To help address this impairment, the Michigan 
Department of Environmental Quality has allocated more than $230,000 
in section 319 funds to a project designed to implement stormwater 
conservation practices at a local library containing a portion of the 
creek on its property. Practices implemented under this project 
included a vegetated green roof on the library to absorb rainwater, as 
well as basins filled with native vegetation to help slow the flow of 
water from the library's parking lot into the creek. Water quality 
testing showed that the project's conservation practices reduced the 
pollutants in stormwater leaving the site, including a greater than 40 
percent reduction in copper, lead, oil, and grease, along with a 66 
percent reduction in zinc.[Footnote 21] 

* Hydromodification: In Washington state, the natural flow of 
Jimmycomelately Creek was altered in the past to facilitate farming 
and the construction of roads and buildings. One effect of this 
alteration was a significant decline in chum salmon in the creek 
during the 1990s. Using approximately $300,000 in section 319 funds 
awarded to the Jamestown S'Klallam Tribe, along with several million 
dollars from other federal, state, and local sources, the effort to 
restore Jimmycomelately Creek involved removing roads and buildings, 
altering the creek channel, and planting native vegetation. A tribal 
official involved with this project told us that the section 319 
funding helped the restoration effort include an additional tributary 
of the creek. The creek is still listed as impaired on Washington 
state's 303(d) list, but the tribe reported that more than 4,000 
summer chum salmon returned to the stream to spawn in 2010 (up from 7 
fish in 1999). The tribal official told us that this biological 
indicator is an important sign that the project was successful at 
reducing pollution and restoring water quality in the creek. 

* Resource extraction: In West Virginia, the Department of 
Environmental Protection has used more than $625,000 in section 319 
funds to install treatment systems that remove metals and neutralize 
acidic water draining from abandoned coal mines in the Lambert Run 
watershed. The acidic water drains from the mines into Lambert Run, a 
stream that Department of Environmental Protection staff described as 
having such high concentrations of iron and aluminum that fish and 
other water-dwelling life had not been able to survive before the 
installation of the treatment systems. The treatment systems channel 
mine drainage into ponds, where the acidic water is neutralized 
through contact with limestone, and metal pollutants are removed. As 
of February 2011, EPA reported that the treatment systems have helped 
to restore approximately 2.3 miles of the original 4.4 miles of 
impaired stream, and additional restoration efforts focused on the 
water body were in progress. State officials we spoke with said that 
monitoring data indicate that within 3 or 4 years, the stream may be 
removed from the state's list of impaired water bodies. 

States Have Also Funded Some Projects that Encountered Preventable 
Challenges: 

Some states have directed section 319 funding toward projects that did 
not achieve their objectives, and many projects that did still faced 
challenges. Specifically, projects that relied on voluntary 
participation sometimes did not achieve goals when third-party buy-in 
was not secured in advance. Others sometimes used indirect approaches 
(e.g., community outreach) that did not have a clear connection to 
achieving tangible water quality results. Nevertheless, in recent 
years some states have adopted more rigorous project selection 
processes to avoid these challenges. 

Reliance on Voluntary Participation: 

According to our survey results, project managers for 28 percent of 
all projects that involved implementing conservation practices or 
pollution remediation techniques reported that their projects were 
unable to accomplish all objectives originally identified in the 
project proposal.[Footnote 22] These projects were generally unable to 
implement the desired number or type of conservation practices or to 
implement them in the originally proposed locations. Moreover, of the 
72 percent of the projects that project managers reported achieved 
their originally proposed objectives, almost half did so only after 
encountering significant challenges that prevented them from finishing 
on schedule, staying on budget, or achieving the desired levels of 
pollution reduction. 

Many of the challenges that project staff reported facing resulted 
from bad weather, staff turnover, or other factors outside their 
control. Nevertheless, of the 132 project managers that submitted 
narrative responses to our survey, 71 (54 percent) cited challenges 
that generally could have been identified and mitigated before 
projects were proposed or selected for funding. For example, project 
staff's inability to secure third-party buy-in, such as landowner 
cooperation to implement the projects as intended, was the most 
commonly identified challenge (49 out of 132 responses to the 
question, or 37 percent).[Footnote 23] 

As a result of the challenges that project staff reported facing, 
managers for 29 percent[Footnote 24] of all projects reported that 
their projects required grant revisions to, for example, change the 
type, number, or location of conservation practices to be implemented 
and that 39 percent[Footnote 25] of all projects required grant 
extensions, with most extensions more than 6 months.[Footnote 26] 
(Additional information on the factors that influenced project 
selection and the challenges that project staff reported some projects 
facing can be found in the e-supplement accompanying this report, 
[hyperlink, http://www.gao.gov/products/GAO-12-377SP]. 

The following examples illustrate projects that project staff reported 
could not be implemented as planned because third-party participation 
was not secured in advance: 

* A project on the Illinois River was to reduce pollution by 
implementing conservation practices in urban and forested areas, such 
as rain gardens planted with native plants, which absorb urban and 
stormwater runoff, and prescribed burns on forested lands.[Footnote 
27] The state of Illinois provided section 319 funds to a regional 
planning organization to (1) put in place 2,500 urban and stormwater 
management practices and (2) implement prescribed burns on 1,000 
forest acres. The organization, however, did not implement the project 
as proposed because, after receiving funding, it was unable to compel 
landowners to implement the practices on private property; the 
organization had not secured the landowners' consent before applying 
for the funding. As a result, of the intended 2,500 urban stormwater 
practices, 11 were ultimately implemented, and forest conservation 
measures were implemented on 282 of the intended 1,000 acres because 
prescribed burns could not be done. 

* In West Virginia, Department of Environmental Protection officials 
selected a project that was to subsidize the cost to homeowners of 
pumping and replacing damaged septic systems in rural areas, among 
other practices. The department awarded a nonprofit organization 
nearly $450,000 to implement the project, of which $285,000 was for 
the septic system component. A departmental official explained that 
the project was designed to be a 5-year project--3 years for 
implementation and 2 years for monitoring. In 2011, the project was in 
its third year of implementation, but as of November 2011, even though 
there was no cost to homeowners for the project, a single homeowner 
had signed up to have a septic system replaced. A representative from 
the nonprofit organization said that when the project was proposed, 
project staff had not verified whether landowners in the project area 
would participate. Upon receiving section 319 funds, the 
representative told us that the project did not receive much interest 
from the community and that her organization may have to return 
unspent funds to the Department of Environmental Protection to be 
reallocated to another project. 

* In Arizona, Department of Environmental Quality officials selected a 
project that aimed to install hundreds of native willows in two areas 
on the shore of a lake impaired by high levels of nitrogen and 
phosphorus. The department awarded section 319 funds to a nonprofit 
organization to plant the willows, whose root systems would absorb 
polluted runoff and thus help prevent nutrients from entering the 
lake. The project manager reported that after section 319 funds were 
received, project staff advertised the project to community members 
and reached out to landowners near the lake. Some landowners, however, 
were unwilling to have the nonprofit organization plant willows on 
their properties, according to the project manager's report. Some 
willow trees were ultimately planted in one of the two areas, but the 
report concluded that the project's water quality goals were not 
achieved and that no measurable reduction in pollutants came from the 
planting. 

Securing voluntary third-party cooperation ahead of time is 
particularly important with agricultural projects, whose success 
largely depends on the implementation of a suite of complementary 
conservation practices. Agricultural projects by their nature can be 
particularly challenging for securing third-party participation to 
implement them in ways that effectively reduce nonpoint source 
pollution.[Footnote 28] According to EPA officials and USDA research, 
[Footnote 29] agricultural projects address nonpoint source pollution 
best when conservation practices are implemented as part of a suite of 
complementary practices. When landowners choose to implement 
conservation practices without all of the proper companion practices, 
however, assurance may be reduced that the practices will result in 
the intended water quality benefits, according to state environmental 
protection officials we spoke with.[Footnote 30] 

The need for complementary conservation practices to protect water 
quality is often evident on lands where livestock graze. On such land, 
for example, installing livestock exclusion fencing in isolation does 
not always ensure that the negative water quality effects of grazing 
will be significantly reduced unless additional practices are put in 
place.[Footnote 31] Livestock exclusion fencing is largely ineffective 
without riparian buffers, which help prevent stream bank erosion and 
absorb nutrients from manure, according to the nonpoint source 
coordinator for West Virginia's Department of Environmental 
Protection. When riparian buffers are absent, livestock still 
congregate near stream banks, altering stream ecosystems and allowing 
sediment and manure to enter the water. 

Pennsylvania's Department of Environmental Protection funded a project 
that demonstrated the success of enlisting landowner cooperation to 
install a suite of complementary practices to protect water quality on 
land where livestock graze. For example, the department awarded $1.2 
million in section 319 funds to a project in Mifflin County, 
Pennsylvania, in which a conservation district installed suites of 
conservation practices on several farms to keep livestock from 
congregating near an impaired stream and entering it. Such practices 
included livestock exclusion fencing, riparian buffers, stream 
crossings,[Footnote 32] and off-stream watering facilities for 
livestock. In this case, conservation district staff said they 
proposed this project because the landowners were willing to install 
all of the complementary practices needed to keep livestock away from 
the stream. Further, this project is likely to help restore the stream 
to such a condition that it will soon be removed from the state's list 
of impaired waters, according to the project manager.[Footnote 33] 

State nonpoint source management program officials in one state we 
visited said that because section 319 is a voluntary program, they 
hesitate to require participating landowners who are willing to 
install a given conservation practice to also install all the 
complementary practices that may be needed to ensure that water 
quality is protected. For example, in Arkansas, the state's nonpoint 
source management program encourages certain agricultural conservation 
practices to be installed together as a suite to maximize water 
quality benefits, but it does not require section 319 fund recipients 
to implement all these practices on agricultural land. One state 
official acknowledged that this process does not always result in 
implementing the conservation practices that would produce the 
greatest water quality benefit and that landowners may resist 
implementing the most effective conservation practices. Consequently, 
less effective practices are sometimes chosen for projects to ensure 
sufficient landowner participation, he said. 

Our analysis of EPA data indicated that section 319 funds may have 
sometimes paid for conservation practices to be put in place without 
all of the proper companion practices. Our analysis of EPA's Grants 
Reporting and Tracking System data showed that when section 319 
funding was used to install livestock fencing, more than half the 
time, the fencing may have been installed without all of the proper 
companion practices--those practices that state environmental 
protection officials and USDA research show are needed to reduce 
nonpoint source pollution from livestock grazing. Specifically, in 
projects from 2004 through 2010 where about 700 separate livestock 
exclusion fences were installed, the following additional practices 
were installed with the fencing: about 50 stream crossings (7 
percent), 260 riparian buffers (37 percent), and 225 watering 
facilities or troughs (32 percent).[Footnote 34] (The data do not 
reflect instances where additional practices may have already been in 
place before the fencing was installed.) 

Indirect Approaches Not Always Linked to Tangible Water Quality 
Results: 

Many section 319-funded projects that use indirect approaches to 
address nonpoint source pollution aim to educate local landowners 
about nonpoint source pollution problems or support other activities, 
such as promoting enrollment in USDA conservation programs.[Footnote 
35] Such projects are allowable under EPA guidelines and are often the 
key to achieving local buy-in and commitment to successfully 
implementing future nonpoint source pollution reduction activities, 
according to EPA headquarters officials. State environmental officials 
in several of the states we visited, however, cautioned that such 
projects can sometimes have limited assurance of generating meaningful 
water quality results when their approaches are not directly aligned 
with addressing nonpoint source pollution or are not linked to 
specific outcomes. Local education and technical assistance activities 
may support improvements to water quality by educating community 
members or developing institutional capacity to address nonpoint 
source pollution, but as shown in the examples below some projects 
offer little assurance that measurable improvements will actually be 
achieved. In some instances, state officials have used the lessons 
learned from funding such projects to help inform their decisions on 
whether to fund similar projects in the future. 

* One section 319-funded project in Arkansas had the stated purpose of 
training teachers and conservation district employees to teach a 
conservation education curriculum, so that they in turn could 
encourage students to participate in a conservation program, thereby 
developing students' appreciation and awareness of natural resources. 
Although designed to teach students about nonpoint source pollution, 
the project did not have a clear link to tangible reductions in 
nonpoint source pollution or changes in behavior stemming from the use 
of section 319 funds. According to the state's nonpoint source program 
director, his staff found that while the curriculum materials this 
project funded were well received by many teachers, staff were unable 
to ascertain the effectiveness of the project and whether it resulted 
in any behavioral change. 

* A project in California aimed to implement conservation practices in 
a creek that was impaired in part by sewage disposal from septic 
systems. To do so, the project sought to increase community education 
through several outreach initiatives, among other activities. Project 
staff held workshops and community events, which about 200 people 
attended, and conducted follow-up surveys, which showed that 80 to 90 
percent of residents had increased their understanding of 
environmental conditions and watershed pollution. Nevertheless, in the 
project's final report, project staff concluded that changing the 
habits of residents to actually implement conservation practices--such 
as pumping septic systems, planting streamside vegetation, and 
limiting fertilizer use--was much more complicated than originally 
anticipated. Despite increased levels of awareness, the report 
concluded, few conservation practices were implemented. 

* In West Virginia, the state nonpoint source program director told us 
that the state's department of environmental protection used section 
319 to fund a project to promote best management practices on oil and 
gas drilling sites and access roads. The project was designed to 
establish a training program for company inspectors that would help 
them identify drilling sites and access roads contributing to sediment 
runoff into impaired waters. The training would help company 
inspectors promote the design and implementation of management 
practices in these areas to reduce runoff and encourage compliance 
with such practices. This project promoted positive practices, but 
according to a departmental official, few management practices were 
ultimately implemented. In addition, he said, the project was not 
directly linked to specific water quality outcomes, such as the number 
of sites and roads on which practices were to be implemented, and he 
will likely not use section 319 funds to fund similar projects in the 
future. 

Projects using section 319 funds to pay for staff to promote 
enrollment in USDA conservation programs have also sometimes lacked 
approaches and intended outcomes that were directly aligned with 
addressing nonpoint source pollution.[Footnote 36] Project managers 
with whom we spoke said that when such projects do not contain 
concrete deliverables, such as the number and specific location of 
acres enrolled in conservation programs, there is limited assurance 
that the most vulnerable land will be protected. For example, Kansas 
awarded $225,000 in section 319 funds to a project whose objectives 
were to provide dedicated assistance to USDA's Conservation Reserve 
Program (CRP) and eliminate delays in providing program enrollment 
assistance to agricultural crop producers.[Footnote 37] Funding was 
used to hire from 24 to 30 part-time staff to help interested 
landowners enroll in the CRP program, complete paperwork, and enroll 
acres in the program in shorter-than-typical time frames. According to 
the state project manager, the project did not include a specific 
focus on promoting CRP in areas of concern to nonpoint source water 
pollution. She also told us that the state Department of Health and 
Environmental Services has since concluded that projects such as these 
are not always the most cost-effective because the number of 
vulnerable acres enrolled has not been proportional to the resources 
invested. As a result, the number of funded staff decreased to 8 the 
next time the project was proposed, and the state project manager said 
that such projects may not be funded in the future. 

Similarly, West Virginia's Department of Environmental Protection used 
section 319 funds for several projects aimed, in part, at enrolling 
landowners in USDA's CRP and Conservation Reserve Enhancement Program 
(CREP).[Footnote 38] According to the state's nonpoint source program 
director, the department has not had much success in reducing nonpoint 
source pollution when section 319 funds are used to enroll land in CRP 
and CREP because pieces of land contributing the largest amounts of 
agricultural runoff are not always enrolled. West Virginia's nonpoint 
source program director told us that as a result, he will be reluctant 
to use section 319 funds on projects aiming to enroll landowners in 
these programs. 

In November 2011, EPA issued a national evaluation of the 319 program 
and found that states' success in controlling agricultural nonpoint 
source pollution when funding these types of indirect projects has 
been mixed.[Footnote 39] The report noted that on one hand these 
projects can help develop and strengthen key partnerships among 
federal programs and are critical for making significant progress in 
remediating large numbers of water bodies impaired by nonpoint source 
pollution. Coordinated efforts between state nonpoint source programs 
and NRCS state conservationists has occurred in about one-half of all 
states' farm bill program funding, which is distributed in whole or in 
part in accordance with the states' nonpoint source program goals and 
priorities, according to the report.[Footnote 40] But EPA also 
reported that many states have had difficulty obtaining significant, 
broad-based, recurring support from USDA programs for nonpoint source 
program priorities. Many states therefore identified improved 
coordination and collaboration with USDA programs as a key nonpoint 
source program goal, according to EPA's report. 

Preventing Challenges through More Rigorous Project Selection: 

Our review of some states' experiences and EPA's 2011 evaluation 
report has shown that the challenges associated with third-party 
participation and projects whose approaches are not clearly linked to 
tangible water quality results can largely be avoided when states use 
more rigorous project selection processes. For example, in 2006, 
Ohio's nonpoint source management program staff examined the types of 
organizations that commonly received its section 319 funding and found 
that more than 70 percent of funding had gone to soil and water 
conservation districts, county health departments, and regional 
planning agencies, which typically have little authority to address 
water quality problems, according to Ohio's nonpoint source program 
director. Ohio's program staff found that these organizations often 
used section 319 funding to pay for staff salaries without getting a 
proportionate improvement in water quality from the projects they 
implemented. Further, when funds were used for implementing projects, 
these organizations' projects were not implemented as anticipated in 
the grant application--typically resulting in substantially less water 
quality improvement than originally intended. 

For example, in 2004 Ohio's EPA allocated $500,000 under section 319 
to one agricultural project to install 15 different complementary 
conservation practices in an impaired watershed. The organization that 
received the grant, however, was unable to convince farmers in the 
impaired watershed to adopt the conservation practices, and after 
numerous grant revisions and extensions, it implemented 3 of the 15 
planned conservation practices, according to Ohio's nonpoint source 
management program director. Instead, he told us, the majority of 
funding was used to purchase 111 pieces of equipment for tilling 
fields and handling and transporting manure, which can help reduce the 
amount of nutrient runoff entering nearby water bodies. What had 
started as a comprehensive approach to changing farmers' behavior and 
farming practices evolved into an agricultural equipment acquisition 
project because agricultural equipment was what the landowners wanted, 
he explained. In addition, Ohio's program review showed that section 
319 funds were being used to support payroll for 27 full-time-
equivalent staff positions in local organizations for some projects 
whose indirect approaches did not contain objectives or deliverables 
that addressed nonpoint source pollution problems, according to the 
nonpoint source management program director. Further, the nonpoint 
source management program director told us that the water quality 
results were not proportionate with the investment. 

Following the 2006 review, Ohio's nonpoint source management program 
changed its project selection criteria to favor grantees with 
authority to implement projects on the ground or projects for which 
any necessary landowner buy-in was secured in advance. Ohio EPA's 
application process now requires that specific properties for proposed 
projects be identified before grants are awarded and that assurances 
from property owners are obtained in advance, so that conservation 
practices can be implemented as planned. These programmatic changes 
have also prompted Ohio's nonpoint source program to fund fewer 
projects that rely on indirect approaches. Consequently, since 2007, 
Ohio's nonpoint source program has funded fewer than 5 full-time-
equivalent positions in local organizations. Ohio EPA's nonpoint 
source program director told us that these changes have helped the 
agency make significant progress in achieving the state's water 
quality goal.[Footnote 41] 

Other states have changed their project selection processes along the 
same lines. In its November 2011 report, EPA found that as part of 15 
states' project selection processes, nonpoint source program staff 
coordinate with project staff at the local level before selecting 
proposed projects for section 319 funding.[Footnote 42] The report 
goes on to say that this preproposal coordination helps increase local 
understanding of state nonpoint source program priorities, identifies 
potential project partners, gauges local receptivity to projects, and 
provides greater assurance of potential project success. These efforts 
typically improve the quality of proposals and, ultimately, water 
quality results from section 319-funded projects, according to EPA's 
report. For example, Colorado's nonpoint source management program 
increased the rigor of its project selection process by working with 
local officials to identify the highest-priority water quality issues 
so that the state can better support the projects that will be most 
effective in addressing them. 

EPA's Oversight and Measures of Effectiveness of States' Programs Have 
Not Consistently Ensured Projects Likely to Yield Measurable Water 
Quality Outcomes: 

EPA regional offices have varied widely in the extent of their 
oversight and the amount of influence they have exerted over state 
nonpoint source management programs. In addition, EPA's primary 
measures of effectiveness of state management programs may not always 
demonstrate the achievement of program goals--which are to (1) 
eliminate remaining water quality problems and (2) prevent new threats 
from creating future impairments--or reflect the achievements of some 
critical state activities for reducing nonpoint source pollution. 

EPA's Regional Offices Have Varied in Their Oversight of States' 
Nonpoint Source Management Programs: 

EPA's 10 regional offices varied in their oversight of states' 
nonpoint source management programs and the extent to which they 
influenced the projects states funded through section 319. This 
variability is seen most notably in regional offices' reviews of 
states' annual work plans and project selection criteria, which are to 
describe the activities that states' nonpoint source management 
programs plan to undertake in the upcoming year and the parameters for 
which projects are eligible to receive section 319 funds from the 
state. To oversee states' nonpoint source management programs, EPA 
regional offices by regulation are to determine, before annually 
awarding section 319 funds to states, that achievement of states' 
proposed work plans is feasible--which means that states are to 
demonstrate that the projects described in the work plan can be 
implemented.[Footnote 43] Officials from most regional offices 
reported to us that they do not assess the feasibility of specific 
projects. Nevertheless, regional offices have almost always determined 
that states have made satisfactory progress in achieving their program 
goals--a condition that must be met for states to receive section 319 
funding the following year. 

Variations in EPA Regional Office Oversight: 

Among their responsibilities for oversight of state nonpoint source 
management programs as provided in regulation and guidance, EPA's 10 
regional offices perform two key functions. First, they review state 
nonpoint source management program plans, which are to identify 
states' goals for addressing nonpoint source pollution. Second, they 
review states' annual work plans and project selection criteria, which 
are to describe the activities that states' nonpoint source management 
programs plan to undertake in the upcoming year to meet program goals 
and the parameters for which projects are eligible to receive section 
319 funds from the state. 

Regional offices have varied in the extent of their review of states' 
nonpoint source management program plans, which are to ensure that 
states align the goals of their programs with the highest-priority 
water quality impairments. Some regional offices have encouraged 
states to modify their plans, whereas other regional offices have not. 
For example, regional office officials reported to us that Region 5 
has encouraged all of its states to revise their management plans 
within the past 5 years,[Footnote 44] whereas Region 1 has not 
encouraged any of its states to update their plans since 
1999.[Footnote 45] Section 319 guidance to states indicates that they 
should update their nonpoint source management plans if EPA finds that 
the practices and measures proposed in such plans are not adequate to 
reduce the level of nonpoint source pollution. Overall, EPA's 2011 
report evaluating states' implementation of the section 319 program 
found that EPA regional offices have not required 28 states to upgrade 
their nonpoint source management program plans since 1999 or 2000. 
[Footnote 46] As a result, according to the report, these states' 
plans play a diminished role or are simply ignored in the current 
implementation of the states' programs and do not adequately reflect 
innovations that have become available during the past decade, 
including watershed-based planning and low-impact development. One 
primary reason for the variations in oversight among the regional 
offices could be that EPA headquarters has not issued specific 
implementing guidance to the 10 regional offices on how they are to 
fulfill their regulatory oversight responsibilities for the 319 
program.[Footnote 47] 

Regional offices have also varied in the extent of their review of 
each state's annual work plans and project selection criteria. 
Officials from three regional offices told us that they reviewed 
annual work plans in depth and played an active role in influencing 
the types of projects selected. For example, Region 4 officials 
reported that they helped ensure that several states within the region 
targeted their section 319 funds to severely impaired watersheds and, 
before granting funding, selected project applicants who were willing 
to implement and capable of implementing projects.[Footnote 48] 
Specifically, state program officials in North Carolina established a 
priority-setting system in collaboration with the regional office, 
which helped state staff review and rank project proposals. Final 
project selections were generally made after proposals were ranked and 
reviewed by various stakeholders, including Region 4 officials. An 
official with another regional office told us that this kind of 
process provides regional offices with the opportunity to guide states 
to spend funds on certain geographic regions or to encourage 
implementation of projects in watersheds that have a watershed 
implementation plan and willing landowners. 

In contrast, of the 7 other regional offices, 4 reported that they 
review project selection criteria on an "as needed" basis, and 3 
regions reported that they have limited to no involvement in reviewing 
or approving the project selection criteria. When reviewing project 
selection criteria and work plans, EPA's regional offices have an 
opportunity to influence how state funding is distributed, determine 
which projects are selected to receive section 319 funds, reject 
proposed projects, or require modifications to the state programs' 
annual goals.[Footnote 49] Nevertheless, most of the 10 regions 
reported that although they provide feedback to states on specific 
project proposals before states select projects, they do not 
systematically assess the merits and feasibility of specific 
projects.[Footnote 50] As noted above, EPA headquarters has not issued 
specific guidance to the 10 regional offices, including on how to 
review states' plans for project feasibility and criteria to ensure 
that funded projects have characteristics that reflect the greatest 
likelihood of effective implementation and tangible water quality 
results. 

EPA Regional Offices Almost Always Approve State Programs for Annual 
Funding: 

Notwithstanding the variation in regional offices' reviews of nonpoint 
source management program plans and annual work plans, regional 
offices have almost always determined that states have made 
satisfactory progress in achieving their program goals, which states 
must do to receive section 319 funding the following year. Regional 
office officials told us that it is more common for regional office 
staff to work with states that are at risk for not achieving 
satisfactory progress than to withhold funds. For example, according 
to Region 8 officials, the regional office almost withheld its 
determination of satisfactory progress from Wyoming for not having 
developed satisfactory watershed-based plans, but it granted the 
determination instead and then worked with Wyoming state program staff 
to direct future funds toward developing TMDLs with implementation 
plans that would include all the elements of satisfactory watershed-
based plans. Officials from regional offices reported to us that 
determinations of satisfactory progress were made according to a 
variety of factors specific to each state's program, such as the 
number of projects completed and reductions in pollutant loads, which 
states commonly reported in nonpoint source management program 
reports.[Footnote 51] Officials with one regional office told us that 
the determination of satisfactory progress is a fairly low bar and 
that they were generally reluctant to withhold this determination 
because states would then not receive funds to address a significant 
water quality problem. 

EPA's Primary Measures of Effectiveness May Not Fully Demonstrate 
Program Goal Achievements: 

In addition to requiring states to report on their progress in meeting 
milestones for their nonpoint source management programs, section 319 
requires states to annually report to EPA on two measures of 
effectiveness resulting from implementation of their management 
programs: (1) reductions in loadings of specific nonpoint source 
pollutants and (2) improvement in water quality of water bodies 
identified on states' lists of impaired waters as requiring nonpoint 
source controls to meet water quality standards. Section 319 does not 
limit EPA to these two measures of effectiveness, but the agency has 
chosen to use these two reporting requirements as barometers of 
success for the section 319 program. Specifically, EPA provides 
nationwide data on these two measures to Congress to report on the 
progress that the section 319 program makes each year toward achieving 
the program's goals of (1) eliminating remaining water quality 
problems and (2) preventing new threats from creating future 
impairments. Reporting on the two measures of effectiveness is 
statutorily required, but as described in the 2003 Federal Register 
guidelines and other EPA documents, states can demonstrate the 
achievements of nonpoint source management programs in additional 
ways--ways that in many respects may provide a more accurate picture 
of environmental outcomes and reflect the achievements of some 
critical state activities for reducing nonpoint source pollution, such 
as the number, kind, and condition of living organisms in the water. 

Reporting Requirements Emphasize Reductions of Specific Nonpoint 
Source Pollutants: 

For the first national performance measure of effectiveness--
reductions in loadings of specific nonpoint source pollutants--EPA 
requires states to provide information on the amounts by which 
nitrogen, phosphorus, and sediment have been reduced in water bodies 
where section 319 projects targeting such pollutants have been 
implemented.[Footnote 52] According to several state environmental 
protection officials, EPA's focus on these reductions as one of two 
primary reporting requirements has inherent limitations. For instance, 
it has encouraged some states to design their nonpoint source 
management programs and select projects to maximize reductions in 
specific pollutants, but the projects and activities associated with 
reducing these pollutants do not always address the root cause of a 
nonpoint source pollution problem that prevents living organisms from 
inhabiting the water. One example of this limitation is the manner in 
which states have used section 319 funds to mitigate the effects of 
channeling streams in areas of agricultural production. Stream 
channeling occurs when farmers straighten creeks and streams to 
maximize the amount of land that can be farmed and make it easier to 
move machinery across fields (see figure 5). Channeling streams, 
however, removes vegetation along stream banks, alters streambed 
configuration and water flows, and disrupts stream food webs and other 
life-supporting systems, according to EPA documents, and such effects 
may extend far downstream. 

Figure 5: An Eastern Washington Stream Channeled to Maximize Crop 
Yield: 

[Refer to PDF for image: photograph] 

Source: GAO. 

[End of figure] 

Projects to mitigate the adverse effects of channeled streams may 
yield large pollutant reductions, but because they do not address the 
straightening of the streams, the goal of healthier water bodies--as 
measured by biological indicators, such as the number, kind, and 
condition of living organisms in the water--is not always achieved. 
For example, to absorb runoff and help reduce loads of nitrogen, 
phosphorus, and sediment, section 319 funds have often been used for 
projects that install 20-to 120-foot-wide grass filter strips as a 
buffer between cropland and adjacent water bodies. These filter 
strips, however, do not solve the water quality problems caused by the 
loss of streamside vegetation and altered streambed configuration, 
according to an engineer with Ohio's nonpoint source management 
program.[Footnote 53] From 2001 to 2005, one midwestern state used 
section 319 funds to install more than 40,000 feet of grass filter 
strips, which were estimated to have reduced nitrogen by more than 
500,000 pounds and phosphorus by 168,000 pounds annually. Yet none of 
the watersheds where the grass filter strips were installed 
demonstrated any measurable improvement in stream health (since they 
were installed 7 to 11 years ago), as indicated by the number, kind, 
and condition of living organisms in the water, according to the 
engineer who monitored the stream. 

Some state officials mentioned their concerns when this performance 
measure drives project selection. For example, an official with Iowa's 
Department of Natural Resources told us that focusing on this 
reporting requirement has compelled his state's nonpoint source 
management program to select projects that are likely to substantially 
reduce nitrogen, phosphorus, and sediment, even though such reductions 
may not address actual causes of stream impairment or improve the 
condition of water bodies for aquatic life. In addition, an official 
with Michigan's Department of Environmental Quality told us that he 
generally has the flexibility to choose projects that are the most 
cost-efficient and effective at addressing nonpoint source pollution. 
To satisfy EPA, however, he told us he feels compelled to use section 
319 funds for several projects each year that may not be the most 
important in addressing nonpoint source pollution problems but that 
are likely to yield large reductions in pollutants. 

In contrast, we also found limited instances in which states funded 
projects (such as stream restoration) that, rather than concentrating 
on reductions of specific pollutants, sought instead to address the 
underlying causes of impairment. One project in Ohio in 2011 involved 
the headwaters of the Big Darby River, which had been channeled more 
than a century ago to help agricultural producers increase crop 
yields. The channeling, however, removed streamside trees and shrubs 
and altered the stream's flow, thereby increasing water temperatures 
and reducing the stream's ability to support life. Furthermore, the 
officials said that numerous local agricultural activities contributed 
nitrogen, phosphorus, and sediment to the stream. Project staff did 
not focus solely on the pollutants used by EPA as key measures; 
instead, their holistic design focused on restoring the natural 
configuration and flow of about one mile of the stream's headwaters. 
Project engineers told us that when this restoration is completed, the 
stream will be able to better assimilate pollutants, water temperature 
will fall, stable stream banks will reduce erosion, and structure and 
habitat for living things will be restored for approximately 75 
downstream miles (see figure 6). 

Figure 6: Restoration of the Headwaters of the Big Darby River in Ohio: 

[Refer to PDF for image: photograph] 

Source: The Nature Conservancy. 

[End of figure] 

EPA research and guidelines over the past several decades acknowledge 
the advantages of incorporating biological indicators (e.g., the 
number, kind, and condition of living organisms) into state water 
quality programs to better reflect environmental outcomes.[Footnote 
54] In 2005, EPA stated that the most direct and effective measure of 
the integrity of the water body is the status of its living systems 
and that the use of biological information can help states improve 
water quality protection.[Footnote 55] Moreover, EPA's 2003 Federal 
Register guidelines list demonstrable improvements in biological or 
physical parameters--such as increased diversity in fish or insect 
populations or improved riparian areas--as a key method for measuring 
the progress and success of state nonpoint source programs.[Footnote 
56] 

Nonetheless, despite the advantages of implementing projects that 
address the varied root causes of water bodies' impairment and 
associated downstream effects, EPA data show that state nonpoint 
source management programs have far more often funded practices that 
generally reduce pollutant loads than those more directly linked to 
improving the number, kind, and condition of living organisms in water 
bodies. For example, the installation of filter strips and other 
similar practices have been funded more often than stream restoration 
projects--at least 1,000 projects compared with about 175--since 2004, 
even though stream restoration projects generally result in more 
living organisms in the water. 

States Must Report on the Number of Water Bodies Removed from State 
Lists of Impaired Waters: 

For the second national performance measure of effectiveness--the 
improvement in water quality of water bodies identified on state lists 
of impaired waters as requiring nonpoint source controls to meet water 
quality standards--EPA asks states to provide information on the 
number of water bodies that are removed from these lists. EPA tracks 
this number to document how states' efforts are improving water 
quality across the nation and to demonstrate to Congress the program's 
annual progress in reducing nonpoint source pollution. Since 2000, 
states have removed more than 350 water bodies from their lists of 
impaired waters.[Footnote 57] 

EPA's focus on this second measure may also influence state project 
selection in ways that may reduce the effectiveness of their nonpoint 
source management programs. According to four officials from the eight 
states we visited, this focus has compelled state program staff to 
choose projects for water bodies that were close to being restored and 
removed from the states' 303(d) lists of impaired waters, rather than 
solely those that are most degraded. For example, Maryland's nonpoint 
source program manager told us that the program's project selection 
committee targeted projects for water bodies close to removal from the 
state's list of impaired water bodies over water bodies with more 
serious water quality problems, where such funding could have had a 
greater impact on nonpoint source pollution. He told us that such 
projects are selected because EPA expects each state to "deliver" 
several such successes each year. According to the official, each case 
has its merits. The former is more likely to result in meeting 
standards for the particular water body. The latter demonstrates 
incremental improvement and typically has a greater pollutant load 
reduction, which benefits downstream waters like the Chesapeake Bay. 

The emphasis on this measure of effectiveness has also encouraged some 
state program staff to focus on restoring impaired water bodies even 
when they have determined that greater benefit could be achieved by 
protecting high-quality water bodies not yet listed as impaired. 
Environmental protection officials from several states we visited told 
us that if EPA put a greater emphasis on protecting high-quality water 
bodies, they would likely select some different projects on water 
bodies that are not yet impaired but are threatened by nonpoint source 
pollution. Instead of focusing solely on EPA's two performance 
measures, Maine's Department of Environmental Protection requested and 
received permission from EPA to fund projects that focus on preventing 
pollution of lakes, streams, and coastal waters by, for example, 
providing training opportunities to advance stream protection efforts. 
The state's nonpoint source program's annual report for 2010 reported 
that such a focus is far more cost-effective than the long-term 
investments needed to restore waters once they become polluted. 

EPA's former nonpoint source pollution chief acknowledged that EPA's 
emphasis on the two statutorily required national performance measures 
makes it difficult to judge which states are making more progress than 
others in addressing nonpoint source pollution. He said that 
protecting undisturbed lakes and streams is critical for protecting 
aquatic life but that such projects rarely demonstrate substantial 
reductions in pollutant concentrations. He also said that, conversely, 
reductions in pollutants are reported each year in some states without 
associated improvements in biological indicators. EPA does not require 
states to provide information on their progress under the nonpoint 
source program in improving water bodies' condition for aquatic life 
or the protection of high-quality water bodies. 

In EPA's November 2011 program evaluation, the agency reported that it 
will issue new section 319 guidelines to states in fiscal years 2012 
to be implemented in 2012 and 2013. EPA reported that these guidelines 
will generally address program accountability but did not specify 
whether such accountability will include measures to more accurately 
reflect the overall health of targeted water bodies (e.g., the number, 
kind, and condition of living organisms) or demonstrate states' focus 
on protecting high-quality water bodies, where appropriate. 

A Key USDA Program Largely Complements EPA's Section 319 Program, 
although It Is Unclear How Closely Its Procedures to Protect Water 
Quality Are Followed: 

USDA's Environmental Quality Incentives Program (EQIP) is the key 
agricultural conservation program that can complement EPA's efforts to 
reduce nonpoint source pollution.[Footnote 58] According to the 
Department's Natural Resources Conservation Service (NRCS), which 
manages the program, it has resulted in substantial pollutant 
reductions in key watersheds across the country.[Footnote 59] 
Nonetheless, certain conservation practices under the program could 
adversely affect water quality if installed without the proper suite 
of companion practices to mitigate these adverse effects. NRCS 
officials maintain that its procedures ensure that conservation 
practices conserving one resource (e.g., soil) do not inadvertently 
harm another (e.g., water), and that its quality control measures 
ensure they are followed at the ground level. Our analysis of the EQIP 
data shows that nutrient management plans and other conservation 
practices of one kind or another have often been put in place. The 
EQIP data, however, is kept at an aggregated level and does not reveal 
which mitigation measures are applied for site-specific conditions on 
the ground--information which is necessary to determine whether the 
mitigation measures are effective. 

NRCS Soil and Water Conservation Practices and Procedures Are Intended 
to Prevent Certain Practices from Adversely Affecting Water Quality: 

Under EQIP, NRCS funds conservation practices throughout the nation 
that are intended to reduce nonpoint source pollution and, among other 
things, soil erosion.[Footnote 60] NRCS has developed standards for 
each conservation practice, which provide science-based criteria that 
participating agricultural producers are supposed to follow to address 
soil, water, air, plant, animal, and energy resource concerns. Each 
practice's potential effects on soil, water, and air quality are 
documented in the agency's assessments of the conservation practices' 
physical effects. Almost one-half the conservation practices can 
address nonpoint source water pollution, according to officials in 
NRCS' Office of Science and Technology. 

According to NRCS technical documents, some conservation practices 
could have unintended, negative effects on water quality if installed 
without the proper "companion practices" capable of mitigating the 
potential negative effects. This is due to the fact that EQIP-funded 
practices may have distinct purposes, such as to reduce soil loss or 
improve soil conditions for agriculture, which are not oriented toward 
improving water quality. For example, NRCS often funds underground 
outlet systems, which help move surface water to a "suitable outlet," 
such as a drainage ditch, to reduce soil erosion. Such systems can 
help conserve soil, but NRCS conservation practice physical effects 
assessments show that such systems can also help transport nutrients 
(nitrogen and phosphorus) and pesticides from nutrient-laden fields 
into outlets that in turn feed nearby water bodies.[Footnote 61] 

According to officials in NRCS' Office of Science and Technology, the 
agency has therefore put procedures in place to help ensure that all 
resources, including water quality, are protected when EQIP funds are 
used. These procedures include the following: 

* Environmental evaluation: NRCS' National Environmental Coordinator 
told us that the agency requires its field planners to perform an 
environmental evaluation for every proposed conservation practice as 
part of NRCS' process to comply with the National Environmental Policy 
Act.[Footnote 62] The environmental evaluation helps field planners 
identify existing soil and water resources on land where a 
conservation practice is proposed and to analyze the effects of the 
proposed practice on the same resources. In short, NRCS field planners 
are to assess whether the proposed actions for improving one resource 
(e.g., soil erosion) will negatively affect another resource (e.g., 
water quality) and document their assessment and determination. 

* Compliance with state plans for impaired waters: With respect to 
water quality, field planners are to identify whether a proposed 
practice is on or near a stream listed by the state as impaired. If it 
is, then field planners are directed to review and comply with any 
existing pollution limits or watershed plans that have been 
established by the state. This process asks field planners to ensure 
that landowners are provided with options to install practices that 
will not further degrade the stream segment. The final decision on 
what practices to install ultimately rests with the landowner. 

* Nutrient management planning: In the event of a significant net 
negative effect on water quality, field planners are to document it, 
which would trigger an environmental assessment or an environmental 
impact statement under the National Environmental Policy Act, 
according to NRCS' National Environmental Coordinator. He said that 
significant net negative effects on water quality are rare because 
landowners instead typically agree to implement other, mitigating 
practices along with the proposed practice. The agency's key method to 
help ensure that the potential negative effect is mitigated is termed 
nutrient management planning. Nutrient management plans describe a 
coordinated combination of conservation practices that help farmers 
manage the amount, form, placement, and timing of fertilizer to 
support crop production while also minimizing polluted runoff. 
[Footnote 63] These plans are site specific and are developed by NRCS 
field planners and implemented by landowners. 

* Alternative mitigation measures: According to NRCS Science and 
Technology officials, if landowners choose not to adopt nutrient 
management plans, NRCS field staff may still work with them to develop 
alternative mitigation measures--that is, additional conservation 
practices, often in a suite, to minimize potential adverse effects on 
water quality. Such mitigation measures may include, for example, 
planting cover crops on agricultural fields or installing filter 
strips on field borders. Each of these practices helps reduce runoff 
of nutrients and pesticides from fields into nearby water bodies by 
absorbing pollutants in plant root systems. 

Instances of Potential Water Impacts Raise Questions about the Extent 
of Adherence to Procedures, but NRCS Data Are Too Highly Aggregated to 
Resolve Them: 

During the course of our field work, we identified instances that 
raised questions as to whether NRCS' procedures to protect water 
quality were always implemented as intended, particularly in 
watersheds where EPA's section 319 funds had been used. To ascertain 
whether such instances were isolated occurrences or indicative of a 
broader problem, we examined NRCS' national level data to determine 
the extent to which NRCS' procedures intending to ensure that all 
resources are protected (including water quality) were followed. 

Observed Instances of Potential Water Impacts from Conservation 
Practices: 

During the course of our field work, we found some indications that 
site-specific mitigating practices were not implemented when practices 
with potentially negative effects on water quality were installed. For 
example, nonpoint source management program staff in two states told 
us that NRCS field staff have sometimes funded soil conservation 
practices without appropriate alternative mitigation measures. 
Specifically: 

* An agriculture specialist in Washington State's nonpoint source 
pollution program told us that NRCS field staff have authorized 
funding of stream crossings for livestock. The departmental official 
told us that stream crossings in Washington were installed on certain 
properties where livestock exclusion fencing prevented cattle from 
entering the stream. The stream crossings were to be installed for 
emergency purposes only, in case off-site watering facilities were 
unable to function, and closed off with gates. He explained, however, 
that some landowners chose not to implement additional conservation 
practices, such as off-site watering facilities in places that would 
keep the cattle away from the stream, and they have kept the gates 
permanently open, meaning that livestock now enter certain streams 
more often than they did before installation of the stream 
crossings.[Footnote 64] A former agriculture specialist with the 
program echoed this concern, stating that on land where livestock 
graze in eastern Washington, it is not uncommon for NRCS conservation 
practices intended to protect water quality to be implemented in such 
a way that actually increases the number of livestock entering streams. 

* An engineer with Ohio's Environmental Protection Agency official 
told us that NRCS promoted hayland buffer strips to help reduce soil 
erosion and absorb nutrient runoff as part of a NRCS water quality 
focused program. The buffer strips were to be 100 to 200 feet wide, 
located adjacent to a water body, and to remain in place for 3 years. 
Landowners received $100 per acre each year for installing these 
buffer strips. But on some properties, riparian vegetation--and the 
water quality protection it provides--was removed so that landowners 
could install hayland buffers. According to an engineer with Ohio's 
Environmental Protection Agency, hayland buffers provide less overall 
protection to water quality than wooded riparian vegetation, such as 
that which was removed. 

Headquarters Data on Projects' Potential Impacts Are Too Highly 
Aggregated to Reveal Site-Specific Information: 

To determine whether the instances we observed when meeting with state 
officials were isolated examples or indicative of a more prevalent 
problem, we examined EQIP data on three key parameters: (1) the 
universe of conservation practices funded by NRCS field units that 
could have negative water quality effects in watersheds in which 
Section 319 projects were funded; (2) for that universe of practices, 
the extent to which nutrient management plans were in place to 
mitigate unintended adverse effects on water quality; and (3) where 
nutrient management plans were not in place, the extent to which 
alternative mitigation practices were in place that could reliably 
serve that same purpose. 

Conservation practices that can affect water quality. EQIP data show 
that from 2005 through 2010, of the 47,000 practices that NRCS field 
units funded in watersheds where states allocated section 319 funds, 
nearly 8,000 were types of individual conservation practices that 
could facilitate agricultural runoff or have other unintended 
consequences unless other mitigating measures were implemented along 
with them.[Footnote 65] These 8,000 practices were funded in about 820 
watersheds. 

Use of nutrient management plans. EQIP data show that nutrient 
management plans--the agency's primary method for ensuring that those 
practices intending to conserve one resource (e.g., soil) do not 
inadvertently harm another resource (e.g., water)--were funded on 
properties in less than one-third of the watersheds where soil 
conservation practices that NRCS acknowledges could degrade water 
quality were also funded. According to NRCS officials, there are 
several reasons why the data on nutrient management plans do not 
provide a complete picture of water protection efforts. For example, 
it is possible that some of the acres within the 820 watersheds where 
these practices were funded were actually not vulnerable--that is, 
were not close to water bodies of concern--and therefore did not 
require a nutrient management plan to ensure that water quality was 
protected. In addition, officials told us that nutrient management 
plans expire after 3 years, so nutrient management plan contracts that 
expired during the time period of our analysis that could have 
possibly addressed the practices with potential adverse effects on 
water quality would not appear in the data. Furthermore, landowners 
often adopt nutrient management practices well beyond the length of a 
contract because they recognize the economical and environmental 
benefits of improved nutrient use efficiency provided by those 
practices, according to NRCS officials.[Footnote 66] 

Use of alternative mitigation measures. As noted above, the absence of 
a nutrient management plan does not mean that an NRCS conservation 
practice will adversely affect water quality, if proper alternative 
mitigation measures are in place. According to NRCS officials, when a 
conservation practice with a potentially negative effect is 
identified, and where a nutrient management plan is not in place, 
conservation planners are alerted so that they know to plan any site-
specific mitigating practices to ensure positive outcomes. However, 
when we examined EQIP data provided by USDA to determine if 
alternative mitigation measures were funded in the two-thirds of 
watersheds where nutrient management plans had not been funded, we 
found the data to be too highly aggregated to allow for a 
determination as to whether the conservation practices reflected in 
the data were the appropriate practices that could mitigate site-
specific problem. For example, the EQIP data show that the agency 
funds an average of 4 conservation practices per EQIP contract. This 
summary information, however, does not shed light on the type of 
practices that are being installed, whether the combination of those 
practices have a water quality focus, or whether they are effective in 
mitigating the potentially adverse effects on water quality of the 
practices in question. 

According to NRCS officials, detailed, project-specific information, 
while not available at the national level, is available in NRCS' many 
field offices across the country. NRCS field staff, for instance, are 
supposed to document their site-specific determinations, which include 
information on, among other things, how conservation practices are to 
be implemented in a way that protects all resources. Moreover, the 
field offices are subject to NRCS' internal quality assurance 
processes designed to ensure that all contracts are structured to 
protect all resources, including water quality, and that projects are 
appropriately tailored to reflect site-specific conditions.[Footnote 
67] Nonetheless, EQIP officials told us they neither ask for nor 
analyze this site-specific information. Without examining such data, 
however, it is difficult to see how NRCS can assure itself or the 
Congress that certain practices are not having unintended effects on 
water quality. 

Conclusions: 

The magnitude, pervasiveness, and dispersed nature of nonpoint source 
pollution make it particularly difficult for states to control. EPA 
has achieved some notable successes through its section 319 program 
and, in recent years, has helped states target their nonpoint source 
pollution reduction efforts in watersheds with the most severe water 
quality problems. Now more than ever, EPA's and states' limited 
budgets make it critical that the most effective projects are selected 
for funding. In some cases, however, states used section 319 to fund 
projects that were not effectively implemented or not clearly linked 
to tangible water quality results. Our review of the experiences of 
some states and EPA's 2011 evaluation report has shown that such 
issues can potentially be avoided when states use more rigorous 
project selection processes. EPA's regional offices can constructively 
influence the types of projects that states fund through the program, 
but they are generally not reviewing states' plans for project 
feasibility or for project selection criteria that would help ensure 
that funded projects have characteristics reflecting the greatest 
likelihood of effective implementation and tangible water quality 
results. As state programs have evolved over the last decade, some 
state programs have shown that certain characteristics of proposed 
projects, such as securing third-party buy-in in advance, can provide 
greater assurance that these projects will achieve tangible water 
quality results, although such lessons learned have not been 
systematically adopted by all states. 

In addition, EPA's emphasis on two statutorily required reporting 
measures as measures of effectiveness--to the exclusion of other 
measures--may not be fully capturing information reflecting program 
achievements and may, in some cases, influence state project selection 
toward narrow measures of nonpoint source pollution over comprehensive 
results. As a result of EPA's focus on these primary measures, states 
are sometimes selecting projects targeted to meet those measures, 
rather than selecting projects that could have larger impacts on 
improving the health of impaired or threatened water bodies. At 
present, certain EPA documents discuss the advantages of additional 
performance measures that may more accurately reflect the overall 
health of water bodies, such as conditions for aquatic life, but the 
agency does not require states to use such measures to provide 
information on their progress under the nonpoint source program. As a 
result, most states report on reductions of specific pollutants, 
rather than on indicators of overall health of targeted water bodies 
(e.g., the number, kind, and condition of living organisms) or on 
protection of high-quality water bodies that are not impaired. EPA 
plans to soon issue new section 319 guidelines to states that 
generally address program accountability, but it is unclear whether 
and to what extent these new guidelines will include measures to more 
accurately reflect the overall health of targeted water bodies or 
demonstrate states' focus on protecting high-quality water bodies, 
where appropriate. 

USDA's Environmental Quality Incentives Program has helped to 
substantially reduce sediment, nitrogen, and phosphorus runoff across 
the country. By their nature, however, some of the conservation 
practices supported by the program have the potential to inadvertently 
conflict with EPA efforts to reduce nonpoint source water pollution. 
While NRCS procedures strive to minimize such problems, state 
environmental officials identified instances where these procedures 
may not always have their intended effect. NRCS has cited highly-
aggregated data to demonstrate that a preponderance of mitigating 
practices--an average of four practices per project--counter any 
possibility that such unintended effects occur. However, the EQIP data 
provided by USDA lack the details needed to assess whether these other 
practices mitigate the potential for negative effects. The most 
meaningful data on the use and effectiveness of mitigating practices 
is site-specific information that resides within NRCS' field offices, 
and has been neither obtained nor analyzed by NRCS program officials. 
Tapping and analyzing these data could more accurately inform NRCS, 
and other interested parties including the Congress, on the extent to 
which EQIP projects may inadvertently affect water quality in areas 
where Section 319 funds are used. 

Recommendations for Executive Action: 

We are making three recommendations to help protect the quality of our 
nation's water resources. 

To strengthen EPA's implementation of its responsibilities under the 
Clean Water Act's section 319 nonpoint source pollution control 
program, we recommend that the Administrator of EPA take the following 
two actions: 

* provide specific guidance to EPA's 10 regional offices on how they 
are to fulfill their oversight responsibilities, such as how to review 
states' plans for project feasibility and criteria to ensure that 
funded projects have characteristics that reflect the greatest 
likelihood of effective implementation and tangible water quality 
results, and: 

* in revising section 319 guidelines to states, and in addition to 
existing statutorily required reporting measures, emphasize measures 
that (1) more accurately reflect the overall health of targeted water 
bodies (e.g., the number, kind, and condition of living organisms) and 
(2) demonstrate states' focus on protecting high-quality water bodies, 
where appropriate. 

To provide assurance that efforts to conserve soil resources do not 
work at cross-purposes with efforts to protect water quality, we 
recommend that the Secretary of Agriculture direct the Chief of the 
Natural Resources Conservation Service to analyze available 
information, and obtain necessary information from field offices, to 
determine the extent to which appropriate mitigation measures are 
implemented when nutrient management plans are not in use, 
particularly in watersheds where states are spending section 319 funds. 

Agency Comments and Our Evaluation: 

We provided a draft of this report to the Administrator of the 
Environmental Protection Agency and to the Secretary of Agriculture 
for their review and comment. EPA provided written comments in an 
April 16, 2012, letter, in which the agency expressed general 
agreement with the report's two recommendations calling for improved 
and more consistent regional oversight, and for improved and more 
comprehensive program measures. The letter also cited GAO's 
"constructive engagement with the EPA headquarters, the EPA regions, 
and state nonpoint source control program staff." It did, however, 
also question our characterization of several points related to 
project selection and effectiveness. The letter is included in 
appendix II along with our responses to the agency's comments. 

USDA's NRCS provided written comments in an April 20, 2012, letter 
that did not specify whether the agency concurred with our 
recommendations. The letter acknowledged that we addressed some of the 
concerns NRCS raised in reviewing an earlier "Statement of Facts" we 
had provided NRCS officials as a means of verifying the factual 
information we had planned to use in drafting our report. However, in 
its letter, NRCS took issue with what it characterized as "several 
inaccuracies" that remained after the draft report was sent to USDA 
for comment. Specifically, NRCS identified two, related concerns with 
the draft report, stating that (1) the message conveys that USDA soil 
conservation practices have unintended negative impacts on water 
quality and (2) this inaccuracy appears to be based on 
misinterpretation and subsequent misuse of a generalized planning tool 
(Conservation Practice Physical Effects matrix), lack of knowledge of 
NRCS conservation planning process, and inferences that exceed the 
limitations of the data on which they are based. Regarding the first 
concern, our draft report acknowledged the goals and accomplishments 
of NRCS' Environmental Quality Incentives Program in mitigating the 
impacts on water quality of certain agricultural practices. We revised 
the language in the draft report to further discuss the program's 
benefits in response to NRCS comments. That said, our field work 
identified instances where the program's goal of mitigating 
agricultural impacts appeared--on occasion--not to have been carried 
out at the ground level, and where water quality may have been 
affected as a result. 

Addressing NRCS' second concern, in an effort to ascertain whether 
instances of ground water quality being affected were anecdotal or 
more prevalent, we examined USDA data and other information. 
Specifically, we examined (1) information from NRCS' Conservation 
Practice Physical Effects matrix; (2) data on EQIP conservation 
practices funded in watersheds where states had spent section 319 
funds; and (3) information on the extent to which "alternative 
mitigation measures" are used as required, when nutrient management 
plans are not in use. In examining this information, we concluded that 
the EQIP data provided by USDA do not contain site-specific 
information on the extent to which alternative mitigation measures are 
effectively employed when nutrient management plans are not used. As 
we state in the report, without examining such information, neither 
we, nor NRCS, could determine that certain practices are not having 
unintended effects on water quality. 

It was for this reason and to ensure that complete data are available 
to allow NRCS and others to assess whether the program has unintended 
water quality impacts that we recommended that NRCS analyze available 
information and obtain necessary site-specific information from field 
offices as necessary. NRCS would then be in a better position to 
determine the extent to which appropriate mitigation measures are 
implemented when nutrient management plans are not in use (and 
particularly in watersheds where states are spending section 319 
funds). We therefore continue to believe this recommendation has 
merit. NRCS' letter is included in appendix III, along with our 
responses to its comments. 

As agreed with your offices, unless you publicly announce the contents 
of this report earlier, we plan no further distribution until 30 days 
from the report date. At that time, we will send copies to the 
Administrator of EPA, the Secretary of Agriculture, the appropriate 
congressional committees, and other interested parties. In addition, 
this report will be available at no charge on the GAO website at 
[hyperlink, http://www.gao.gov]. 

If you or your staff members have any questions about this report, 
please contact me at (202) 512-3841 or trimbled@gao.gov. Contact 
points for our Offices of Congressional Relations and Public Affairs 
may be found on the last page of this report. GAO staff who made major 
contributions to this report are listed in appendix IV. 

Signed by: 

David C. Trimble: 
Director, Natural Resources and Environment: 

[End of section] 

Appendix I: Objectives, Scope, and Methodology: 

The objectives of our work were to examine (1) states' experiences in 
funding projects that effectively address nonpoint source pollution 
problems, (2) the extent to which the Environmental Protection Agency 
(EPA) oversees the section 319 program and measures program 
effectiveness in reducing the adverse impacts of nonpoint source 
pollution on water quality, and (3) the extent to which key 
agricultural conservation programs complement EPA's efforts to reduce 
nonpoint source pollution. 

To conduct this work, we reviewed relevant laws, regulations, and 
agency guidance. In addition, we visited 8 states in four EPA regions: 
Arkansas, Louisiana, Maryland, Michigan, Ohio, Pennsylvania, 
Washington, and West Virginia. We chose these states on the basis of 
their varied types of nonpoint source pollution and proximity to some 
of the nation's premier watersheds, such as the Chesapeake Bay, Great 
Lakes, and Mississippi River. On our site visits, we met with state 
nonpoint source management program officials, conservation districts, 
nonprofit organizations, watershed associations, and research 
officials. 

We also interviewed agency officials, including officials from EPA, 
the U.S. Department of Agriculture (USDA), EPA regional offices, the 
Natural Resources Conservation Service's Office of Science and 
Technology, and the Environmental Quality Incentives Program. We also 
interviewed representatives from the National Association of 
Conservation Districts, Association of State and Interstate Water 
Pollution Control Administrators, New England Interstate Water 
Pollution Control Commission, and state abandoned mine land programs. 
We discussed with these officials their observations on efforts to 
reduce nonpoint source pollution and the challenges associated with 
such efforts. We also interviewed subject-matter experts from industry 
and academia. 

To examine states' experiences in selecting projects that effectively 
address nonpoint source pollution, we e-mailed a 10-question data 
collection instrument to nonpoint source program management officials 
in all 50 states, which solicited information from them on project 
selection processes, project selection criteria, types of 
organizations receiving funding, program organization and 
responsibilities, and program oversight practices. We received 
responses to this data collection instrument from every state. We also 
examined summary information, including objectives, methods, and 
outcomes, for more than 1,500 projects in EPA's Grants Reporting and 
Tracking System database. We reviewed this information for all 
projects in the database that were (1) awarded funding during or after 
fiscal year 2004 and completed before December 31, 2011; (2) 
categorized as nonstatewide projects; (3) received section 319 funds; 
and (4) had complete information on objectives, methods, and outcomes. 
Before reviewing project summary information and drawing the sample 
for our survey, we interviewed EPA officials to discuss the 
reliability of the data contained in EPA's Grants Reporting and 
Tracking System; we also checked for outliers and determined that the 
data were sufficiently reliable for our purposes. 

In addition, we surveyed a random sample of 524 managers of projects 
that have been implemented with section 319 funds. The purpose of this 
survey was to examine topics such as general project information, 
project proposal and selection, conservation practice selection, 
project implementation and goals, challenges associated with projects, 
monitoring and oversight, and funding sources. To identify issues 
pertaining to section 319-funded projects and to develop the survey 
questions, we reviewed state annual reports and interviewed 
headquarters and regional agency officials and subject-matter experts. 

We selected an initial simple random sample from EPA's Grants 
Reporting and Tracking System out of a universe of 1,584 projects. We 
selected projects that received section 319 funding; were completed 
between January 1, 2004, and June 9, 2011; and involved implementing 
conservation practices or remediation techniques, rather than, for 
example, projects that focused primarily on planning or monitoring. 
After drawing the initial sample, we removed duplicates and excluded 
projects that we learned were statewide, because such projects did not 
always involve implementing conservation practices or remediation 
techniques. We also excluded projects from North Dakota and South 
Dakota because some incomplete projects and some that never started in 
these states were listed as complete in EPA's Grants Reporting and 
Tracking System. After obtaining contact information for the sampled 
projects, we also excluded those projects for which knowledgeable 
officials were no longer available. After making these adjustments, we 
estimated that the number of projects meeting our criteria was 1,273. 
For those project managers who had more than 2 projects sampled, we 
randomly selected 2 projects for the survey in order to reduce 
respondent burden. After these adjustments, the final number of 
projects in the sample was 524. The results of our survey are 
generalizable to the population of section 319 projects that meet our 
criteria. That is, they are not generalizable to projects, for 
example, that did not implement conservation practices or those that 
were completed before January 1, 2004. 

The survey was conducted using self-administered electronic 
questionnaires posted on the World Wide Web. While developing the 
survey questions, we conducted two rounds of pretests with section 319 
project managers over the phone. We conducted six first-round 
exploratory pretests with managers to help develop the scope of the 
questionnaire and key concepts. After refining our concepts and 
questions, we pretested a draft version of the questionnaire with five 
project managers. We conducted pretests to check that (1) the 
questions were clear and unambiguous, (2) terminology was used 
correctly, (3) the questionnaire did not place an undue burden on 
agency officials, (4) the information could feasibly be obtained, and 
(5) the survey was comprehensive and unbiased. We made changes to the 
content or format of the questionnaire after each pretest according to 
the feedback we received. A draft of the questionnaire was also 
reviewed by independent GAO survey experts, and we revised the 
questionnaire to reflect that review. 

We contacted survey respondents by sending the survey through an e-
mail notification to each. We e-mailed each potential respondent a 
unique password and username to ensure that only members of the target 
population could participate in the survey. The survey data were 
collected from September 2011 through November 2011. We sent follow-up 
e-mail messages to those who had not responded by the deadline to our 
original e-mail. We then telephoned all remaining nonrespondents for 
whom contact information was available, beginning in October 2011. We 
received a total of 298 responses, accounting for an overall 
unweighted response rate of 57 percent. Estimates produced from the 
sample of projects are subject to sampling error. We express our 
confidence in the precision of our results as a 95 percent confidence 
interval. This interval would contain the actual population value for 
95 percent of the samples we could have drawn. As a result, we are 95 
percent confident that each of the confidence intervals in this report 
includes the true values in the study population. Additionally, to 
encourage honest and open responses, we pledged in the introduction to 
the survey that we would report information in the aggregate and not 
report data that would identify a particular respondent. This report 
does not contain all the results from the survey; the survey and a 
more complete tabulation of the results are provided in a supplement 
to this report (see GAO-12-377SP). To eliminate data-processing 
errors, we independently verified the computer program that generated 
the survey results. 

In addition to tabulating and analyzing the frequencies of survey 
responses, we conducted a content analysis of all of the open-ended 
narrative responses received to survey questions 33, 34, and 35. We 
analyzed the content of the 180 responses to question 33 and the 106 
responses to 34 to identify the types of challenges faced and reasons 
for grant revisions and extensions. Question 33 was coded using the 
following categories: property access, lack of identified project 
sites, staff, weather, budget, technical, the National Environmental 
Policy Act, administrative barriers, coordination, other, exempt, and 
unclear. Projects coded as exempt included those for which no 
challenge was encountered or the project was found to be outside of 
the scope of the sample. Projects were coded as unclear if it was 
unclear whether a challenge was encountered or if the nature of the 
challenge was unclear. Question 34 was coded using the following 
categories: time, budget, location, project scope or specifications, 
other, exempt, and unclear. Projects coded as exempt included those 
for which no grant extension was required or the project was found to 
be outside of the scope of the sample. Projects were coded as unclear 
if the length of the extension was unclear. Question 34 was coded to 
reflect the number of months of the grant extension; unclear and 
exempt categories were also used. We also analyzed the 129 responses 
to question 35, which was coded to reflect the number of months of the 
grant extension, including use of unclear and exempt categories. 
Coding was performed independently by two coders; team members then 
met to discuss the coding categories and reached consensus on the 
final coding category assignment for each response. Measures of 
interrater reliability were calculated before codes were reconciled 
and found to be sufficiently high for the purposes of this analysis. 
The numbers of responses in each content category were then summarized 
and tallied. 

To examine the extent to which EPA oversees the section 319 program 
and measures program effectiveness in reducing the adverse impacts of 
nonpoint source pollution on water quality, we obtained from EPA's 10 
regional offices information on the nature and extent of their 
oversight of state programs, including the extent to which they 
examine states' project selection processes, annual plans, and program 
objectives and the criteria they use to annually award funds. In 
addition, we examined section 319's statutorily required reporting 
requirements, which EPA uses as national measures of program 
effectiveness. We evaluated the water quality benefits derived from 
projects that address these measures, compared with the water quality 
benefits of projects that address other EPA-approved measures of state 
program effectiveness, primarily by reviewing EPA documents and 
interviewing state nonpoint source program officials. We also obtained 
annual reports from 42 states' nonpoint source management programs and 
reviewed 25 of them to determine how they reported the achievements of 
section 319-funded projects during the most recent fiscal year for 
which the report was available. 

To examine the extent to which key agricultural conservation programs 
complement EPA's efforts to reduce nonpoint source pollution, we 
analyzed data on USDA's conservation practices funded under the 
Environmental Quality Incentives Program that have been implemented in 
watersheds where states have allocated section 319 funds. We obtained 
these data from the Program Contracts System, known as ProTracts, 
which is used to manage Natural Resources Conservation Service 
conservation program applications, cost-share contracts, and program 
funds. We also examined USDA reports on the effectiveness of 
conservation practices, including those produced by the Conservation 
Effects Assessment Project. To assess the reliability of this 
database, we performed electronic data testing for missing data, 
outliers, and obvious errors. In addition, we interviewed 
knowledgeable agency officials and compared summary information from 
the database with published reports. On the basis of this assessment, 
we determined that ProTracts data were sufficiently reliable for our 
purposes. We also interviewed USDA officials in the Natural Resources 
Conservation Service's Office of Science and Technology. 

We conducted this performance audit from December 2010 through May 
2012, in accordance with generally accepted government auditing 
standards. Those standards require that we plan and perform the audit 
to obtain sufficient, appropriate evidence to provide a reasonable 
basis for our findings and conclusions based on our audit objectives. 
We believe that the evidence obtained provides a reasonable basis for 
our findings and conclusions based on our audit objectives. 

[End of section] 

Appendix II: Comments from the Environmental Protection Agency: 

Note: GAO comments supplementing those in the report text appear at 
the end of this appendix. 

United States Environmental Protection Agency: 
Office Of Water: 
Washington, D.C. 20460: 

April 16, 2012: 

Mr. David C. Trimble: 
Director, Natural Resources and Environment: 
U.S. Government Accountability Office: 
Washington, D.C. 20548: 

Dear Mr. Trimble: 

Thank you for the opportunity to review and comment on the draft 
report Nonpoint Source Water Pollution: Greater Oversight and 
Additional Data Needed for Key EPA Water Program, which examines 
aspects of the EPA's Clean Water Act Section 319 grant program. 
Overall, the EPA believes GAO's findings highlight the fact that the 
EPA and the states have done well with the resources Congress has 
provided in achieving nonpoint source pollution reductions, and we are 
in general agreement with GAO's two recommendations to the EPA to 
improve the effectiveness of this program. Although modestly funded 
relative to the challenges of nonpoint source pollution nationwide, 
the Section 319 grant program has used partnerships and leveraging of 
additional resources to achieve significant watershed protection and 
restoration. This includes the restoration of more than 360 water 
bodies documented as "Section 319 Nonpoint Source Success Stories.”
[Footnote 1] However, the EPA suggests that the continued success of 
the nonpoint source program depends on the right balance of sufficient 
resources, legal authorities, and tree partnerships with the United 
States Department of Agriculture (USDA), the states, and other key 
stakeholders to achieve a common goal of water quality protection and 
improvement. The EPA's general comments are contained in this letter, 
with more detailed comments and editorial recommendations in the 
enclosure. 

GAO Finding Highlights Success of EPA's Clean Water Act Section 319 
Program: 

The EPA appreciates GAO's acknowledgment of the program successes in 
the Section 319 program. This is especially notable given the 
program's primary emphasis on non-regulatory approaches that rely on 
voluntary participation by local stakeholders and private landowners. 
GAO's finding that over 70 percent of watershed projects accomplished 
all objectives originally identified in the proposal indicates that 
the majority of Section 319-funded projects are carefully planned and 
well managed during implementation. We believe the characterization 
and language in the GAO report should better align with this generally 
positive finding about the Section 319 program (see detailed comments 
in the attachment). [See comment1] 

Although GAO did find that some Section 319-funded projects 
encountered difficulties, the report indicates a number of instances 
in which states learned from funding unsuccessful projects and 
adjusted the state project selection process accordingly. This type of 
adaptive management is essential in watershed-based implementation 
where local leaders, landowners, partners and resources vary in each 
new location. 

Adaptive Management Necessary for Section 319 Implementation Projects: 

GAO's discussion under "preventing challenges through more rigorous 
project selection" suggests that all projects should have necessary 
landowner "buy-in" secured in advance. While this may be feasible for 
some smaller projects when project coordinators are able to fund 
recruitment up front, the EPA does not believe it is feasible to 
require all grant applicants to have landowner participation secured 
in advance. Despite best efforts in project selection and 
outreach/recruitment of landowners, Section 319-funded projects are 
based on voluntary participation and in many cases will require 
adjustments during a project period. The EPA does rot believe GAO 
should broadly characterize projects that use adaptive management as 
facing "preventable challenges" and suggests some wording changes in 
the detailed comments. [See comment 2] 

Nonpoint Source Control Practices are Implemented in the Context of 
Watershed-Based Plans: 

The GAO report correctly noted that waters impaired by agriculture are 
best addressed through a suite of conservation practices. [See comment 
3] We are concerned by GAO's assertion that "...the EPA data indicated 
that Section 319 funds paid for conservation practices to be put in 
place in isolation" and don't believe the GAO analysis supports this 
conclusion. Reviewing projects in the Section 319 Grants Reporting and 
Tracking System (GRTS) gives a snapshot of project activity but may 
not provide a full picture of watershed restoration. GRTS provides 
data for past or ongoing Section 319 projects but will not represent 
practices that have been put in place already or are planned, or that 
are being implemented via other agencies and funding mechanisms (e.g., 
EQIP). The Section 319 guidelines require a watershed-based plan be in 
place prior to funding implementation projects for impaired waters. 
The purpose of this requirement is to ensure that water quality 
problems are analyzed at a watershed scale, critical areas are 
identified, and a suite of necessary practices are identified and 
implemented through the plan. [See comment 4] 

The Role of Section 319 Funding in Restoring and Protecting the 
Nation's Waters: 

The GAO review is described as stemming from concerns about the 
persistence of nonpoint source pollution "particularly given more than 
20 years of funding for the Section 319 program." This preface to the 
GAO report lacks critical context related to the pervasiveness, 
variety, and magnitude of NPS pollution nationwide relative to the 
federal funding levels and scope of the Section 319 program. NPS 
pollution encompasses a wide range of sources that are not subject to 
federal or even state regulation. The scope of the problem expands as 
a result of population growth and land use changes; even as waters are 
restored, others are listed as impaired as a result of development 
pressures and other factors such as recent identification of existing 
water quality problems. The vast extent and continuous nature of NPS 
pollution requires that the problem be addressed through a variety of 
federal, state and local non-regulatory and regulatory approaches, as 
well as through multiple funding sources. [See comment 5] 

As one point of reference, the Clean Watersheds Needs Survey 2008 
Report to Congress compiles state estimates of costs to address NPS 
pollution and decentralized/onsite wastewater systems at $46.7 billion 
over a 20-year period. This is likely an underestimate as not all 
states provided information for the NPS/onsite systems portion of the 
survey. The reported funding needs are equivalent to an annual need of 
$2.3 billion per year. 

Section 319 funding is most appropriately viewed as a critical source 
of support for state NPS management programs and for watershed 
projects that demonstrate successful NPS pollution controls.
While Section 319 funding is an essential part of the solution, it is 
not the entire remedy. Many states successfully leverage additional 
state, federal and local funds in support of their NPS programs, and the
EPA continues to work with federal partners to leverage the greater 
resources available through other programs. In this regard, we are 
pleased about a joint FY13 initiative with USDA's Natural Resources
Conservation Service in which we will collaborate in watersheds across 
the country to address waters impaired by excess nutrients and 
sediments. Given that GAO's study looked at aspects of USDA programs, 
we welcome any insights GAO may have for additional opportunities for 
partnering and leveraging with USDA. 

EPA's Response to GAO's Recommendations: 

The EPA completed the National Evaluation of the Clean Water Act 
Section 319 Program in November 2011. Appendix C of this study or 
tunes a number of potential Section 319 program enhancements. We are 
moving forward this year with several of these program revisions aimed 
at strengthening the strategic focus of state NPS programs, providing 
more consistent review of state programs nationally, and improving our 
ability to document the progress and success of the Section 319 
program. We will be revising our Nonpoint Source Program and Grants 
Guidelines for States and Territories in November 2012 for use in 
Fiscal Year 2013 end beyond. These program efforts align well with the 
two GAO recommendations for the EPA. 

Additionally, as part of an Agency Water Quality Priority Goal for 
Fiscal Year 2012-2013, the EPA has committed that 50 percent of states 
will revise their NPS management programs plans by September 30, 2013. 
Within the next few months we will be providing guidance to states and 
the EPA regions on updating NPS program plans. The EPA will also 
provide guidance to the EPA regions en conducting annual satisfactory 
progress determinations of states' NPS program progress, increasing 
national consistency in the conduct of these reviews. The EPA's 
response to the recommendations and additional program refinements are 
described below. 

(1) Provide Specific Guidance to EPA Regional Offices on Oversight: 

GAO's first recommendation for the EPA is to provide guidance to the 
EPA regional offices on how they are to fulfill their oversight 
responsibilities. GAO's report mentioned the variable rate of updates 
to state NPS management program plans, and recommends reviewing state 
workplans for project feasibility and criteria to ensure that funded 
projects have characteristics that reflect the greatest likelihood of 
effective implementation and tangible water quality results. 

The EPA agrees with this recommendation. As noted above we will 
provide guidance on the EPA regional review of updated State NPS 
management program plans, as well as for the EPA regional satisfactory 
progress determinations of state NPS programs. In terms of project 
selection, we would like to reiterate the finding in the GAO report 
that all regions review annual state grant workplans. Further, some 
EPA regions play a greater role in project selection by, for example, 
participating in state project selection committees. We believe, and 
GAO's analysis seems to support, that successful projects have been 
associated with differing levels of EPA regional involvement unique to 
state and regional circumstances. We will engage the states and the 
EPA regions to identify best practices for project selection in the 
Section 319 program, and we will incorporate these best practices into 
our program guidance as appropriate. 

Review Section 319 Program Measures: 

GAO's second recommendation to the EPA is to emphasize measures that 
(1) more accurately reflect the overall health of targeted water 
bodies, and (2) demonstrate states' focus on protecting high quality 
water bodies, where appropriate. The EPA is in agreement with the idea 
that more or different measures would better represent the progress 
and accomplishments of the Section 319 program, especially given the 
inevitably long lag time from project initiation to positive water 
quality results. 

GAO cites some limitations with the existing program measure to 
quantify pollutant load reductions (nitrogen, phosphorus, sediments). 
The EPA recognizes that reductions in pollutant loadings do not always 
capture the full water quality benefits of various projects. 
Nonetheless we believe reductions in pollutant loadings are an 
effective way to realize water quality results and are an important 
measure of success for many projects. Pollutant reductions are often a 
necessary component, even if other actions are needed to achieve full 
restoration. This measure provides valuable information in the 
implementation year to give some indication of recent progress, and is 
called for by statute. We have a strong interest in making this 
measure more meaningful, for example by linking annual pollutant 
reductions to the total reductions needed to reach water quality 
goals. Later this year, the EPA will undertake a process to develop 
program measures that will gauge incremental progress in addressing 
water quality in NPS-impaired waters. 

The EPA appreciates GAO's interest in the need to focus on the overall 
health of water bodies in the context of measuring the number, kind 
and condition of living organisms. The EPA encourages states to 
consider biological assessments and the adoption of biocriteria as 
part of their water quality standards program. Where states do gauge 
their water quality progress against biocriteria, this is usually 
noted in the descriptive summaries of the EPA's measure WQ-10 (i.e., 
Section 319 Success Stories). Moving forward we expect to continue the 
practice of gauging the success of Section 319 restoration projects 
based on state-established water quality standards, including 
biocriteria where these standards exist and states monitor for them. 

We appreciate GAO's emphasis on the importance of protecting 
unimpaired/healthy waters and expanding the role of the Section 319 
program to do so accordingly. The current (2003) Section 319 grant 
guidelines are focused on restoring impaired waters. While we expect 
that restoration of impaired waters will continue to be a key feature 
of the Section 319 program, we are actively considering ways to 
provide greater emphasis on protecting high quality waters and will 
address this issue when we revise the grant guidelines later this year. 

Measuring the benefits of protection (e.g., pollution prevention) 
presents a more difficult set of challenges than measuring restoration 
of already-impaired waters. While we cannot commit to adopting such a 
program measure in the near future, we are very interested in 
addressing how to meaningfully measure the prevention of impairment 
with data and methods that are accessible to most states. We are 
hopeful that we could move in this direction and adopt such a measure 
in the coming years. 

Again, thank you for the opportunity to comment. The EPA expresses its 
appreciation for the thoughtful work of GAO staff members during this 
review, and their constructive engagement with the EPA headquarters, 
the EPA regions, and state nonpoint source control program staff. 

Sincerely, 

Signed by: 

Nancy K. Stoner: 
Acting Assistant Administrator: 

Enclosure: 

Footnote: 

[1] [hyperlink, http://water.epa.govi/polwaste/nps/success319/.] 

The following are GAO's comments on the Environmental Protection 
Agency's letter dated April 16, 2012. 

GAO Comments: 

1. EPA stated that our report should better align with the generally 
positive finding that over 70 percent of watershed projects 
accomplished all of their originally identified objectives. As EPA 
suggested, we made several changes to our report to put the 70 percent 
figure in context. In response to the EPA comment are two additional 
considerations. First, the fact that 72 percent of projects eventually 
achieved their goals does not suggest that they all did so in either a 
timely fashion or without significant complications. For example, as 
we state in our report, almost half of the projects that achieved 
their goals did so after encountering significant challenges that 
prevented them from finishing on schedule, staying on budget, or 
achieving the desired levels of pollution reduction. Second, we did 
not intend to arbitrarily identify a specific pass/fail threshold for 
the success of surveyed projects or intend to suggest that 70 percent 
is an acceptable or unacceptable share of projects that achieved 
originally proposed goals. Instead, for those projects that did not 
achieve their originally proposed goals and for those that did so 
while encountering challenges, we described the challenges that 
projects most often faced--and their main causes--to provide 
information that may assist EPA and the states in developing or 
modifying project selection criteria or, through other means, better 
ensure that projects receiving section 319 funds in the future will 
have a high likelihood of achieving pollution reduction and other 
project goals. 

2. EPA stated that we should not broadly characterize projects that 
use adaptive management (i.e., projects in which some issues cannot be 
identified in advance but rather need to be addressed after a project 
is under way) as facing "preventable challenges." EPA noted in 
particular that it is not feasible for grant applicants to obtain 
landowner buy-in in advance of project selection. We acknowledge the 
difficulty that EPA and the states face in trying to reduce nonpoint 
source water pollution and protect threatened waters using primarily 
voluntary methods and that numerous unforeseen factors can affect the 
success of project implementation. Our survey results showed, however, 
that the chief reason why some projects did not achieve their 
originally proposed goals was because third-party buy-in was not 
secured in advance. Moreover, a key component of EPA's strategy under 
the section 319 program is to have states lay the groundwork in 
advance, using "indirect" projects (e.g., education and outreach 
activities) to obtain local support so that direct implementation 
projects can succeed. The experiences of some states, such as Ohio, 
have shown that where rigorous project selection criteria have been 
put in place--such as requirements to secure landowner participation 
in advance--the quality of projects has increased over time (as 
measured by the ability of project applicants to actually implement 
their projects as intended), and local partners have been more 
effective in meeting such requirements. 

3. EPA stated it was concerned by our statements about EPA data 
indicating that section 319 funds paid for conservation practices "to 
be put in place in isolation" and that it did not believe the analysis 
supported this conclusion. The draft report had acknowledged that 
EPA's Grants Reporting and Tracking System (GRTS) data do not provide 
a complete picture of multi-agency efforts to implement conservation 
practices. In response to EPA's comment, we revised the report to 
further acknowledge that, for the properties reflected in our analysis 
of GRTS data, complementary practices may have been installed by NRCS, 
landowners, or others. Notwithstanding the incompleteness or the GRTS 
data, however, our field work showed that section 319-funded projects 
to reduce agricultural runoff were ineffective because the proper 
suites of companion practices were not always installed. We noted that 
environmental protection officials from several of the states we 
visited told us that they had often encountered this problem with 
section 319-funded projects to reduce agricultural runoff. 

4. EPA cites a requirement in its guidelines that a watershed-based 
plan be in place before funding implementation projects for impaired 
waters, as adding further assurance that in selecting projects, "water 
quality problems are analyzed at a watershed scale, critical areas are 
identified, and a suite of necessary practices are identified and 
implemented through the plan." We acknowledge the guidelines' 
requirement for a watershed-based plan and its value. Nonetheless, 
despite the good intentions behind such watershed-based plans, EPA 
does not have regulatory authority to compel either the implementation 
of the plans or the suites of practices described within them, 
particularly on agricultural land. That choice is ultimately left to 
the landowners within the plans' geographic areas. Thus, while having 
a watershed-based plan could help promote positive outcomes, the plan 
in and of itself is no guarantee that its requirements will be 
fulfilled. 

5. EPA stated that our allusion to the persistence of nonpoint source 
pollution, "particularly given more than 20 years of funding for the 
Section 319 program," does not account for the pervasiveness, variety, 
and magnitude of nonpoint source pollution nationwide relative to the 
federal funding levels and scope o the section 319 program. We 
acknowledge the massive scope of the nation's nonpoint source water 
pollution problem and that its persistence and pervasiveness is not 
something that EPA alone can be expected to solve with Section 319 
funding alone. We have adjusted the text accordingly. At the same 
time, we believe that the funding levels allocated to the section 319 
program are not insignificant and that the improvements suggested in 
this report can enhance the program's contributions toward alleviating 
the nation's nonpoint source pollution problem. 

[End of section] 

Appendix III: Comments from the U.S. Department of Agriculture: 

Note: GAO comments supplementing those in the report text appear at 
the end of this appendix. 

United States Department of Agriculture
Natural Resources Conservation Service: 
Post Office Box 2890: 
Washington, D.C. 20013: 

April 20, 2012: 

Mr. David C. Tremble: 
Director, Natural Resources and Environment: 
U.S. Government Accountability Office: 
441 G Street, N.W., Room 2T23: 
Washington, D.C. 20548: 

Dear Mr. Tremble: 

Thank you for allowing the Department of Agriculture (USDA) to review 
and comment on the draft report entitled "Nonpoint Source Water 
Pollution—Greater Oversight and Additional Data Needed for Key EPA 
Water Program." 

We were pleased to see that you addressed several of our concerns in 
the "Statement of Facts for GAO's Review of EPA's Section 319 Program" 
(Job Code: 361251). However, it is clear that several inaccuracies 
still remain in the report. Therefore, we request your attention to 
their correction. 

Overall Comment: 

Throughout this report, the message is conveyed that USDA soil 
conservation practices have unintended negative impacts on water 
quality. This message disregards years of university and Federal 
research findings that document the positive benefits of soil 
conservation practices on water quality. This inaccuracy appears to be 
based on misinterpretation and subsequent misuse of a generalized 
national planning tool (the "Conservation Practice Physical Effects 
(CPPE) matrix"), lack of knowledge of the Natural Resources 
Conservation Service (NRCS) conservation planning process, and 
inferences that exceed the limitations of the data on which they are 
based. [See comment 1] 

The following comments are referenced by page and paragraph: 

* Highlights of "What GAO Found," last paragraph: "USDA data show that 
from 2005 through 2010, the program funded nearly 8,000 practices 
that —if implemented in isolation-can increase nonpoint so tree 
pollution." 

NRCS does not implement practices in isolation. In fact, NRCS analysis 
of its Environmental Quality Incentives Program (EQIP) data show an 
average of four practices funded per EQIP contract. [See comment 2] 
NRCS uses a site-specific conservation planning process that accounts 
for critical soil, site, cropping system, management, and other 
factors that impact water quality (e.g. factors such as slope 
gradient, soil properties influencing infiltration, surface residue 
impacts on runoff, and distance to water bodies). It is true that a 
single practice standard may have benefits for one resource concern 
and not another. However, NRCS implements suites of practice standards 
in an overall (comprehensive) Conservation Plan. In this way multiple 
resource concerns are addressed and multiple conservation practices 
are implemented to achieve the landowners' and environmental goals. As 
described more fully later in this review, the perception that USDA 
implements practices in isolation apparently arises from 
misinterpretation of the CPPE matrix. The CPPE matrix is a 
generalized, national tool that provides a first approximation of 
potential effects for each practice. If a potentially negative effect 
is identified, then the conservation planner is alerted to this 
possibility so that they know to plan any site-specific mitigating 
practices (if needed) to ensue positive outcomes. Interpretation of 
the effects of individual practices using the CPPE matrix, therefore, 
is not appropriate for assessing overall effects of site-specific 
conservation plans. Consequently, we request that you delete such 
statements (here and on pages 37, 38, and 40), [Now on pages 41, 42, 
and 46] as they are inaccurate within the context of how NRCS develops 
and implements site-specific conservation plans. [See comment 3] 

* Highlights of "What GAO Found", last paragraph: "...but USDA data 
showed that these plans had not been funded for use on properties in 
about two-thirds of the 800 watersheds." 

This information is misleading and inaccurate. [See comment 4] The 
report context from where this summary statement is drawn (page 40, 
paragraph 2) [Now on page 46] shows that nutrient management is the 
particular focus of the statement. Although nutrient management 
(practice 590) is a frequently used approach for enhancing water 
quality, it is not the only practice used for mitigating water quality 
issues. NRCS has identified over 80 practices with potential water 
quality improvement attributes. Implying that NRCS' procedures for 
protecting water quality have not been followed in two-thirds of the 
watersheds without an analysis of the use of the other 80 water 
quality improving practices is not only a misinterpretation of the 
statistics, but this conclusion violates a basic tenet of statistical 
analysis by inferring a conclusion that far exceeds the limitations of 
the data. 

Because NRCS does not know which watersheds GAO used in their 
analysis, NRCS examined all 30,000 watersheds and found about 80 
percent had water quality improving practices in place with a similar 
share reporting; "Nutrient Management Plans." This percentage does not 
mean that the other 20 percent were sub-standard (i.e. water quality 
may not have been identified as a resource concern). However, failure 
to consider practices other than "nutrient management" in this 
analysis clearly understates the number of water quality improving 
practices that arc in place on the landscape. In addition, EQIP 
contracts typically last for about 3 years. Producers often adopt 
nutrient management practices well beyond the length of a contract 
because they recognize the economical and environmental benefits of 
improved nutrient use efficiency provided by those practices. This 
adoption would not likely appear in an analysis of existing contracts. 
USDA requests that either the above qualifiers be added to improve 
accuracy of the report, or this section be deleted. [See comment 5] 

* Page 5, paragraph 1: "For example pesticide use in farming has 
resulted in toxic runoff." 

USDA requires that growers enrolled in USDA programs follow all 
national, State and local regulations. This requirement would include 
the need to follow pesticide labels. Pesticide labels are developed 
post extensive multiple year testing programs, and when followed, 
should not typically result in toxic runoff. Pesticide application is 
governed by the Environmental Protection Agency (EPA) and is 
controlled specifically by the active ingredient label restrictions.
NRCS takes this resource concern a step further and if soil and 
pesticide interactions indicate an elevated pesticide risk, that risk 
is addressed by conservation mitigation measures designed to lower the 
pesticide risk. Further, there is often a suite of conservation 
practices applied to agricultural settings in the form of an 
Integrated Pest Management (IPM) plan. Indeed, agrochemicals are very 
expensive and their over-application is neither a wise environmental 
nor economical decision. Most producers strive to balance the most 
efficacious pesticide with the least amount of application. Pesticides 
used in accordance with labeling are frequently applied with little 
threat to the environment. Pesticide use should be planned as part of 
an overall strategy designed to improve economic production and 
protect the environment. 

* Page 37, paragraph 1: [Now on page 41] "USDA and EPA documents have 
identified NRCS's Environmental Quality Incentives Program as the key 
agricultural conservation program that can complement EPA's efforts to 
reduce nonpoint source pollution under section 319, but certain soil 
conservation practices under the program may hinder the success of
EPA's efforts." 

NRCS disagrees that soil conservation practices may hinder the success 
of EPA's efforts. Upon speaking with GAO, it appears such conclusions 
are largely drawn from misinterpretation and subsequent misuse of the 
CPPE matrix. NRCS implements suites of practice standards in an 
overall (comprehensive) Conservation Plan. In this way multiple 
resource concerns are addressed and multiple conservation practices 
are implemented to achieve the landowners' and environmental goals. 
Interpretation of the effects of individual practices using the CPPE 
matrix is not appropriate for assessing overall effects of site-
specific conservation plans. Consequently, we respectfully request 
that you delete such statements as they are inaccurate within the 
context of how NRCS develops and implements site-specific conservation 
plans. 

* Page 38, first complete paragraph: "...some soil conservation 
practices could have unintended, negative effects on water quality if 
installed in isolation." [Now on page 42] 

NRCS analysis of EQIP data finds an average of four practices per 
contract, indicating that single practice implementation is very rare 
(please see table below). 

Table: Fiscal Year 2011 Financial Assistance Programs Contracts and 
Number of Practices in Contracts (Source: ProTracts 10/1/2011): 

EQIP: 
Total Number of Practice Occurrences: 168,500 
Number of Active or Completed Contracts: 38,352 
Average Number of Practices in Contracts: 4. 

WHIP: 
Total Number of Practice Occurrences: 17,569 
Number of Active or Completed Contracts: 3,856 
Average Number of Practices in Contracts: 5. 

AWEP: 
Total Number of Practice Occurrences: 4,866 
Number of Active or Completed Contracts: 1,299 
Average Number of Practices in Contracts: 4. 

CBWI: 
Total Number of Practice Occurrences: 11,340 
Number of Active or Completed Contracts: 1,716 
Average Number of Practices in Contracts: 7. 

AMA: 
Total Number of Practice Occurrences: 1,119 
Number of Active or Completed Contracts: 275 
Average Number of Practices in Contracts: 4. 

Total: 
Total Number of Practice Occurrences: 203,394 
Number of Active or Completed Contracts: 45,498 
Average Number of Practices in Contracts: 4. 

[End of table] 

Good conservation planning requires that the planner understands the 
site conditions and stated resource concern. Adequate planning avoids, 
minimizes, and mitigates negative consequences associated with 
conservation plan implementation. The statement about the effects of 
practices installed in isolation indicates that GAO again used the 
CPPE matrix. The CPPE matrix is simply a very rough first 
approximation of effects that is most often used as a training tool or 
part of EQIP ranking and is not the only effects assessment tool used 
during site-specific conservation planning. When water quality is a 
concern, other effects assessments include such evaluations as soil 
erosion potential, risk of nitrogen leaching, and risk of phosphorus 
loss. Because the CPPE matrix includes a single rating for all parts 
of the country and all situations, it is well recognized that its site-
level accuracy is low. Also, "facilitating practices" may be employed 
that do not directly affect a resource concern, but have a beneficial 
effect when used in a system of practices. For these and other reasons 
stated elsewhere, USDA requests that this statement be deleted. [See 
comment 8] 

* Page 38, paragraph 1: The report refers to underground outlets in 
saying that "Such systems can also increase the quantity of nutrients 
(nitrogen and phosphorus) and pesticides transported into outlets that 
in turn feed nearby water bodies". [Now on page 42] 

Again, this is a misinterpretation of the CPPE matrix, and the 
statement by GAO is an inaccurate description of the practice. [See 
comment 8] Underground outlets are used to allow water to change 
elevation without erosion. The practice is not used as part of a field 
drainage system. The practice does not increase the possibility of 
nutrients in the water, nor change the flow of water eventually 
reaching surface water systems; the practice prevents erosion while 
allowing the water to move down slope. The "negative entry" in the 
CPPE matrix is included to alert the planner that this practice allows 
water to move more quickly and thus reduces the time that natural 
processes have to reduce the nutrients or other agricultural 
chemicals. Please delete or clarify this section to reflect the above 
information. 

* Page 38, paragraph 1: The report states that "...if such (livestock) 
crossings are installed in isolation...(they) can increase the number 
of livestock entering the stream." [Now on page 45] 

This statement lacks perspective. NRCS analysis of the full EQIP data 
finds an average of four practices per contract, implying that single 
practice implementation is very rare. NRCS realizes that according to 
this GAO report, a Dept. of Ecology official at Washington State 
provided an example of landowners keeping the gates open rather than 
using them for emergency purposes only, as intended. USDA feels this 
is the exception, as miles of exclusion fencing are contracted every 
year to keep livestock out of streams, and stream crossings are 
appropriately utilized to help attain those water quality benefits. 
[See comment 10] 

As an example, NRCS fenced cattle out of 2.5 miles of a stream on a 
Maryland farm (below, left picture). Adding a stream crossing (see 
gate in left picture) allows cattle to travel from one side of the 
farm to another, but concentrates their crossing at one place, thereby 
reducing streambank erosion and nutrient losses to the stream. NRCS 
worked with the landowner to then plant buffers to stabilize the banks 
and capture runoff nutrients before they reach the stream (right 
picture). For balance, please add these pictures and associated 
descriptions to the report. 

[Refer to PDF for image: 2 photographs] 

* Page 39, paragraph 2: "...that help farmers manage the amount, form 
and timing of fertilizer". [Now on page 44] 

Please also include the "placement" factor in the 4Rs concept of 
nutrient management (i.e. "the amount, form, placement, and timing of 
fertilizer"). 

* Page 40, paragraph 1: The report states that "...17 percent, were 
types of individual conservation practices that could degrade water 
quality, according to NRCS's analysis of its conservation practices' 
physical effects." [Now on page 46] 

This statement is misleading and conveys an inaccurate message for at 
least two reasons. First, it is based on misuse of the CPPE matrix. 
[See comment 11] The CPPE matrix is used in the planning stage to 
alert planners of potential issues so they can plan accordingly. They 
then use this information in developing their conservation plans (i.e. 
they do not prescribe practices detrimental to water quality). Because 
the CPPE matrix provides a single rating for all parts of the country 
and all situations, it is well recognized that its results are 
generalized. Again, its use in the conservation planning process is 
only as a preliminary tool to let planners know the potential effects 
and to plan accordingly (.e. offer mitigating practices if needed). 
Secondly, this statement does not report the number of practices 
actually installed in isolation. These statistics are needed by GAO 
before a conclusion can be reached on whether conservation practices 
are installed singly without other mitigating practices. As previously 
provided, NRCS analysis of all EQIP data shows an average of four 
practices employed per EQIP contract. Therefore, USDA requests 
deletion of this misleading message. 

* Page 40, paragraph 2: the report states that "...nutrient management 
plans were funded on properties in less than one-third of the 
watersheds where soil conservation practices were funded that NRCS 
acknowledges could degrade water quality." [Now on page 46] 

Again, NRCS does not agree that its soil conservation practices are a 
threat to water quality. Numerous Federal and university research 
studies have documented the benefit of those practices for reducing 
nutrient and sediment transport to water bodies. Again, this statement 
appears to arise from the auditors' misinterpretation and misuse of 
the CPPE matrix. NRCS does not recommend practices in isolation. 
Instead, NRCS uses a comprehensive, site-specific conservation 
planning process to ensure that multiple resource concerns are 
addressed, including water quality. [See comment 12] In addition, 
although nutrient management is a frequently used approach for 
enhancing water quality, it is not the only practice used for 
mitigating water quality issues. NRCS has identified over 80 practices 
with potential water quality improvement attributes. Implying that 
NRCS' procedures for protecting water quality have not been followed 
in two-thirds of the watersheds without an analysis of the use of the 
other 80 plus water quality improving practices is not only a 
misinterpretation of the statistics, but this conclusion violates a 
basic tenet of statistical analysis by inferring a conclusion that far 
exceeds the limitations of the data on which it is based. [See comment 
13] 

Because NRCS does not know which watersheds GAO used in their 
analysis, NRCS examined all 30,000 watersheds and found about 80 
percent had water quality improving practices in place with a similar 
share reporting "Nutrient Management Plans." The failure to consider 
practices other than "nutrient management plans" clearly understates 
the number of water quality improving practices on the landscape. In 
addition, EQIP contracts typically last for about 3 years. Producers 
often adopt nutrient management practices well beyond the length of a 
contract because they recognize the economical and environmental 
benefits of improved nutrient use efficiency provided by those 
practices. This adoption would not show up in an analysis of existing 
contracts. USDA requests that either the above qualifiers be added to 
improve the accuracy of the report, or that this section be deleted. 

* Page 43, third paragraph: [Now on page 49] 

This concluding paragraph summarizes the same points presented in the 
text, asserting that NRCS practices may inadvertently conflict with 
pollution. In summary, it is clear that this assertion is based on 
misinterpretation and misuse of the CPPE matrix, a lack of knowledge 
of the NRCS conservation planning process, inferences that exceed the 
limitations of the data, and a disregard for university and research 
findings. For these reasons, USDA requests changes in this concluding 
paragraph consistent with those requested in this review. 
Specifically, please remove any assertion that USDA practices conflict 
with water quality improvements.[See comment 14] 

Additional Information for Objective #3: [See comment 15] 

The third stated objective for this report was to examine "the extent 
to which key programs complement EPA efforts to control such 
(nonpoint) pollution." Based on this objective, it would seem 
appropriate for GAO to provide such an evaluation. For several years, 
university and Federal research has documented the positive benefits 
of agricultural conservation practices at watershed scales as part of 
the
Conservation Effects Assessment Project. In all, 42 watersheds 
covering 13 million acres combined have been studied (14 by USDA-ARS, 
Institute of Food and Agriculture NIFA, and 11 by NRCS). As modeling 
results show in the table below, conservation practices on cultivated 
cropland have reduced sediment loss with runoff by 47-73 percent, 
nitrogen loss in runoff by 35-58 percent, nitrogen loss through 
leaching by 9-45 percent, and phosphorus loss by 33-59 percent.  

Table: Percent Reductions in Loss of Sediment, Nitrogen, and 
Phosphorus from Conservation Practice Use in Five Regions: 
     
Constituent: Sediment; 
Upper Mississippi: 61%; 
Chesapeake Bay: 55%; 
Great Lakes: 47%; 
Ohio-Tennessee: 52%; 
Missouri: 73%. 

Constituent: N (surface); 
Upper Mississippi: 45%; 
Chesapeake Bay: 42%; 
Great Lakes: 43%; 
Ohio-Tennessee: 35%; 
Missouri: 58%. 

Constituent: N (subsurface); 
Upper Mississippi: 9%; 
Chesapeake Bay: 31%; 
Great Lakes: 30%; 
Ohio-Tennessee: 11%; 
Missouri: 45%. 

Constituent: Total P; 
Upper Mississippi: 42%; 
Chesapeake Bay: 41%; 
Great Lakes: 39%; 
Ohio-Tennessee: 33%; 
Missouri: 59%. 

[End of table] 

These reports are available at the following Web site: [hyperlink, 
http://www.nrcs.usda.gov/wps/portal/nrcs/main/national/technical/nra/cea
p/pub]. 

To improve accuracy and balance of this GAO report, USDA requests that 
the above evaluations of conservation practices and associated Web 
site link be included in this report.      

* Page 2, paragraph 2 and footnote 2: 

GAO indicates USDA's farm commodity programs indirectly contributed to 
nonpoint source water pollution through policies that encouraged the 
use of fertilizers and pesticides. This statement makes it seem as 
though commodity programs are creating pollution. Farmers are now free 
to plant whatever they want There are little if any commodity program 
provisions that encourage excess crop production. If anything, given 
conservation compliance rules, commodity programs may actually reduce 
nonpoint pollution. 

* Page 24, footnote 39: [Now on page 28] 

Please reword the Conservation Reserve Enhancement Program (CREP) 
footnote to read: "Part of CRP, CREP authorizes State-Federal 
conservation partnerships that address specific State and nationally 
significant water quality, soil erosion, and wildlife habitat concerns 
related to agriculture." 

Concluding Comments: 

USDA appreciates the opportunity to review and comment on this draft 
report, but has serious concerns over several inaccuracies stated in 
the current version of the report. USDA requests that you please 
address the inaccuracies in the manner identified herein. The EPA 319 
program is important and addresses critical non-point source water 
quality issues. USDA will continue to work with our federal partners 
at EPA and the State water quality agencies to coordinate programs and 
efforts to address non-point source water quality issues. Please 
direct any questions pertaining to this review to Dr. Wayne Honeycutt, 
NRCS, Deputy Chief for Science and Technology, at 
Wayne.Honcycutt@wdc.usda.gov or (202) 720-4630. 

Sincerely, 

Signed by: 

Homer L. Wiles, Acting for: 

Dave White: 
Chief: 

The following are GAO's comments on the Department of Agriculture's 
letter dated April 20, 2012. 

GAO Comments: 

1. NRCS commented on what it referred to as "several inaccuracies" 
that remained after the draft report was sent to USDA for comment. As 
discussed in the Agency Comments section of this report, in examining 
NRCS information, we concluded that the EQIP data provided by USDA do 
not contain site-specific information on the extent to which 
alternative mitigation measures are employed when nutrient management 
plans are not used. We continue to believe, as we state in the report, 
that without examining such information, neither we nor NRCS could 
determine that certain practices are not having unintended effects on 
water quality. 

2. NRCS commented on our finding that from 2005 to 2010 the 
Environmental Quality Incentives Program funded nearly 8,000 practices 
that--if implemented in isolation--can increase nonpoint source 
pollution. Specifically, in its letter, the agency stated that "NRCS 
does not implement practices in isolation. In fact, NRCS analysis of 
its Environmental Quality Incentives Program data show an average of 
four practices funded per contract." We acknowledge that NRCS does not 
fund practices in isolation and have revised the report to clarify 
this statement. Nonetheless, our statement is correct in that the 
nearly 8,000 practices we identified have the potential to have a 
negative effect on water quality if, depending on site-specific 
conditions, the proper companion practices are not installed along 
with them. We note in the report that NRCS generally funds multiple 
practices per contract--an average of 4, according to NRCS' letter--
but that this figure does not shed light on the type of practices that 
are being installed and whether the additional practices properly 
mitigate the potentially negative effects on water quality. It was for 
this reason that we recommended that NRCS analyze available 
information, which may involve obtaining site-specific information 
from field offices, to determine the extent to which appropriate 
mitigation measures are implemented when nutrient management plans are 
not in use. 

3. NRCS stated that we misinterpreted the agency's use of its 
conservation practice physical effects (CPPE) assessments, noting 
that, "The CPPE matrix is a generalized, national tool that provides a 
first approximation of potential effects for each practice… 
Interpretation of the effects of individual practices using the CPPE 
matrix is not appropriate for assessing overall effects of site-
specific conservation plans." We acknowledge that USDA generally 
implements suites of practices as part of site-specific conservation 
plans and have made changes in the report to clarify this point. 
Nevertheless, as NRCS' letter states, the CPPE matrix provides an 
"approximation of potential effects" for each practice and, as such, 
can give an indication of the kind of effects that can occur if site-
specific conservation plans are not implemented as intended. For 
instance, NRCS' letter further states, "If a potentially negative 
effect is identified, then the conservation planner is alerted to this 
possibility so that they know to plan any site-specific mitigating 
practices to ensure positive outcomes." In our field work, we found 
indications that site-specific mitigating practices were not always 
implemented when practices with potentially negative effects on water 
quality were installed. In an effort to determine whether the 
instances we observed were anecdotal or more prevalent, we then 
examined EQIP data but found these data not sufficiently detailed to 
determine if appropriate site-specific mitigation practices were also 
implemented. 

4. NRCS commented on our finding that USDA data showed that nutrient 
management plans had not been funded for use on properties in about 
two-thirds of the 800 watersheds, stating that "this information is 
misleading and inaccurate…Although nutrient management is a frequently 
used approach for enhancing water quality, it is not the only practice 
used for mitigating water quality issues." We disagree that this 
information is misleading and inaccurate. In meetings in July and 
August 2011, NRCS officials stated that nutrient management planning 
was the agency's primary method for ensuring that those practices 
intending to conserve one resource (e.g., soil) do not inadvertently 
harm another resource (e.g., water). As we stated in our draft report, 
nutrient management plans are the mechanism that field planners often 
use to ensure water quality is protected on a property in an impaired 
watershed, and such plans might direct field planners to propose one 
or more of the 80 conservation practices that can improve water 
quality, according to NRCS officials. Our draft report also 
acknowledged that where nutrient management plans are not in place, 
other conservation planning procedures exist. For example, field 
planners might propose alternative mitigation measures--which would 
likely include one or more of the 80 conservation practices that can 
improve water quality--without requiring a landowner to adhere to a 
formal nutrient management plan. As we state in the report, because of 
limitations in the precision of EQIP data, the extent could not be 
ascertained to which alternative mitigation measures were in place for 
the roughly two-thirds of watersheds where nutrient management plans 
were not funded and where the data also showed that practices had been 
funded that could potentially degrade water quality. 

5. NRCS stated that because we did not "consider practices other than 
nutrient management," we "understate the number of water quality 
improving practices that are in place on the landscape." We are not 
questioning the number of water quality practices funded by NRCS; 
rather, it was the limitations of USDA's data that prevented us, and 
NRCS, from identifying the full extent to which those water quality 
improving practices were implemented where they should be. Several 
state environmental protection officials we spoke with echoed this 
concern, telling us that because they do not have information on where 
NRCS funds practices, they do not know whether, and where, they need 
to use section 319 funds to implement mitigating practices to protect 
water quality or other practices to complement NRCS efforts. The 
agency's summary data showing that "80 percent [of watersheds] had 
water quality improving practices" does not mean that such practices 
were always properly implemented with other companion practices to 
mitigate the effects of those practices that have the potential to 
degrade water quality. 

6. NRCS disagreed with our finding that soil conservation practices 
may sometimes adversely affect efforts to protect water quality. In 
addressing this issue above, we acknowledged that, in general, such 
practices have substantial environmental benefits, but that in certain 
instances, it is possible that certain conservation practices--
including some designed to minimize soil erosion--can negatively 
affect water quality when the proper companion practices are not also 
implemented. 

7. NRCS provided information showing that under EQIP, the agency 
rarely implements conservation practices in isolation and instead 
funds "an average of four practices per contract." We acknowledge, as 
previously stated, that NRCS rarely funds practices in isolation and 
have revised the report to clarify this statement. We also acknowledge 
that NRCS generally funds multiple practices per contract--an average 
of four, according to NRCS' letter--but this number alone does not 
shed light on the type of practices that are being installed and 
whether the additional practices mitigate the potentially negative 
effects on water quality in the watersheds we analyzed. 

8. NRCS commented on how we used information in its conservation 
practice physical effects assessments, asserting that "The CPPE matrix 
is simply a very rough first approximation of effects that is most 
often used as a training tool … and is not the only effects assessment 
tool used during site-specific conservation planning." We acknowledge, 
as we have above, that the CPPE matrix is not the only tool used 
during site-specific planning and that field planners generally 
propose "site-specific mitigating practices to ensure positive 
outcomes," according to NRCS' letter. Nevertheless, in our field work 
we found some indications that site-specific mitigating practices were 
not implemented when practices with potentially negative effects on 
water quality were installed. As noted above, it is the limitations of 
the EQIP data available to NRCS in illuminating whether such instances 
were isolated occurrences, or indicative of a broader issue, that led 
us to recommend that NRCS analyze available information, which may 
involve NRCS obtaining site-specific information from field offices, 
to determine the extent to which appropriate mitigation measures are 
implemented where nutrient management plans are not in use, 
particularly in watersheds where states are spending section 319 funds. 

9. NRCS stated that we misinterpreted the agency's CPPE matrix in 
describing underground outlet systems. This information, however, came 
directly from NRCS field office technical guides and other NRCS 
documents. We nonetheless added information, as NRCS suggested in its 
letter, to clarify that underground outlets "allow water to move more 
quickly and thus reduce the time that natural processes have to reduce 
nutrients or other agricultural chemicals." 

10. NRCS stated that our example of stream crossings in Washington 
state is an exception and that "miles of exclusion fencing are 
contracted every year to keep livestock out of streams, and stream 
crossings are appropriately utilized to help attain those water 
quality benefits." We acknowledge that the example from Washington 
State may well be an exception. Nevertheless, because of limitations 
in the availability of data, neither we nor NRCS could determine 
whether examples such as these were exceptions or more prevalent. As 
we stated earlier, it was for this reason, and to ensure that complete 
data are available to allow NRCS and others to assess whether the 
program has unintended water quality impacts, that we recommended that 
NRCS analyze information on the extent to which mitigation measures 
are implemented in situations where NRCS-funded conservation practices 
may negatively affect water quality (and particularly in watersheds 
where states are spending section 319 funds). 

11. NRCS stated that our statement that 17 percent of conservation 
practices were types of individual conservation practices that could 
degrade water quality was misleading and conveyed an inaccurate 
message. Specifically, NRCS stated that this finding is (1) based on 
misuse of the CPPE matrix, and (2) does not report the number of 
practices actually installed in isolation. To respond to NRCS' first 
point, as we stated above, we acknowledge that the CPPE matrix is not 
the only tool used during site-specific planning, but in our field 
work we found some indications that site-specific mitigating practices 
were not implemented when practices with potentially negative effects 
on water quality were installed. To respond to NRCS' second point, 
because of limitations in the precision of EQIP data, we could not 
ascertain the extent to which alternative mitigation measures were in 
place for the roughly two-thirds of watersheds where nutrient 
management plans were not funded and where the data also showed that 
practices had been funded that could potentially degrade water 
quality. Upon our request for information on alternative mitigation 
measures, NRCS officials told us that such information was documented 
and stored at the field level but that such information was not 
catalogued and available at the headquarters level. 

12. NRCS stated that we misinterpreted and misused the CPPE matrix. As 
we stated above, NRCS' letter states that the CPPE matrix provides an 
"approximation of potential effects" for each practice and, as such, 
can give an indication of the kind of effects that can occur if site-
specific conservation plans are not implemented as intended. In our 
field work, we found indications that site-specific mitigating 
practices were not always implemented when practices with potentially 
negative effects on water quality were installed. 

13. NRCS stated that we misinterpreted USDA statistics on the use of 
nutrient management plans and "violated a basic tenet of statistical 
analysis by inferring a conclusion that far exceeds the limitations of 
the data on which it is based." We acknowledge, as we did in our draft 
report, that where nutrient management plans are not in place, other 
conservation planning procedures exist. Because of limitations in the 
precision of EQIP data, however, the extent could not be ascertained 
to which alternative mitigation measures were in place for the roughly 
two-thirds of watersheds where nutrient management plans were not 
funded and where the data also showed that practices had been funded 
that could potentially degrade water quality. Upon our request for 
information on alternative mitigation measures, NRCS officials told us 
that such information was documented and stored at the field level, 
but that such information was not catalogued and available at the 
headquarters level. Rather than drawing "inferences that exceed the 
limitations of the data on which they are based," as NRCS stated, we 
concluded that neither we, nor NRCS, could draw such inferences 
without additional information on the extent to which alternative 
measures are employed--or if site-specific conditions make them 
unnecessary--on these EQIP contracts when nutrient management plans 
are not used. 

14. NRCS stated, "In summary, it is clear that this assertion is based 
on misinterpretation and subsequent misuse of the CPPE matrix, a lack 
of knowledge of the NRCS conservation planning process, inferences 
that exceed the limitations of the data, and a disregard for 
university and Federal research findings." We disagree. Our draft 
report acknowledged the noteworthy goals and accomplishments of EQIP 
in mitigating the impacts on water quality of certain agricultural 
practices. In response to the NRCS comments, we revised the language 
in the draft report to further discuss the program's benefits in 
finalizing the report. However, we found instances where NRCS 
conservation planning process did not mitigate certain practices' 
effects at the ground level, and where water quality had been impacted 
as a result. In an effort to ascertain whether the instances we 
observed were anecdotal or more prevalent, we examined USDA data and 
other information, including the CPPE matrix, and concluded that 
neither we nor NRCS could draw such inferences without additional 
information on the extent to which alternative measures are employed 
on Environmental Quality Incentives Program contracts when nutrient 
management plans are not used. It was for this reason--and to ensure 
that complete data are available to allow NRCS and others to assess 
whether the program has unintended water quality impacts--that we 
recommended that NRCS analyze available information and obtain 
necessary site-specific information from field offices as necessary. 
NRCS would then be in a better position to determine the extent to 
which appropriate mitigation measures are implemented when nutrient 
management plans are not in use (and particularly in watersheds where 
states are spending section 319 funds). 

15. NRCS provided additional information on EQIP's water quality 
benefits, specifically requesting that we include information from 
published studies on the benefits of agricultural conservation 
practices. The draft report cited EQIP's contributions toward 
improving water quality, but we nonetheless added language as 
suggested on the program's pollutant reductions in key watersheds and 
on NRCS reports describing the agency's water quality improvement 
efforts. 

[End of section] 

Appendix IV: GAO Contact and Staff Acknowledgments: 

GAO Contact: 

David C. Trimble, (202) 512-3841, or trimbled@gao.gov: 

Staff Acknowledgments: 

In addition to the individual named above, Steve Elstein; Assistant 
Director; Nathan Anderson; Elizabeth Beardsley; Mark Braza; Ellen Chu; 
Emily Eischen; Mitch Karpman; Jill Lacey; Dae Park; Kiki 
Theodoropolous; Jason Trentacoste; and Josh Wiener made key 
contributions to this report. 

[End of section] 

Footnotes: 

[1] In fiscal year 2012, the section 319 program received $165 million. 

[2] GAO, Water Pollution: Greater EPA Leadership Needed to Reduce 
Nonpoint Source Pollution, [hyperlink, 
http://www.gao.gov/products/GAO/RCED-91-10] (Washington, D.C.: Oct. 
15, 1990). 

[3] We visited Arkansas, Louisiana, Maryland, Michigan, Ohio, 
Pennsylvania, Washington, and West Virginia. 

[4] EPA's Grants Reporting and Tracking System is a database of 
section 319-funded projects awarded by states; the database does not 
include information on most section 319 projects awarded by Indian 
tribes. To assess the reliability of this database, we performed 
electronic data testing for missing data, outliers, and obvious 
errors. In addition, we reviewed internal controls for the database, 
interviewed knowledgeable agency officials, and compared summary 
information from the database with published reports. On the basis of 
this assessment, we determined that EPA's Grants Reporting and 
Tracking System data were sufficiently reliable for our purposes. 

[5] To obtain this information, we sent a data collection instrument 
to nonpoint source program coordinators in all 50 states and obtained 
responses from all states. 

[6] We obtained these data from the Program Contracts System, known as 
ProTracts, which is used to manage Natural Resources Conservation 
Service conservation program applications, cost-share contracts, and 
program fund management. To assess the reliability of these data, we 
performed electronic data testing for missing data, outliers, and 
obvious errors. In addition, we interviewed knowledgeable agency 
officials. On the basis of this assessment, we determined that 
ProTracts data were sufficiently reliable for our purposes. 

[7] EPA groups the primary types of nonpoint source pollution into 
seven categories--agriculture, urban and stormwater runoff, 
hydromodification (e.g., altering water patterns by straightening 
streams or building dams), resource extraction, silviculture (e.g., 
timber harvesting), construction, and land disposal--of which the 
first three have been identified by the agency as the greatest 
contributors to nonpoint source pollution. 

[8] For consistency, we refer to the statute and its amendments as the 
Clean Water Act throughout this report. 

[9] Environmental Protection Agency, Report to Congress: Nonpoint 
Source Pollution in the U.S., NEPIS 841R84100 (Washington, D.C.: 1984). 

[10] 33 U.S.C. §1329(h)(1), (2), (10). 

[11] 68 Fed. Reg. 60,653 (Oct. 23, 2003). 

[12] By regulation, EPA regional administrators can deny the award of 
section 319 funds to states in instances where a work plan does not 
briefly describe each significant category of nonpoint source activity 
and commitments to be produced for each category, among other things. 

[13] EPA's guidelines encourage states to select projects that 
complement USDA's conservation program efforts because the regulations 
for these programs assigned a priority to reducing nonpoint source 
pollution in impaired watersheds. In fiscal year 2012, USDA 
implemented a National Water Quality Initiative in which state 
conservationists reserved about $33 million in EQIP funds to help 
producers implement conservation practices to protect water quality in 
impaired or threatened watersheds. 

[14] State nonpoint source programs may also implement regulatory 
activities rooted in state authority. States may use section 319 funds 
to support such regulatory efforts, but section 319 does not provide 
regulatory authority to EPA or to states with respect to the control 
of nonpoint source pollution. 

[15] We characterized projects as supporting either direct or indirect 
approaches by analyzing EPA data on the functional categories of 
activities listed for section 319-funded projects. 

[16] Although the distinction between direct and indirect approaches 
is helpful for understanding the primary purposes of different 
projects, it is important to note that some section 319-funded 
projects supported elements of both approaches. In addition to listing 
a primary functional category of activity, some projects also listed 
secondary functional categories in EPA's Grants Reporting and Tracking 
System. For example, a project may have been designed primarily to 
install conservation practices (an example of a direct activity), but 
a secondary goal of the project may have involved conducting water 
quality monitoring (an example of an indirect activity). The data 
presented here are based exclusively on the primary functional 
category of activity listed for each project in EPA's Grants Reporting 
and Tracking System. 

[17] We developed these groups largely on the basis of how EPA 
categorizes different indirect activities, although they also reflect 
modifications based on the results of our analysis. 

[18] TMDL is a calculation of the maximum amount of a pollutant that a 
body of water can receive and still meet water quality standards. 
States can use up to 20 percent of section 319 funds to develop 
nonpoint source TMDLs, watershed-based plans to implement the nonpoint 
source TMDLs, and watershed-based plans in the absence of or before 
completing development of TMDLs for waters listed as impaired under 
section 303(d) of the Clean Water Act, among other things. The 
relevant EPA region can authorize a state to use over 20 percent of 
the incremental funding to develop watershed-based plans in section 
303(d)-listed waters, but the region is to ensure that a proper 
balance exists between funding the development of watershed-based 
plans and the implementation of watershed-based plans. 

[19] In instances where a water body has multiple segments listed by 
the state as impaired, each fully or partially restored segment is 
counted separately within the total number of restored water bodies 
reported by EPA. The number of partially or fully restored water 
bodies is based on those that had been listed as impaired during the 
1998-2000 listing cycle or in subsequent years. EPA considers a water 
body that has been included on the Clean Water Act section 303(d) list 
of impaired water bodies to be "fully restored" once the water body 
meets all water quality standards or designated uses. "Partially 
restored" water bodies are those that, as a result of restoration 
efforts, meet some, but not all, of their initially impaired water 
quality standards or designated uses. 

[20] We surveyed a random sample of managers of projects receiving 
section 319 funding that were completed from January 1, 2004, through 
June 9, 2011. The response rate was 57 percent. The 95 percent 
confidence interval for the 72 percent of projects is 67-78 percent 
and denoted (67, 78). For more detail on survey methodology, see 
appendix I. The full results of this survey are available at 
[hyperlink, http://www.gao.gov/products/GAO-12-377SP]. 

[21] According to the Michigan Department of Environmental Quality's 
Nonpoint Source Unit Chief, this project served as a demonstration for 
the rest of the community of effective low-impact development 
practices, and other local government buildings and businesses in the 
area have subsequently implemented similar practices, at least in part 
because of this project's success. 

[22] Estimates produced from the sample of projects are subject to 
sampling error. We express our confidence in the precision of our 
results as a 95 percent confidence interval. This interval would 
contain the actual population value for 95 percent of the samples we 
could have drawn. As a result, we are 95 percent confident that each 
of the confidence intervals in this report includes the true values in 
the study population. The 95 percent confidence interval for this 
estimate lies between 22 percent and 33 percent and is expressed as 
(22, 33). 

[23] States often select projects proposed by organizations that rely 
on voluntary third-party participation for implementation. Such 
organizations typically include soil and water conservation districts, 
county health departments, and nonprofit organizations. For example, 
project managers we surveyed reported that 81 percent of the 
organizations that implemented section 319-funded projects were 
organizations that typically rely on voluntary participation from 
third parties, such as landowners, to achieve project goals. Nonprofit 
organizations typically rely on voluntary participation from third 
parties, but in some cases they own the land on which projects are to 
be implemented. 

[24] The 95 percent confidence interval for this estimate is (23, 34). 

[25] The 95 percent confidence interval for this estimate is (33, 44). 

[26] Some projects may have required both grant revisions and 
extensions. 

[27] Prescribed burns were chosen to remove excessive invasive shrubs 
and trees, which would then allow the growth of desired native 
vegetation for erosion control. 

[28] Our survey results indicate that projects addressing agricultural 
runoff are more challenging than those addressing other forms of 
nonpoint source pollution: 86 percent of projects implemented in 
impaired watersheds that did not accomplish all their objectives were 
aimed at addressing agricultural runoff, whereas 14 percent of 
projects that did not accomplish their objectives addressed other 
forms of nonpoint source pollution. The 95 percent confidence 
intervals for these estimates are (74, 93) and (7, 26), respectively. 

[29] In June 2010, NRCS reported that conservation practices may 
sometimes be inadequate to address soil and water conservation when 
installed in isolation, and instead, suites of complementary practices 
should be used where possible to better ensure protection of both soil 
and water resources. Department of Agriculture, Natural Resources 
Conservation Service, Assessment of the Effects of Conservation 
Practices on Cultivated Cropland in the Upper Mississippi River Basin, 
Draft (Washington, D.C.: June 2010). 

[30] According to state environmental protection officials, 
agricultural issues are complicated and take a thorough assessment 
from trained and skilled specialists to determine what practices will 
address the observed degradation, as well as fit into farmers' 
management limitations. Once the proper suite of practices is 
determined, local and state specialists need to identify what program 
is the best fit to help support farmers' efforts to install the 
necessary practices. 

[31] Livestock exclusion fences, for instance, are often installed 
because when livestock graze close to water bodies and drink from 
them, they trample stream bank vegetation, break down the stream bank, 
and deposit manure--which can increase the amount of sediment, 
nutrients, and bacteria within the water bodies and alter the physical 
structure of streams as well. Fencing to exclude livestock may or may 
not need companion practices to fully reach its pollution control 
potential. According to state environmental protection officials, 
there are a number of reasons why livestock exclusion fencing may not 
need to include any companion practices. For example, such fencing may 
not need companion practices when off-stream watering facilities or 
areas already exist for a pasture. In addition, riparian buffers may 
not be needed if the stream corridor will naturally generate a 
riparian buffer once fencing is installed. 

[32] Stream crossings are stabilized areas or structures that allow 
livestock, people, equipment, or vehicles to cross a stream in a safe 
and environmentally sound manner. Stream crossings help prevent stream 
bank erosion, water pollution, and damage to stream beds and channels. 

[33] In 2010, in a draft Water Quality Manual for Livestock Grazing, 
Washington state's Department of Ecology identified additional 
practices that should be implemented together to reduce negative water 
quality impacts when exclusion fencing is installed, which may include 
riparian buffers, stream crossings, and off-stream water facilities 
where livestock can drink. An Ecology official explained that if 
section 319 funds are used to install such fencing without these 
additional practices, there is limited assurance that water quality 
will improve. 

[34] If complementary practices were installed together, one would 
generally expect the data to show a similar number for all practices. 
It is possible, however, that on some of these properties, the 
complementary practices were already in place or funded by the 
landowner or another organization. 

[35] In November 2011, EPA issued a national evaluation of the 319 
program. According to this report, projects that use indirect 
approaches provide funds to enable the states to work with federal, 
state, local, private-sector, and watershed groups to gain cooperation 
and to leverage dollars, authorities, and other resources to solve or 
prevent nonpoint source pollution problems. These funds also provide 
critical support for state staff to conduct project planning and 
selection, monitoring, and building of partnerships, which are 
critical to ensure successful implementation of watershed-based plans. 
Environmental Protection Agency, A National Evaluation of the Clean 
Water Act Section 319 Program (Washington, D.C.: November 2011). 

[36] Since 2004, 20 states have spent nearly $30 million on nearly 100 
projects that were designed, at least in part, to provide 
administrative support to USDA programs and activities or to directly 
promote these USDA programs and activities to landowners. The $30 
million includes four projects approved by Michigan's Department of 
Environmental Quality, totaling more than $10 million, which were 
designed to fund the state's nonpoint source management program staff, 
who are generally responsible for assisting with Conservation Reserve 
Enhancement Program agreements, among other activities. 

[37] CRP is a voluntary program for agricultural landowners through 
which they can receive annual rental payments and cost-share 
assistance to establish long-term, resource-conserving covers on 
eligible farmland. 

[38] CREP, as part of CRP, is a voluntary land retirement program that 
helps agricultural producers protect environmentally sensitive land, 
decrease erosion, restore wildlife habitat, and safeguard ground and 
surface water. CREP authorizes state-federal conservation partnerships 
that address specific state and nationally significant water quality, 
soil erosion, and wildlife habitat concerns related to agriculture. 

[39] Environmental Protection Agency, A National Evaluation of the 
Clean Water Act Section 319 Program (Washington, D.C.: November 2011). 

[40] In other words, USDA's project-ranking systems include 
significant weighting of water quality considerations, such as 
projects focused in impaired watersheds, nonpoint source program 
priority watersheds, or watersheds that have watershed-based plans. 

[41] Ohio's water quality goal is for 100 percent of large rivers and 
80 percent of small rivers to attain their designated use by 2020. 

[42] The report notes that 20 states' project selection processes 
explicitly take into consideration a project's feasibility for 
successful implementation. 

[43] 40 CFR §§ 35.111(a)(4) , 35.101(a)(10) (2012). 

[44] Region 5 comprises Illinois, Indiana, Michigan, Minnesota, Ohio, 
and Wisconsin. 

[45] Region 1 comprises Connecticut, Maine, Massachusetts, New 
Hampshire, Rhode Island, and Vermont. Massachusetts added an amendment 
to its management plan in 2007. 

[46] EPA's 2011 report does not specify which states have not upgraded 
their nonpoint source management program plans. 

[47] EPA's existing national water program guidance for fiscal year 
2013 provides little direction to EPA regional offices. EPA's 2003 
guidelines for section 319 grants are focused on state 
responsibilities. To the extent they discuss regional office 
oversight, the 2003 guidelines identify some responsibilities but 
generally do not advise regions on how they are to fulfill them. For 
example, the guidelines state that the region will work with a state 
to ensure that the work plan has programmatic, technical, and 
scientific merit, but there is no further direction on what regions 
should consider in ensuring this is achieved. 

[48] Region 4 states include Alabama, Florida, Georgia, Kentucky, 
Mississippi, North Carolina, South Carolina, and Tennessee. 

[49] In addition, regional offices also review proposed projects and 
activities for synopses explaining the state's strategy for using 
section 319 funds in the current fiscal year, including the nonpoint 
source problems to be addressed, the types of practices that will be 
implemented, and indicators of performance, among other things. 

[50] Because of the way EPA records information on the achievements of 
states' nonpoint source management programs, we could not determine 
whether water quality outcomes differed in the states that received 
more oversight than in those that did not. 

[51] We also found that regional offices use specific criteria in 
making these determinations about half the time. 

[52] EPA reported that in fiscal year 2010, projects funded at least 
in part by section 319 funds had achieved estimated nationwide 
reductions during the year of 9.7 million pounds of nitrogen, 2.6 
million pounds of phosphorus, and 2.1 million tons of sediment. 

[53] Grass filter strips do not absorb polluted runoff as well as 
natural riparian vegetation, according to the engineer. 

[54] EPA has singled out Ohio and Maine as states that have adopted 
biologically based information in how they measure water quality. Both 
Maine and Ohio developed and adopted their programs to incorporate 
ecologically relevant endpoints into decisions, inform water quality 
management decisions, and quantify water quality improvements, among 
other things. Maine and Ohio scientists have identified a sequence of 
steps and milestones that EPA has compiled as a template other states 
may use to develop biologically based measures of water quality. 

[55] Environmental Protection Agency, Use of Biological Information to 
Better Define Designated Aquatic Life Uses in State and Tribal Water 
Quality Standards: Tiered Aquatic Life Uses, (Washington, D.C.: Aug. 
10, 2005). 

[56] Water quality standards are established by each state, however, 
so the extent to which biological indicators are incorporated as 
measures of program success may depend in part on how states define 
their water quality standards. 

[57] According to EPA, this number does not reflect the net change in 
the number of impaired water bodies over this time period; water 
bodies are also being added to state lists of impaired waters each 
year. 

[58] The program funds conservation practices throughout the nation 
that are intended to reduce nonpoint source pollution and, among other 
things, soil erosion. 

[59] Reports describing such pollutant reductions are available at 
[hyperlink, 
http://www.nrcs.usda.gov/wps/portal/nrcs/main/national/technical/nra/dea
p/pub]. 

[60] EQIP offers cost-share payments and technical assistance to 
agricultural producers through 1-to 10-year contracts to implement 
conservation practices. Cost-share payments require agricultural 
producers to pay some percentage of the total cost of the conservation 
practices, which can include money or in-kind payments (e.g., labor or 
use of equipment). According to NRCS, the program funds an average of 
4 conservation practices per contract. 

[61] According to NRCS, underground outlet systems allow water to 
change elevations and move more quickly, which reduces the time that 
natural processes have to reduce nutrients or other agricultural 
chemicals. 

[62] Under the National Environmental Policy Act, federal agencies are 
to assess, among other things, the effects of major federal actions 
that significantly affect the environment and prepare a detailed 
statement on the environmental impacts of those actions. 

[63] In addition, nutrient management plans take into account critical 
factors that may affect water quality, including a property's slope 
gradient, soil composition and density, and distance to a water body. 
For some sites that are at high risk for contributing pollution to 
nearby water bodies, NRCS' nutrient management plans require the 
landowner to limit fertilizer application to the amount that can be 
taken up by the crop, among other things, to protect water quality. 

[64] According to NRCS, this scenario is likely an exception. Adding 
stream crossings generally allows cattle to travel from one side of a 
property to another and concentrates their crossing at one place, 
thereby reducing stream bank erosion and nutrient runoff in streams. 
Stream crossings are generally part of suites of companion practices 
whose net effect on water quality is positive, according to NRCS. 

[65] Underground outlets, access roads, and subsurface drains, 
accounted for 80 percent of the nearly 8,000 conservation practices. 

[66] NRCS officials also commented that in some instances a landowner 
might have taken the initiative to choose an alternative mitigation 
measure that protects soil and water quality, without needing to 
adhere to a formal nutrient management plan. 

[67] Specifically, NRCS emphasized the rigor of its quality control 
process noting, for example, that the program uses a ranking process 
for proposed projects that gives greater weight to projects 
implementing complete suites of practices over individual practices. 
In addition, NRCS field staff review each contract to ensure it 
complies with the agency's procedures, and that practices are planned 
and applied appropriately for site-specific concerns and conditions. 
After conservation practices are implemented, NRCS state or field 
staff spot-check 5 percent of all program contracts to ensure 
practices are appropriately planned and applied for site-specific 
conditions. 

[68] The initial random sample was 618. After adjustments were made to 
the eligible survey population, the random sample was 524 managers of 
projects. 

[End of section] 

GAO’s Mission: 

The Government Accountability Office, the audit, evaluation, and 
investigative arm of Congress, exists to support Congress in meeting 
its constitutional responsibilities and to help improve the 
performance and accountability of the federal government for the 
American people. GAO examines the use of public funds; evaluates 
federal programs and policies; and provides analyses, recommendations, 
and other assistance to help Congress make informed oversight, policy, 
and funding decisions. GAO’s commitment to good government is 
reflected in its core values of accountability, integrity, and 
reliability. 

Obtaining Copies of GAO Reports and Testimony: 

The fastest and easiest way to obtain copies of GAO documents at no 
cost is through GAO’s website [hyperlink, http://www.gao.gov]. Each 
weekday afternoon, GAO posts on its website newly released reports, 
testimony, and correspondence. To have GAO e-mail you a list of newly 
posted products, go to [hyperlink, http://www.gao.gov] and select “E-
mail Updates.” 

Order by Phone: 

The price of each GAO publication reflects GAO’s actual cost of 
production and distribution and depends on the number of pages in the 
publication and whether the publication is printed in color or black 
and white. Pricing and ordering information is posted on GAO’s 
website, [hyperlink, http://www.gao.gov/ordering.htm]. 

Place orders by calling (202) 512-6000, toll free (866) 801-7077, or 
TDD (202) 512-2537. 

Orders may be paid for using American Express, Discover Card, 
MasterCard, Visa, check, or money order. Call for additional 
information. 

Connect with GAO: 

Connect with GAO on facebook, flickr, twitter, and YouTube.
Subscribe to our RSS Feeds or E mail Updates. Listen to our Podcasts.
Visit GAO on the web at [hyperlink, http://www.gao.gov]. 

To Report Fraud, Waste, and Abuse in Federal Programs: 

Contact: 
Website: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]; 
E-mail: fraudnet@gao.gov; 
Automated answering system: (800) 424-5454 or (202) 512-7470. 

Congressional Relations: 

Katherine Siggerud, Managing Director, siggerudk@gao.gov, (202) 512-4400
U.S. Government Accountability Office, 441 G Street NW, Room 7125
Washington, DC 20548. 

Public Affairs: 
Chuck Young, Managing Director, youngc1@gao.gov, (202) 512-4800
U.S. Government Accountability Office, 441 G Street NW, Room 7149 
Washington, DC 20548.