Decision Not To Regulate Forest Road Discharges Under the Clean Water Act; Notice of Decision, 43492-43510 [2016-15844]

Agencies

• ENVIRONMENTAL PROTECTION AGENCY

[Federal Register Volume 81, Number 128 (Tuesday, July 5, 2016)]
[Rules and Regulations]
[Pages 43492-43510]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 2016-15844]


-----------------------------------------------------------------------

ENVIRONMENTAL PROTECTION AGENCY

40 CFR Chapter I

[EPA-HQ-OW-2015-0668; FRL-9948-62-OW]


Decision Not To Regulate Forest Road Discharges Under the Clean 
Water Act; Notice of Decision

AGENCY: Environmental Protection Agency (EPA).

ACTION: Decision.

-----------------------------------------------------------------------

SUMMARY: The Environmental Protection Agency (EPA) is providing notice 
of the Agency's decision that no additional regulations are needed to 
address stormwater discharges from forest roads under Section 402(p)(6) 
of the Clean Water Act (CWA) at this time. This document responds to 
the remand in Environmental Defense Center, Inc. v. U.S. EPA, 344 F.2d 
832 (9th Cir. 2003) that requires EPA to consider whether the CWA 
requires the Agency to regulate stormwater discharges from forest 
roads.

DATES: This decision shall be considered issued for purposes of 
judicial review at 1 p.m. Eastern time on July 11, 2016.

FOR FURTHER INFORMATION CONTACT: Prasad Chumble, EPA Headquarters, 
Office of Water, Office of Wastewater Management via email at 
chumble.prasad@epa.gov or telephone at 202-564-0021.

SUPPLEMENTARY INFORMATION: 

I. General Information

A. Applicability

    This document does not impose requirements on any entity.

B. Obtaining Copies of This Document and Related Information

1. Docket
    EPA has established a docket for this action under Docket ID No. 
[EPA-HQ-OW-2015-0668; FRL-9948-62-OW]. Publicly available docket 
materials are available either electronically through 
www.regulations.gov or in hard copy at the EPA Docket Center, (EPA/DC) 
EPA West, Room 3334, 1301 Constitution Ave. NW., Washington, DC. The 
EPA Docket Center Public Reading Room is open from 8:30 a.m. to 4:30 
p.m.,

[[Page 43493]]

Monday through Friday, excluding legal holidays. The telephone number 
for the Public Reading Room and the Docket Center is (202) 566-1744.
2. Electronic Access
    You may access this Federal Register document electronically from 
the Government Printing Office under the ``Federal Register'' listings 
at FDSys (https://www.gpo.gov/fdsys/browse/collection.action?collectionCode=FR).
3. Dates
    In accordance with 40 CFR part 23, this decision shall be 
considered issued for purposes of judicial review at 1 p.m. Eastern 
time on July 11, 2016. Under Section 509(b)(1) of the CWA, judicial 
review of this decision can be had only by filing a petition for review 
in the U.S. Court of Appeals within 120 days after the decision is 
considered issued for purposes of judicial review.

II. Executive Summary

    EPA has determined not to designate stormwater discharges from 
forest roads for regulation under Section 402(p)(6) of the Clean Water 
Act (CWA) at this time. EPA's decision is based on several interrelated 
factors. First, state, federal, regional, tribal government, and 
private sector programs already exist nationwide to address water 
quality problems caused by discharges from forest roads. Many of these 
programs have been improved and updated in recent years. Program 
implementation rates are generally high and have been shown to be 
effective in protecting water quality when properly implemented. These 
programs employ a variety of approaches, based in part on variations in 
regional topography and climate. While EPA recognizes that existing 
programs vary in their degree of rigor, the Agency has concluded that 
efforts to help strengthen existing programs would be more effective in 
further addressing forest road discharges than superimposing an 
additional federal regulatory layer over them.
    Some commenters have asserted that federal regulatory requirements 
could, in theory, promote national consistency and improvements in less 
effective programs. In practice, however, federal forest roads 
regulation presents a number of challenges that make achievement of 
that result unlikely. Wide variations in topography, climate, 
ownership, management, and use across the nation's network of forest 
roads make the establishment of any nationwide regulatory program a 
complex and difficult endeavor. Mechanisms for implementation and 
enforcement of any federal regulatory requirements are limited, as 
recent amendments to CWA Section 402(l) preclude both the use of 
National Pollutant Discharge Elimination System (NPDES) permits to 
regulate most discharges from forest roads and citizen suit enforcement 
of any Section 402(p)(6) requirements. Some commenters discussed the 
failings of existing best management practices (BMP) programs, 
including insufficient compliance rates and compliance monitoring, but 
a federal EPA-administered program would not necessarily be able to 
address these challenges more effectively than entities with regional 
expertise overseeing existing forestry management practice programs, 
especially without the accountability mechanisms afforded by a 
permitting program or third-party enforcement.
    For these reasons, elaborated upon below, EPA is exercising the 
``broad discretion the CWA gives the EPA in the realm of stormwater 
runoff,'' in deciding not to regulate stormwater discharges from forest 
roads. See Decker v. Nw. Envtl. Def. Ctr., 133 S. Ct 1326, 1338 (2013) 
(affirming EPA's determination not to regulate stormwater discharges 
from logging roads in its industrial stormwater rule). Instead, EPA 
intends to work in consultation with state and local officials, as well 
as other federal agencies and interested stakeholders, to help 
strengthen their existing programs and improve awareness and 
implementation of forestry best management practices. In reaching this 
conclusion, the Agency is cognizant that the CWA reserves for states 
``the primary responsibilities and rights . . . to prevent, reduce, and 
eliminate pollution [and] to plan the development and use (including 
restoration, preservation, and enhancement) of land and water resources 
. . .'' 33. U.S.C. 1251(b).

III. Legal Background

    The objective of the CWA is to restore and maintain the chemical, 
physical, and biological integrity of the nation's waters. 33 U.S.C. 
1251(a). To that end, the CWA provides that the discharge of any 
pollutant by any person shall be unlawful, except in compliance with 
other provisions of the statute. The CWA provides for a permit program, 
in general, for the discharge of a pollutant from a ``point source,'' 
which is defined in Section 502 of the CWA as ``any discernible, 
confined and discrete conveyance, including but not limited to any 
pipe, ditch, channel, tunnel, conduit, well, discrete fissure, 
container, rolling stock, concentrated animal feeding operation, or 
vessel or other floating craft, from which pollutants are or may be 
discharged.'' 33 U.S.C. 1362(14). In 1987 Congress added Section 402(p) 
to the CWA, which required NPDES permits for certain specified 
stormwater discharges and provided EPA with discretion to determine 
whether and how discharges from other stormwater sources should be 
addressed ``to protect water quality.'' See Northwest Environmental 
Advocates v. EPA, 640 F.3d 1063, 1083 (9th Cir. 2011) (``[i]t is within 
the discretion of EPA to promulgate Phase II regulations requiring, or 
not requiring, permits for such discharges'').
    For the initial phase of stormwater regulation, Section 402(p)(1) 
created a temporary moratorium on NPDES permits for point sources 
except for those listed in Section 402(p)(2). Section 402(p)(2) 
includes discharges already required to have a permit; discharges from 
municipal separate storm sewer systems serving a population of 100,000 
or more; and stormwater discharges ``associated with industrial 
activity.'' Congress did not define discharges associated with 
industrial activity, allowing EPA to interpret the term. For other 
stormwater discharges, Section 402(p)(5) directs EPA to conduct 
studies, in consultation with the states, for ``identifying those 
stormwater discharges or classes of stormwater discharges for which 
permits are not required''; ``determining to the maximum extent 
practicable, the nature and extent of pollutants in such discharges''; 
and ``establishing procedures and methods to control stormwater 
discharges to the extent necessary to mitigate impacts on water 
quality.''
    Section 402(p)(6) authorizes the Administrator to issue 
regulations, in consultation with state and local officials, based on 
the studies prescribed by Section 402(p)(5). It provides EPA discretion 
in selecting which discharge sources to regulate and how to regulate 
them; it does not require the use of NPDES permits. Specifically, the 
section states that the regulations ``shall establish priorities, 
establish requirements for state stormwater management programs, and 
establish expeditious deadlines'' and may include ``performance 
standards, guidelines, guidance, and management practices and treatment 
requirements, as appropriate.'' 33 U.S.C. 1342(p)(6). This flexibility 
is unique to stormwater discharges regulated under Section 402(p)(6) 
and differs from the requirement for NPDES permits for stormwater 
discharges listed in Section 402(p)(2) of the Act.

[[Page 43494]]

    In 1990, EPA promulgated the Phase I stormwater regulations (55 FR 
47990, November 16, 1990) (``Phase I Rule''), following the 1987 CWA 
amendments which directed the Agency to develop regulations requiring 
permits for large and medium municipal separate storm sewer systems and 
stormwater ``discharges associated with industrial activity.'' In March 
1995, EPA submitted to Congress a report on the results of the Section 
402(p)(5) study that evaluated the nature of stormwater discharges from 
municipal and industrial facilities not already regulated under the 
Phase I regulations (EPA, 1995). On December 8, 1999, EPA promulgated 
the Phase II stormwater regulations to address stormwater discharges 
from small municipal separate storm sewer systems and construction 
sites that disturb one to five acres. 64 FR 68722. Under CWA Sections 
402(p)(2)(E) and 402(p)(6), EPA retains the discretionary authority to 
designate additional stormwater discharges for regulation.
    The Phase II stormwater regulations were challenged in 
Environmental Defense Center v. US EPA, 344 F.3d 832 (9th Cir. 2003) 
(EDC v. EPA). In that case, petitioners contended that EPA arbitrarily 
failed to regulate discharges from forest roads under the Phase II 
rule. The court held that EPA failed to consider petitioners' comments 
and remanded the issue to EPA ``so that it may consider in an 
appropriate proceeding Petitioner's contention that Section 402(p)(6) 
requires the EPA to regulate forest roads. The EPA may then either 
accept Petitioners' arguments in whole or in part, or reject them on 
the basis of valid reasons that are adequately set forth to permit 
judicial review.'' Id. at 863.
    In the years following the decision in EDC v. EPA, EPA undertook 
research to improve the Agency's knowledge of the water quality impacts 
of forest road stormwater discharges and the programs that exist to 
reduce those impacts. During that period, the Northwest Environmental 
Defense Center initiated litigation concerning logging road stormwater 
discharges. In 2011, the U.S. Court of Appeals for the Ninth Circuit 
issued a decision in Northwest Environmental Defense Center v. Brown, 
640 F.3d 1063 (9th Cir. 2011) (``NEDC''), a citizen suit alleging 
violations of the CWA for unpermitted discharges of stormwater from 
ditches alongside two logging roads in state forests. The court held 
that because the stormwater runoff from the two roads in question is 
collected by a system of ditches, culverts, and channels and then 
discharged into waters of the U.S., there was a point source discharge 
of stormwater associated with industrial activity for which an NPDES 
permit is required.
    On May 23, 2012, EPA published a Notice in the Federal Register 
summarizing known water quality impacts related to forest roads and 
discussing existing state, tribal, and voluntary programs designed to 
address those impacts. (77 FR 30473). The Notice expressed EPA's intent 
to specify that only stormwater discharges associated with rock 
crushing, gravel washing, log sorting, and log storage are discharges 
associated with silvicultural activity that are subject to permitting 
under the stormwater regulations pertaining to industrial activity. The 
Notice also discussed the Agency's consideration of non-permitting 
approaches to address other stormwater discharges from forest roads. On 
December 7, 2012, EPA promulgated a rule (77 FR 72970) clarifying that 
discharges of stormwater from silviculture activities other than rock 
crushing, gravel washing, log sorting, and log storage do not require 
an NPDES permit. On March 20, 2013, the Supreme Court reversed the 
Ninth Circuit's ruling in NEDC, holding that discharges of stormwater 
that ran off logging roads into ditches, culverts, and channels did not 
require an NPDES permit as stormwater from industrial activity. See 
Decker v. Nw. Envtl. Def. Ctr., 133 S. Ct 1326 (2013).
    In January 2014, Congress amended CWA Section 402(l) to effectively 
prohibit the requirement of NPDES permits for the discharge of runoff 
``resulting from the conduct of the following silviculture activities 
conducted in accordance with standard industry practice: nursery 
operations, site preparation, reforestation and subsequent cultural 
treatment, thinning, prescribed burning, pest and fire control, 
harvesting operations, surface drainage, or road construction and 
maintenance.'' 33 U.S.C. 1342(l). In addition, the amendment prohibits 
third-party lawsuits (``citizen suits'') authorized by CWA Section 
505(a) for any requirements established under Section 402(p)(6) for the 
silviculture activities listed above.
    In December 2014, EDC and the Natural Resources Defense Council 
filed a petition with the Ninth Circuit to compel EPA to respond, 
within six months, to the question remanded in the 2003 EDC v. EPA 
decision of whether Section 402(p)(6) requires federal regulation of 
stormwater discharges from forest roads. Following execution of a 
settlement agreement filed with the court on August 26, 2015, the court 
entered an order establishing a schedule requiring EPA to issue a final 
determination by May 26, 2016. The parties subsequently extended the 
deadline by joint stipulation to June 27, 2016.

IV. Background on Forest Roads and Their Water Quality Impacts

    Forests cover about one-third of the continental U.S. 
(approximately 816 million acres). Over half are privately owned (58% 
or approximately 475 million acres) (USFS, 2016). Of private forest 
land, 63% is owned by families and individuals and is commonly referred 
to as ``family forests.'' Most of the family forest owners (around 62%) 
own fewer than 10 acres of forest land. Owners of the remaining private 
forest land include corporations, Real Estate Investment Trusts 
(REITs), conservation organizations, clubs, and Native American tribes 
(USFS, 2016). Over 300 Native American reservations are significantly 
forested, and Native American tribal lands include 18.6 million acres 
of forest land, including 1.5 million acres of productive timberland 
(Bureau of Indian Affairs, 2009). Private forest land owners invest 
considerable resources in forest road construction and maintenance, as 
they are critical assets that enhance property values, maintain 
economic viability, and facilitate sustainable forestry.
    Forty-two percent of forest land, or approximately 341 million 
acres, is publicly-owned. The federal government administers an 
estimated 74% of the public forest land. State forestry, park, and 
wildlife agencies account for most of the 22% of state-owned public 
forest land. The remaining 4% of public forest land is owned by local 
governments, such as counties and towns (USFS, 2016). Within the U.S., 
the distribution of public versus private forests differs greatly among 
the various regions of the country. For example, forest ownership in 
the Northwest is dominated by public ownership, primarily by the U.S. 
Forest Service (USFS) and the Bureau of Land Management (BLM). Private 
ownership is more prevalent in the Southeast and Northeast (Id.).
    Forests are connected by a vast network of forest roads built over 
the course of more than a century. Roads exist in forests for all land 
ownership categories, enabling activities as varied as timber 
operations, recreation, fire protection and general transportation. 
Originally some were built to allow mining or agriculture. The network 
of forest roads includes both active and inactive roads that vary in 
age and condition, and which often serve multiple purposes by multiple 
users at

[[Page 43495]]

the same time. Because of the nature of timber growing, timber roads 
are often used just once every fifteen or twenty years. Endicott (2008) 
noted that:

[e]ach forest road network commonly contains a collection of older 
and newer roads, designed to different standards, for various 
purposes, and crossing terrain of differing sensitivities. This 
mosaic of road segments has implications for how the forest road 
network will interact with the forest watershed, streams, and other 
downstream aquatic resources.

    A single road may be subject to different owners and managers and 
used for different activities at different points. Often the owner of 
the road is not the owner of the forest land over which the road 
travels. For example, a BLM-owned road may pass through private 
property or a timber company-owned road may pass through a state-owned 
public forest. The purpose of a road may also change at different 
points; for example, most of a road may be used for recreation but a 
small part of it may service a timber operation. Legacy roads pose 
particular concerns for water quality. Built prior to the adoption of 
modern BMPs, they may be poorly sited or designed and frequently no 
owner or operator assumes responsibility for those roads.
    As previously discussed in 80 FR 69655-69656 (November 10, 2015) 
and 77 FR 30476 (May 23, 2012), the Agency's research indicates that 
improperly designed, constructed, maintained, or decommissioned forest 
roads can impact water quality. These impacts are variable and may 
include increased sediment load and changes in stream network 
hydrology, which can cause physical, biological, and ecological impacts 
to water quality and aquatic organisms.
    Erosion from many forest roads does not affect water quality. 
First, roads that are not hydrologically connected to a stream do not 
deliver sediment to water bodies. For example, Dube et al. (2010), 
found that in an inventory of forest roads in 60 random four-square-
mile sections of forests in the Washington State, only 11% were 
connected to streams; Skaugset and Allen (1998) surveyed 287 miles of 
forest roads in 5 regions of Oregon and determined that 25% of forest 
roads drained directly to streams while another 6% were rated 
``possible'' for sediment delivery. Second, a variety of factors play a 
role in how water quality is impacted by forest roads, including road 
design, road surfaces, construction, maintenance, rate of use, 
topography, soil characteristics, precipitation patterns, and proximity 
of roads to surface water. The source of water quality impacts tends to 
be localized.
    Available data suggest that the number of surface waters impacted 
by silvicultural operations, including forest roads, is a small 
percentage of Section 303(d) listed impaired waters. EPA's analysis of 
the data shows that this trend has been consistent over time, 
indicating that water quality impacts appear to have persisted over 
time, but comprise only a small percentage of all sources of 
impairment. Specifically, results of nationwide waterbody assessments 
from the EPA's Assessment and Total Maximum Daily Loads (TMDL) Tracking 
and Implementation System (ATTAINS),\1\ which contains the most 
currently available data reported by states to the EPA under Sections 
305(b) and 303(d) of the CWA, found silviculture, which includes a 
broad spectrum of forestry activities including regulated 
activities,\2\ contributed to impairment of 40,637 miles of rivers and 
streams (7% of the total of 614,153 miles impaired) and 159,920 acres 
of lakes, reservoirs and ponds (1% of the total of 13,009,273 acres of 
impaired) (ATTAINS 2016). ``Forest roads (road construction and use)'' 
or ``logging roads'' are listed as the ``probable source'' of 
impairment for 31,076 miles of rivers and streams (5% of total 
impaired) and 7,627 acres of lakes, reservoirs and ponds (less than 1% 
of total impaired).
---------------------------------------------------------------------------

    \1\ https://iaspub.epa.gov/waters10/attains_index.home
    \2\ Non-point source silvicultural activities include nursery 
operations, site preparation, reforestation and subsequent cultural 
treatment, thinning, prescribed burning, pest and fire control, 
harvesting operations, surface drainage in addition to road 
construction and maintenance from which there is natural runoff at 
issue here.
---------------------------------------------------------------------------

    The extent of the impacts of silvicultural activities on water 
quality varies by region. Impairment data from states that report 
probable sources of impairments suggest that forest roads constitute a 
relatively low percentage of impairments. Examples of states where 
silviculture (a broader category that includes forest roads) is 
identified as a probable source of impairment and that document a 
percentage of the total river and stream miles impaired by `forest 
roads' or `logging roads' include: Idaho (0.62%; forest roads); 
Kentucky (0.04%; forest roads); Montana (5.71%); New Mexico (1.97%); 
and Pennsylvania (0.01%) (ATTAINS 2016). Road-related pollutant loading 
and impairments, however, may represent a higher percentage of 
impairments within specific regions. For example, within federal lands 
in the interior Columbia Basin, roads were identified as the largest 
source of sediment from any land management activity.\3\
---------------------------------------------------------------------------

    \3\ https://www.fs.fed.us/pnw/publications/icbemp.shtml
---------------------------------------------------------------------------

    EPA recognizes that the national water quality data discussed above 
have certain limitations. One limitation is that some states, when 
compiling their Section 305(b) reports, may not report the probable 
source of an impairment or may list probable impairment sources as 
unspecified, unknown, or in some other category, which may lead to 
underreporting of the source of the impairment. Additionally, some 
states may not assess all of their waters or may use different 
methodologies to collect or report water quality data, limiting the 
ability of drawing national-scale conclusions.
    ATTAINS data indicating the effect of discharges from forest roads 
on water quality impairments may therefore not be fully representative 
due to reporting differences among states. For example, of the 40,637 
miles of rivers and streams that ATTAINS indicates are impaired by 
silviculture, the database shows that California accounts for 34,443, 
or 85%, nationally (ATTAINS, 2016). Some regions in California use a 
particular approach toward classifying impairments that increases the 
reported percentage of impaired miles. Unlike other states, if a given 
reach of river is identified as impaired for a particular pollutant, 
some California regions categorize all of the river miles in the entire 
watershed as impaired.
    It is also important to recognize that EPA's data collection 
methods have changed over time. While ATTAINS compiles state-level 
data, it relies on the states for this information. The National Water 
Quality Initiative (NWQI), conducted by EPA, provides very specific 
information on impairments and sources, but EPA no longer collects 
these data. EPA currently uses probabilistic approaches (such as the 
Wadeable Streams Assessment and the National Rivers and Streams 
Assessment) to collect national-scale data on water quality. While 
these assessment approaches are sound, they do not reveal specific 
impairments and causes and therefore are less informative for purposes 
of this analysis.
    Estimating sedimentation specifically related to forest road 
discharges is also difficult as a practical matter. Unlike industrial 
and wastewater facilities, which typically have water quality 
monitoring to provide background data for assessing compliance with 
water quality standards, there is little to no regular monitoring of 
water quality in waters affected by forest road

[[Page 43496]]

discharges. Endicott (2008) noted that ``[e]ven a well-designed erosion 
experiment frequently results in variations from the mean of up to 
50%.'' Investigators may also be unable to differentiate among sediment 
generated from forest roads and sediment generated from other 
silvicultural activities, background erosion rates, or other sources. 
Endicott (2008) further explains that: ``Numerous studies have 
demonstrated that the biotic and chemical ``noise'' in larger streams 
renders the water quality effects of forestry activities using BMPs 
undetectable.'' Finally, Endicott (2008) recognizes that quantitative 
data can be difficult to obtain because ``impairments can be difficult 
to detect and/or measure'' and ``[e]rosion only usually occurs during 
wet weather.''

V. Role and Effectiveness of Forestry Best Management Practices

    The U.S. Forest Service defines Best Management Practices (BMPs) as 
the following:

    A practice or a combination of practices, that is determined by 
a State (or designated area-wide planning agency) after problem 
assessment, examination of alternative practices and appropriate 
public participation to be the most effective, practical (including 
technological, economic, and institutional considerations) means of 
preventing or reducing the amount of pollution generated by nonpoint 
sources to a level compatible with water quality goals (USFS, 1988).

    In the context of forest roads, BMPs focus on preventing and 
mitigating water quality impacts that may stem from the construction, 
maintenance and use of forest roads. Forest road BMPs are on the ground 
activities and structures that, in most cases, aim to prevent 
discharges of sediment from roads to streams. BMPs may also target 
other suspended solids, spills and residues, changes in water 
temperature, and alterations to flow regimes. In some cases they are 
designed to protect stream geomorphology and habitat for certain 
species.
    BMPs for forest roads generally fall into three categories: BMPs 
addressing road planning and design, road construction and 
reconstruction, and road management (e.g., Endicott 2008). Over the 
past several decades BMPs have been developed, evaluated, and improved 
based on ongoing research and technical innovation. BMPs are now widely 
implemented as standard elements of most private, state, and federal 
forestry programs (Ice et al., 2010). State-specific BMP programs and 
guidelines are available in most states (NCASI, 2009). Although the 
primary purpose of BMPs is to reduce environmental impacts, they must 
also be feasible and practical (Ice, 2004).
    BMPs are generally selected based on site-specific needs and 
conditions, which vary tremendously. Proximity of the road to the 
stream, size of the road, local geology and climate all influence the 
occurrence and magnitude of erosion and consequently the types of BMPs 
that will be most effective. For example, use of gravel to cover a road 
surface can be a highly effective erosion control BMP in steep terrain. 
In flat terrain, that same BMP would be less effective and much more 
expensive than a properly maintained continuous roadside berm 
(Appelboom et al., 2002).
    While BMP design is site-specific, many documents describe the most 
common BMPs (e.g., NCASI, 2001; EPA, 2005; NCASI, 2009; USFS, 2012; 
NCASI, 2012). This document does not provide a detailed discussion of 
the BMPs themselves; a number of comprehensive sources regarding 
different types of BMPs are available and included in the record for 
this decision (e.g., NCASI, 2009; Endicott, 2008; North Carolina 
Forestry BMP Manual; Montana Forestry BMP Manual). Most BMPs are based 
on relatively few guiding principles (Megahan and King, 2004; Olszewski 
and Jackson, 2006). These include:
     Use existing roads when practicable;
     Inventory road and stream conditions;
     Identify and avoid high-erosion hazard areas;
     Minimize the total land area disturbed;
     Minimize road crossings and other incursions into 
waterbodies;
     Engineer stable road surfaces, drainage features and 
stream crossings to reduce erosion;
     Separate bare ground from surface waters and minimize 
delivery of road-derived sediments to streams;
     Provide a forested buffer around streams;
     Design and install stream crossings to allow passage of 
fish, other aquatic biota, and large wood;
     Anticipate and mitigate erosion from precipitation events, 
including especially large ones;
     Regularly inspect all BMPs and erosion-prone areas, 
including during and/or immediately following precipitation and 
snowmelt events that may generate runoff; and
     Maintain forest roads and all BMPs.
    EPA notes that BMPs currently play and historically have played a 
significant role in wet weather \4\ and non-point source control 
programs. The scientific literature increasingly demonstrates the 
effectiveness of BMPs in preventing, minimizing, and mitigating 
discharges affecting water quality and aquatic habitats (Ice, 2004; 
Anderson and Lockaby, 2011; NCASI, 2012; Cristan et al., 2016; Endicott 
(2008)). Although existing research has significantly improved the 
effectiveness of forest road BMPs, reducing water quality impacts from 
road construction and other practices, many discharges still occur 
(Anderson and Lockaby, 2011). Further research would help to optimize 
operation and maintenance and provide guidelines for adapting BMP 
implementation to site-specific needs.
---------------------------------------------------------------------------

    \4\ 40 CFR 122.44(k).
---------------------------------------------------------------------------

    Several commenters cited a report by Cristan et al. (2016) --
``Effectiveness of Forestry Best Management Practices in the United 
States: Literature Review''--which summarized 81 BMP effectiveness 
studies: 30 studies of southern states, 20 studies of northern states, 
and 31 studies of western states.
    The review concluded generally that:
     Forestry BMPs minimize water quality effects of forest 
operations when implemented as recommended by state forestry and water 
quality agencies.
     Forest roads, skid trails, and stream crossings warrant 
considerable attention because they have the greatest potential for 
erosion and sediment delivery.
     Many studies across the U.S. have shown BMPs to be 
effective and reduce sediment delivery to streams.
    Several of the studies in the review assessed BMP performance and 
effectiveness in tandem and individually, including:
     Appelboom et al. (2002) sampled runoff from seven road 
practices in North Carolina and found that roads with continuous berm 
treatment had a 99% reduction in sediment loss compared to roads that 
did not have a continuous berm.
     Aust et al. (2011) evaluated four types of operational 
forest stream crossings at 23 crossings and approaches for total 
dissolved solids, pH, conductivity, temperature, and sediment 
concentration in the Piedmont region of Virginia during initial, 
installation, harvest, and closure stages. The authors found that 
bridge crossings had the least impact on water quality, that the 
installation and harvest phases had the greatest impact on water 
quality, and that BMPs should be followed during all phases.
     Wisconsin DNR (2006) published a BMP manual in 1995 and 
assessed the first ten years of their water quality program. The 
average BMP compliance rate was 83% and BMP effectiveness

[[Page 43497]]

was 99% when the appropriate BMPs were applied and maintained. When 
BMPs were not applied, water quality was affected 71% of the time.
     Pannill et al. (2000) evaluated Maryland BMPs in a paired 
watershed study and, based on TSS, stormflow, stream temperature, and 
macroinvertebrate data, found no significant water quality differences 
between pre-harvest and post-harvest, i.e., proper BMPs will help 
protect water quality, biology, and habitat.
     Vowel (2001) conducted stream bioassessments using a 
stream condition index (SCI) for sites before and after silvicultural 
treatments incorporating Florida BMPs and found no significant 
differences in the SCI. The study concluded that Florida BMPs were 
effective in protecting water quality.
    Cristan et al. (2016) also indicated that, in certain conditions, 
water quality effects can occur even when BMPs are used.
     Maryland DNR (2009) evaluated state BMPs from 2004-2005 on 
75 forest harvested sites using a Maryland-specific BMP implementation 
checklist. Maryland found that 81% of those sites were in compliance 
with state BMPs standards. Maryland also found that BMPs were 77% 
effective in protecting water quality; however, they found that 19% of 
the sites evaluated delivered measurable sediment to waterways.
     Rice (1999) estimated the mean erosion rate from older 
logging roads (installed in the 1950s, maintained to standards of the 
1980s) in the Redwood Creek watershed (northern California) to be 177 
m\3\ km\-1\ from 1980 to 1997, mainly from the road cut banks, but 
noted that changes in forest practice rules (especially proper 
placement of culverts and sizing of culverts) reduced erosion on 
logging roads.
     Bilby et al. (1989) assessed road surface sediment 
production from five roads in two southwestern Washington watersheds 
including two heavily trafficked roads built in the 1950s and three 
haul roads built between 1968 and 1974 and found that sediment entered 
first and second order streams 34% of the time.
     Nolan et al. (2015) examined the effectiveness of BMPs at 
a number of stream crossings in Virginia. The study conducted an audit 
of BMP implementation rates, which it found can often function as 
surrogates for BMP effectiveness. In general, the study found that the 
majority of stream crossings were performing properly, but that 
performance varied. The study also cited Edwards and Williard (2010), 
which ``found only three studies that provided BMP efficiencies with 
regard to sediment loading reductions and reported BMP efficiencies 
ranging from 53%-94%.''
     The USFS evaluated its Pacific Southwest Region BMP 
program from 2008-2010, conducting 2,237 BMP inspections, and found 
that BMP implementation was 91% and effectiveness was 80%, with stream 
water quality impacts at 12% of the sites (USFS, 2013). BMPs for timber 
harvesting, fuels treatments, and vegetation management were effective; 
BMPs for roads, range management, recreation, and mining were not as 
effective, although effectiveness could be increased by imposing 
erosion control plans and wet weather standards.
    EPA also considered other recently-published literature. Below are 
some of the major findings:
     The literature review Assessing the Effectiveness of 
Contemporary Forestry Best Management Practices (BMPs): Focus on Roads 
(NCASI, 2012) reviewed hundreds of studies and found that 
``implementing a suite of contemporary BMPs reduces sediment loads to 
streams by 80% or more relative to uncontrolled forestry operations.'' 
The document further concluded that ``Specific BMPs for roads have been 
tested in controlled studies and proven effective by road inventories 
conducted by forestry agencies in several states. Those inventories 
show that road BMPs are being implemented at high rates and are 
effective in reducing risks to water quality; road drainage structures 
are being disconnected from streams; poor road/stream crossings are 
being identified and corrected; and landslides from forest roads are 
being reduced.''
     The USFS (2012) National Best Management Practices for 
Water Quality Management on National Forest System Lands (Volume 1: 
National Core BMP Technical Guide), provides highly detailed guidance 
on silvicultural BMPs, including those for forest roads. BMP 
effectiveness ratings were 93% (Pacific Southwest Region) and 98% 
(Montana), with North Carolina effectiveness rates showing an increase 
from 73% to 93% between 1992 and 2010. Guidance to standardize BMP 
monitoring protocols is under development.
     Ice et al. (2010) estimated national BMP implementation 
rates at 89%.
     Sugden et al. (2012) found that BMP implementation rates 
in Montana have increased over time, corresponding with a significant 
drop in the number of observed water quality impacts.
    Below are findings from national-scale studies:
     Cristan et al. (2016) concluded that BMPs implementation 
rates and quality are critical to BMP effectiveness for reduction of 
erosion and sediment yield. Important BMP practices for forest roads 
include proper drainage structures, surfacing, erosion control of cut 
and fill slopes, traffic control, and closure. Sediment control 
structures applied to stream crossing approaches can significantly 
reduce runoff and sediment delivery.
     Ice et al. (2010) concluded that the combination of 
effective BMPs and a high rate of BMP implementation helps protect the 
water quality and beneficial uses of streams, lakes, and wetlands in 
forested environments.

VI. Existing BMP-Based Programs and Other EPA Tools

    A broad array of BMP-based programs--including state and federal 
programs and private third-party certification programs--has been 
established to address forest roads in every state with significant 
forestry operations in the country. The following sections outline the 
nation's current landscape of state, federal, and third-party BMP based 
programs designed to control discharges from forest roads, and discuss 
the role of existing EPA tools in addressing stormwater discharges from 
forest roads. As highlighted below, available information indicates 
that these programs are tailored to address regional and local 
differences, that implementation rates are generally high, and that 
meaningful improvements have been and continue to be made in these 
programs over time. EPA did not obtain significant data about tribal 
programs addressing discharges from forest roads, so does not report on 
tribal programs in this section. EPA will seek to learn more about 
efforts to address stormwater discharges from forest roads on tribal 
lands as part of its continuing efforts to gather best practices data 
going forward.

A. State BMP-Based Programs

    Data EPA obtained during the comment period indicates that all 
states with significant forestry operations have developed BMP manuals 
and most states have established forest management programs tailored to 
state-specific conditions (e.g., topography, climate, and industry 
activity) that address runoff from forest roads. The data also 
indicates that BMPs are being implemented at increasing rates across 
the nation. A team of researchers from Virginia Polytechnic Institute 
and State University (Virginia Tech), in consultation with the National 
Association of State Foresters (NASF),

[[Page 43498]]

surveyed all 50 states in 2013 to identify silvicultural activities 
addressed by BMPs, characterize the approaches to BMP implementation 
adopted by each state, determine the extent to which states are 
implementing BMP e[fflig]ectiveness monitoring, and summarize BMP 
implementation rates (NASF, 2015). The survey showed that most states 
have established forestry BMPs designed to protect water quality. 
According to the survey, these programs are a mix of regulatory (11 
states), quasi-regulatory (19 states), and non-regulatory (20 states) 
programs. Those states with regulatory programs generally have some 
form of forest practices law or silvicultural BMP legislation. In 
states with quasi-regulatory programs, state law specifies desired 
outcomes but does not require specific BMPs to achieve that outcome.\5\
---------------------------------------------------------------------------

    \5\ Such programs can include states where BMPs are not 
mandatory but enforcement actions can be taken against polluters.
---------------------------------------------------------------------------

    Existing state programs vary because they are designed to address 
state and site-specific factors. Prior assessments of state forestry 
BMP programs have found similar, generally consistent information.\6\ 
\7\ The following number of states have established forest road 
specific BMPs (Table 1).
---------------------------------------------------------------------------

    \6\ See 80 FR 69657-69658 (Nov. 10, 2015). Characterizations of 
state forestry BMP programs differ in some ways because of the way 
reviewers categorize the programs, aspects of the programs they 
review, different interpretations of program elements, and the fact 
that state forestry BMP programs have evolved and continue to evolve 
over time.
    \7\ Endicott, 2008. See Section 4 and Tables 4-1 and 4-2.

 Table 1--States With Forest Road BMP Programs Based on Endicott (2008)
------------------------------------------------------------------------
                                                             Number of
               Category of forest road BMP                    states
------------------------------------------------------------------------
Construction............................................              44
Drainage................................................              41
Location/Spacing........................................              38
Maintenance.............................................              40
Road Closure............................................              24
Stabilization/Soils/Slope...............................              32
Stream Crossings........................................              40
SMZs/Bank Stabilization/Buffer Strips...................              36
Wet Weather Use.........................................              10
Winter Operations.......................................              10
Training/Technical Assistance...........................              23
Implementation/Effectiveness Monitoring.................              32
Compliance/Enforcement..................................              30
------------------------------------------------------------------------

1. Existing State Programs Are Tailored To Address State and Site-
Specific Factors
    One of the primary mechanisms for addressing water quality impacts 
of forest roads is individual states' forest practices polices, which 
generally establish standards for the design, operation and maintenance 
of forest roads applicable to conditions in their state. State forest 
road programs vary to some degree in their structure, requirements, and 
administration. Differences are based on legal, and socioeconomic 
factors as well as variations in climate, soils, topography, and 
aquatic biota. State programs generally establish both guiding 
principles and specific management practices that must be applied and 
adapted to a broad range of settings and conditions. Site-specific 
flexibility is important because no single set of requirements will be 
effective across the country. As EPA stated in its November 10, 2015 
notice, ``[t]he diversity of the forest road networks, the different 
classes of roads, the different local physical conditions, and the 
broad range of road conditions and uses indicate the importance of site 
specific BMP selection and implementation to protect water quality'' 
(80 FR 69656). For example, commenters correctly pointed out that 
Florida's forest road BMPs need not recommend or discuss full-bench 
road construction and end hauling techniques, as Oregon's rules do, 
because Florida does not have landslide-prone terrain, while Oregon has 
steep terrain with the potential for landslides, where such 
construction and end hauling techniques would be appropriate (EPA-HQ-
OW-2015-0668-0089).
2. State Programs Show High Implementation Rates
    Data from the 2013 NASF survey indicated that both forestry and 
forest road BMPs are implemented broadly. BMP implementation surveys in 
32 states (i.e., those with significant forest management activity) 
between 2005 and 2013 showed an average forestry BMP implementation 
rate of 91% (NASF, 2015). Nationally, the survey suggests that 
implementation rates for forest road BMPs averaged 91.5% and stream 
crossing BMPs averaged 86.7% (NASF, 2015). The 2012 Southern Region 
Report published by the Southern Group of State Foresters (SGSF) found 
forest road BMP implementation rates for 11 states \8\ range from 78-
99%, with an average of 88%. In the SGSF report, stream crossing BMP 
implementation rates ranged from 72-98% and averaged 89% (SGSF BMP 
Report, 2012).
---------------------------------------------------------------------------

    \8\ Alabama, Arkansas, Florida, Georgia, Mississippi, North 
Carolina, Oklahoma, South Carolina, Tennessee, Texas, and Virginia.
---------------------------------------------------------------------------

    The NASF survey also indicated that forest road BMP implementation 
rates do not vary significantly regardless of whether the state program 
is regulatory, quasi-regulatory, or non-regulatory. The NASF survey 
indicated that implementation of forest roads BMPs in 8 regulatory 
reporting states averages 93.9%, while the implementation rates in the 
11 quasi-regulatory reporting states and 13 non-regulatory reporting 
states averages 90.6% and 90.5%, respectively (NASF, 2015).
    Plus, BMP implementation rates have improved and continue to 
improve over time. For example, from 2008--2012, the implementation 
rates for all forestry BMPs (including forest road and stream crossing 
BMPs) trended upward in the SGSF report. This included forest road BMP 
implementation rates and stream crossings BMP implementation rates, 
which increased from 87 to 90%, and from 85 to 89%, respectively (SGSF 
BMP Report, 2012).
    In addition to state forest road BMP programs, several efforts have 
emerged over the past 10 years to improve monitoring of BMP programs. 
Regional groups have undertaken efforts to promote consistent and 
comparable forestry BMP program monitoring data. The SGSF and the 
Northeastern Area Association of State Foresters (NAASF) have developed 
regional BMP monitoring protocols that states in those regions are 
using.
    SGSF developed Silviculture Best Management Practices 
Implementation Monitoring, A Framework for State Forestry Agencies 
(2007) to improve and maximize the integrity of BMP implementation 
monitoring in southern states (SGSF Regional BMP Framework Protocol, 
2007). The framework, which is implemented by 13 southern states, 
Puerto Rico, and the U.S. Virgin Islands, is designed to provide 
guidance for monitoring forestry BMP implementation that results in 
data that are statistically sound, objective, and promote analytical 
consistency among states. The framework addresses monitoring frequency, 
site selection, practices to be evaluated, the basis for practice 
evaluation and reporting, scoring methodology, risk assessment, and 
follow-up actions.
    Similar to the SGSF BMP monitoring framework, the USFS Northeastern 
Area State and Private Forestry and the Northeastern Area Association 
of State Foresters--Water Resources Committee have developed the 
Forestry BMP Protocol Project. The BMP Protocol is a

[[Page 43499]]

standard method for monitoring the use and effectiveness of BMPs 
commonly used in timber harvesting. The BMP Protocol, which is 
available to 20 states, serves three functions: (1) Data collection, 
(2) data analysis, and (3) report generation. It collects data using a 
branched question set designed to address those areas of the timber 
harvest with the greatest potential to impact water resources 
(including haul roads and water crossings). The protocol was developed 
to document the use and effectiveness of BMPs in protecting water 
resources during forest harvesting operations; document the degree of 
compliance with the CWA, as well as the Coastal Zone Management Act and 
various state laws and regulations; assess water resource protection 
based on the effectiveness of a collective set of BMPs; increase 
credibility through the measurement of results; respond to public 
concerns regarding the potential effects of timber harvesting based on 
measured evidence; and identify opportunities for improvement in water 
resource protection by identifying causes of BMP failure. Both a Desk 
Reference and Field Guide have been developed for the monitoring 
protocol (BMP Manual Desk Reference, 2007; BMP Field Guide, 2007).
    Other factors are also facilitating the increasing rate of BMP 
implementation. For example, third-party certification programs, as 
discussed in detail in section VI.C of this document, all require BMP 
implementation and third-party audits to verify that timber companies 
conform to state standards. Forest certification programs have made 
important contributions to improved BMP implementation through logger 
training, landowner outreach, and water quality requirements. Other 
examples are the logger training and certification programs established 
by states and third-party programs, such as the SFI Logger Training and 
Education (2015) program, to ensure loggers are educated about the use 
and maintenance of appropriate forest road BMPs. Training is 
particularly important given the site-specific customization BMPs 
require. The best way to ensure optimal BMP selection and installation 
is through localized knowledge of climate, soils, forestry operations, 
and other factors, in combination with state-specific BMPs. Some 
commenters noted that the Forest Resources Association reports having 
trained more than 150,000 logging professionals since the inception of 
the forest certification program (EPA-HQ-OW-2015-0668-0089). For fiscal 
year 2015, West Virginia noted that 1,454 loggers received 
certification to supervise logging operations and assure BMPs were 
applied (EPA-HQ-OW-2015-0668-0075). Also, as one commenter noted, 
effective outreach and training programs have served to foster a 
culture of high BMP implementation rates such that BMPs have largely 
been institutionalized in the forestry community.
3. State Programs Continue To Evolve and Improve
    States frequently revise their forest roads management guidance/
regulations. States with significant forestry operations have 
mechanisms in place to evaluate the effectiveness of forestry BMPs and 
use monitoring and research results to revise these practices when 
necessary (typically by government appointed forestry boards, forestry 
commissions, or a mix of agencies, councils, or departments). For 
example, California Department of Forestry and Fire Protection revised 
its Forest Practice Rules in 2015 to better manage drainage and erosion 
from logging roads (EPA-HQ-OW-2015-0668-0055); Wisconsin DNR-Division 
of Forestry revised its Forest Management Guidelines in 2011,\9\ 
including updating forestry BMPs for water quality; and the Oregon 
Board of Forestry increased the riparian zone buffer width for fish-
bearing streams in 2015 (Oregon Riparian Rule, 2015). States, federal 
agencies and various stakeholder groups continue to enhance BMP 
prescriptions and identify the site-specific factors that influence 
their effectiveness. For example, industry commenters identified 36 
states that have revised their forest road BMPs within the last ten 
years (EPA-HQ-OW-2015-0668-0089), and according to a recent state 
survey conducted by the National Association of State Foresters, 31 
states (62%) have updated their forest roads management guidance/
regulations since 2006.\10\ EPA's own analysis also indicates that many 
states have revised their programs, with some being revised as recently 
as 2016 (State Program Summary, 2016).
---------------------------------------------------------------------------

    \9\ https://dnr.wi.gov/topic/forestmanagement/guidelines.html.
    \10\ https://www.stateforesters.org/action-issues-and-policy/state-forestry-BMPs-map-o-o.
---------------------------------------------------------------------------

B. Federal BMP-Based Programs

    At the federal level, the USFS and the BLM have established 
programs to manage stormwater discharges from forest roads on federal 
lands. These agencies manage large tracts of forested lands, including 
lands that are actively being used for road building, road maintenance, 
logging operations, public and recreational use or other activities, 
and generally demonstrate sound environmental stewardship in managing 
these lands.
1. Summary of U.S. Forest Service Programs
    The 193 million acres (780,000 km\2\) of public land that are 
managed as national forests and grasslands are collectively known as 
the National Forest System. These lands are located in 44 states, 
Puerto Rico, and the Virgin Islands and comprise about 9% of the total 
land area in the U.S. The USFS manages approximately 20% of the 
Nation's forested area and nearly 10% of the Nation's rangelands (USFS 
Strategic Plan FY: 2015-2020). The lands are organized into 154 
National Forests and 20 National Grasslands. The mission of the 
National Forest System is to manage the national forests and grasslands 
to meet the Agency's sustainable multiple-use mandate.
    The USFS uses several tools and strategies, such as the Legacy 
Roads and Trails program, Watershed Condition Framework, and the 
National Best Management Practices Program, in addition to local 
programs, to maintain and improve watershed health and manage 
discharges from forest roads.
    The Legacy Roads and Trails program assists the USFS in identifying 
legacy roads in national forests and grasslands. USFS targets projects 
that will minimize the discharge of stormwater by decommissioning, 
maintaining, or upgrading various roads. From 2009-2015, the USFS 
decommissioned 5,504 miles of National Forest System Roads and an 
additional 6,714 miles of unauthorized roads; reconstructed 13,413 
miles of roads; and maintained 57,333 miles of roads per year during 
that period.
    The USFS Watershed Condition Framework helps the USFS to assess 
watershed health in national forests and grasslands, identify and 
implement protective measures, and conduct ongoing watershed 
monitoring. Watershed conditions are categorized into three discrete 
categories or classes that reflect the health of the watershed. One 
primary emphasis of the watershed assessment is indicators that 
directly or indirectly impact soil and hydrologic functions as well as 
riparian and aquatic ecosystems. Initial watershed condition framework 
assessments for all watersheds on USFS lands were completed in 
2011.\11\
---------------------------------------------------------------------------

    \11\ https://www.fs.fed.us/biology/watershed/condition_framework.html.
---------------------------------------------------------------------------

    In 2012 the USFS also initiated and began to implement a National 
BMP

[[Page 43500]]

program integrating water resource protection into landscape management 
activities. The National BMP program is designed to improve agency 
performance, accountability, consistency, and efficiency in protecting 
water quality. The program consists of National Core BMPs, standardized 
monitoring protocols to evaluate BMP implementation and effectiveness 
of the National Core BMPs, and a data management system to store and 
analyze the resulting monitoring data. National Core BMPs address 11 
subject areas affecting water quality. One of those subject areas is 
road management activity, which includes BMPs for travel management 
planning and analysis, road location and design, road construction, and 
stream crossings (USFS, 2012). The National BMP based program enables 
the USFS to document compliance with the management of nonpoint source 
pollution at local, regional, and national scales as well as address 
the 2012 land management planning rule requirement for national BMPs at 
36 CFR 219.8(a)(4).
    The USFS monitors road management BMP implementation and its 
effectiveness at protecting water, aquatic, or riparian resources 
through nine evaluation categories and/or time periods, some of which 
include: Construction and reconstruction of USFS system roads and/or 
waterbody crossings; after construction or reconstruction has been 
completed; long-term management and maintenance of USFS system roads; 
decommissioned roads after decommissioning activities have been 
completed; and roads, parking areas, and snow storage areas during snow 
removal and storage activities.
    The USFS has also developed a National Core BMP Technical Guide 
intended to improve USFS accountability and performance in managing 
water quality programs. Many of the core BMPs in the National Core BMP 
Technical Guide address water quality. The Technical Guide also 
provides administrative directives to allow for the use of state, 
tribal, and local requirements and information to develop site-specific 
BMPs where needed (USFS, 2012). The USFS is currently developing a 
second volume of the National Core BMP Technical Guide that will 
provide standardized protocols for monitoring BMP implementation and 
effectiveness across all USFS lands.
    Further, USFS has developed a suite of tools to identify and 
prioritize road segments at risk of impacting water quality. These 
tools operate at scales of detail ranging from using corporate road 
databases and digital elevation data to using detailed GPS surveys. 
These tools apply in watershed sediment load reduction plans for waters 
listed as impaired under the CWA and in forest restoration projects 
under the Collaborative Forest Landscape Restoration Program in the 
states of Idaho, Montana, and California. For example, the Geomorphic 
Road Analysis and Inventory Package (GRAIP) tool includes methods to 
inventory roads and analyze the inventory for surface erosion, and 
risks for gullies, landslides, and stream crossing failures. This tool 
can be used in combination with other field observations to assess 
forest roads.
    As an example of implementation of the USFS's BMP programs, the 
USFS evaluated its Pacific Southwest Region BMP program from 2008-2010 
through 2,237 BMP inspections. It found that BMP implementation was 91% 
and effectiveness was 80%, with water quality affected at streams on 
12% of sites. The USFS is continually improving and updating its 
programs and tools as accomplishments are monitored and verified. In 
2013, the USFS completed an interim National BMP monitoring database 
for the National BMP program. The USFS expects to integrate this 
interim database into an enterprise data management system in the 
future which will extend reporting and analysis capabilities of the 
database.
    In fiscal year 2014, 97 USFS administrative units completed a total 
of 600 BMP evaluations as part of implementing in the National BMP 
monitoring program. As discussed above, the USFS national core BMPs 
address 11 subject areas that potentially could affect water quality, 
including ``road management activities.'' Nine monitoring protocols 
have been developed for the road management activity BMPs. At least 1 
BMP evaluation was completed on 87% of the USFS administrative units; 
over 100 evaluations were conducted for road management activity BMPs. 
Of the 600 total evaluations, 94% included implementation assessments, 
90% included effectiveness assessments, and 85% included both 
implementation and effectiveness assessments.
    Overall, 61% of the BMP implementation evaluations were rated as 
``fully implemented'' or ``mostly implemented.'' In addition, 65% of 
the BMP effectiveness evaluations were rated as ``effective'' or 
``mostly effective.'' For sites where BMP implementation and 
effectiveness were both evaluated, 56% had composite ratings of 
``excellent'' or ``good.'' For road management activities, 
approximately 70% of the evaluations identified BMPs that were fully or 
mostly implemented. With regard to road management BMP effectiveness, 
approximately 50% of the completed evaluations were found to be 
effective or mostly effective. In the study the USFS acknowledges that 
these data show room for improvement in BMP implementation and 
effectiveness but observes that prior to development of the National 
BMP Program, it was impossible to report on BMP implementation and 
effectiveness on a national scale in a coherent, understandable, and 
useful way.
    In December 2015, the USFS published the National Best Management 
Practices Monitoring Summary Report for the two-year BMP phase-in 
period of fiscal years 2013 and 2014 following the launch of the 2012 
National Best Management Practices program. That report summarizes the 
national results of the two year phase-in period of national BMP 
monitoring. The report demonstrates the capabilities of a consistent 
nationwide monitoring program to document BMP performance (USFS, 2015). 
In addition, as part of the Watershed Condition Framework, the USFS is 
currently undertaking a five year re-assessment to assess changed 
conditions of USFS watersheds.
    For example, USFS is using outputs from the GRAIP tool, mentioned 
previously, in combination with associated field observations to assess 
the effectiveness of road decommissioning in Idaho, Montana (Cissel et 
al., 2014a), Oregon, Utah, and Washington. BMPs implemented as part of 
the decommissioning efforts resulted in a 79% reduction in fine 
sediment delivery to streams (Cissel et al., 2014b).
    The USFS implements best practices to control stormwater from 
forest roads on a program-wide scale in a number of ways, as well as 
ensuring that specific projects are implemented properly. Where a USFS 
road crew is in place, the agency performs maintenance and 
construction/reconstruction to the extent the law allows. BMPs are 
followed according to USFS policy, incorporating any national, 
regional, and local level BMPs. Crews work closely with local resource 
specialists to ensure work is being performed according to BMPs. When a 
project is awarded under a contract, clauses, provisions, mitigation 
measures, and BMPs are incorporated into the plans, specifications, and 
contract documents. For example, some contract provisions require the 
contractor to preserve, protect, and minimize the impacts from soil 
erosion to streams, lakes, and

[[Page 43501]]

reservoirs.\12\ A Contracting Officer or their certified designees 
monitor work performed by the contractor to ensure work compliance with 
the terms and conditions set forth in the contract.
---------------------------------------------------------------------------

    \12\ See BLM. (2011). Contract for the Sale of Timber and Other 
Wood Products Lump Sum Sale.
---------------------------------------------------------------------------

    The USFS is a recognized leader in establishing road crossing 
techniques that provide for aquatic organism passage, or the ability 
for fish and other aquatic life to move up or downstream under roads. 
In 2005, the USFS created the National Inventory and Assessment 
Procedure to evaluate the effectiveness of current and remediated fish 
passages (USFS, 2005). Over 1,600 miles of habitat were restored in 
fiscal years 2011-2013 by aquatic organism passage projects funded 
through the USFS Legacy Roads and Trails Restoration program among 
others (USFS, 2014).
2. Summary of Bureau of Land Management Programs
    BLM manages approximately 246 million acres of public lands (BLM, 
2015). Most BLM lands are concentrated in 11 western states with 
scattered tracts in the various eastern states. Of the 246 million 
acres, approximately 50 million acres are forest or woodlands where 
approximately 6-7 million acres are managed for sustainable timber 
harvests. These areas are generally mesic sites with annual average 
precipitation that usually exceeds 15 inches per year. Traditional 
timber harvesting on BLM property occurs primarily in northern 
California, Colorado, Idaho, Montana, Oregon, and Wyoming, with minimal 
harvest occurring in Alaska, Arizona, Nevada, New Mexico, and Utah. BLM 
uses several tools including land use plans, Memoranda of Understanding 
(``MOU'') with states and other federal agencies, timber sale 
contracts, and training to ensure protection of water resources.
    Most BLM lands are managed pursuant to the Federal Land Policy and 
Management Act of 1976 (FLPMA), at 43. U.S.C. 1712, which requires 
public lands to be managed under the principles of multiple-use and 
sustained yield. BLM's land use planning regulations at 43 CFR part 
1600 establish a land use planning system for BLM-managed public lands. 
Similar to the USFS, a full suite of activities are authorized and 
managed on BLM forests and woodlands, including timber harvesting, 
hazardous fuel reduction treatments, recreation, fish and wildlife 
conservation, oil and gas activities, and grazing. Authorized uses in 
forests and woodlands such as timber harvesting often include road 
construction and maintenance \13\ which are broadly governed by 
policies, standards, and right-of-way agreements that ensure proper 
design and upkeep.\14\
---------------------------------------------------------------------------

    \13\ Bureau of Land Management estimates that as of 2014 there 
were approximately 72,300 miles of roads on Bureau of Land 
Management lands (Public Land Statistics Table 6.2, pg. 246). Only a 
subset of these roads are located in forested environments that 
would have the potential to contribute to stormwater runoff (Bureau 
of Land Management Supplemental Response 3/29/16).
    \14\ https://www.blm.gov/wo/st/en/prog/more/forests_and_woodland.html.
---------------------------------------------------------------------------

    One source of guidance for proper development of BLM land use plans 
is BLM's Land Use Planning Handbook. The Handbook provides broad agency 
direction for BLM to use BMPs to meet the standards and goals of the 
CWA and address various protection measures to mitigate impacts to 
human health concerns, ecosystem health, riparian areas, and overall 
watershed conditions, and to meet state and local water quality 
requirements (BLM, 2005).
    BLM state offices enter into interagency MOUs with state and other 
federal agencies designed to ensure that they cooperatively meet state 
and federal BMPs and water quality rules and regulations related to 
point and nonpoint source water pollution from BLM managed lands.\15\ 
These MOUs clarify such issues as jurisdictional and statutory 
authorities, monitoring responsibilities, implementing effective BMPs, 
prioritizing restoration activities, and developing strategies to meet 
water quality standards. The Idaho Nonpoint Source Management Plan 
provides one example of such an MOU (Idaho DEQ, 2015). In addition, 
several components of BLM state and national level manuals apply to 
ground-disturbing activities and provide for consistent implementation 
of BMPs.\16\
---------------------------------------------------------------------------

    \15\ An example of an interagency MOU between Bureau of Land 
Management, other federal agencies and the Idaho Department of 
Environmental Quality, can be found at https://www.deq.idaho.gov/media/1041346-nps_program_implementation_mou_2013.pdf.
    \16\ Bureau of Land Management Manual 9113 (Roads), 9115 
(Primitive Roads including BMPs from the Surface Operating Standards 
and Guidelines for Oil and Gas Exploration and Development), 7240 
(Water Quality), Manual 5000 Forest Management (pertaining to timber 
sale contracts and specific contract provisions to apply to forest 
roads to address water quality protection).
---------------------------------------------------------------------------

    Finally, all BLM timber sales contracts contain standard contract 
requirements that expressly require that the purchaser must comply with 
all applicable state and federal laws and regulations pertaining to 
water quality. Often, they include special provisions deemed necessary 
(e.g., restrictions on wet weather operations, conditions addressing 
Endangered Species Act requirements, soil and aquatic protection 
requirements, etc.).\17\ Individual BLM offices consistently add 
special provisions to timber sales as well as other ground disturbing 
activity contracts to ensure effective BMP implementation. Appropriate 
BMPs are identified at the Resource Management Plan level, analyzed 
during site-specific NEPA review process, and implemented in various 
ways such as direct performance by BLM crews or through a timber sale 
contract.
---------------------------------------------------------------------------

    \17\ ``Bureau of Land Management Standard Timber Sale Contract 
Language,'' Bureau of Land Management Form 5450-004, Sections 26, 
27, & 28.
---------------------------------------------------------------------------

    BLM also provides training for their specialists in all aspects of 
resource management including engineering (to include roads and 
facilities), forest management, fish and wildlife management, and 
hydrology. Training curricula include: Review of existing and new state 
and federal regulations, manuals, handbooks, and policies including 
compliance with BMPs; preparing and administering contracts; review of 
interagency agreements or MOUs; review of updates on monitoring, 
evaluating, and reporting protocols and agency monitoring databases; 
review of Resource Management Plans and amendments; and conducting 
National Environmental Policy Act reviews.
    BLM incorporates BMPs into land use plans that include management 
of forest roads. The recently released western Oregon Proposed Resource 
Management Plan/Final Environmental Impact Statement, Appendix J 
provides one example of such a plan (BLM RMPWO Vol. 3 Appendix J, 
2016). The BMPs for the western Oregon Proposed Resource Management 
Plan address various anticipated resource management actions including: 
Road and landing maintenance and construction, timber harvest 
activities, silviculture activities, surface source water for drinking 
water, and recreation management. These BMPs were developed in 
coordination with Oregon Department of Environmental Quality to 
cooperatively meet state and federal water quality regulations. 
Additional BMPs could be required for a particular project depending on 
site-specific needs and subsequent implementation and effectiveness 
monitoring. BLM field offices review the land use plan BMPs and select 
and apply the appropriate and applicable BMPs for a particular project. 
Those BMPs are incorporated into on-the-ground operations like timber 
sales, road maintenance, road construction, and riparian restoration 
projects.

[[Page 43502]]

    Although the BLM does not have a national BMP monitoring database 
like the USFS, it works closely with a number of state and federal 
agencies to annually monitor, evaluate, and report BMP compliance and 
effectiveness. One example demonstrating the success of resource 
management plans to protect water quality is the Northwest Forest Plan 
(NWFP). Approximately 2.5 million acres of forested BLM land falls 
within the area covered by the NWFP and those acres have been managed 
consistent with the NWFP standards and guidelines. All of those 
standards and guidelines were incorporated into the 1995 western Oregon 
resource management plans.
    The Aquatic Conservation Strategy is an important element of the 
NWFP, which incorporates into the resource management plans the 
implementation of a riparian reserve system (e.g., buffers) along 
streams as well as reducing road densities. Since 1995, western Oregon 
BLM Districts have decommissioned or obliterated over 883 miles of 
roads.
    As mentioned above, BLM has released a proposed resource management 
plan and a final environmental impact statement for western Oregon BLM 
Districts to revise the 1995 resource management plans. Under the 
proposed resource management plan, the riparian reserve system, along 
with a late successional forest reserve system, would increase from 57% 
following the 1995 resource management plan to 64% following new 
guidelines. BLM has worked closely with over 20 cooperating agencies 
including U.S. Fish and Wildlife Service, National Marine Fisheries 
Service, and EPA to continue a comprehensive and regional strategy to 
maintain and improve aquatic resources in alignment with the 
overarching ecosystem principles and intent of the Aquatic Conservation 
Strategy of the NWFP under the new RMP.
    The recently released ``Northwest Forest Plan Interagency Regional 
Monitoring: 20 Year Report, Status and Trends of Watershed Condition'' 
report summarizes the results of the twenty year interagency effort to 
implement an array of water quality protective measures in the Aquatic 
Conservation Strategy to maintain watershed health in that region 
(Northwest Forest Plan, 2015). The NWFP Aquatic Conservation Strategy 
consists of four components: Riparian reserves, key watersheds, 
watershed analysis, and watershed restoration. Once watershed 
conditions were evaluated and resource needs were identified, multiple 
agencies, as well as public stakeholders, partnered to complete 
millions-of-dollars' worth of watershed restoration work include: 
Providing fish passages through culvert removals, replacements, or 
bridge construction; obliterating, closing, or relocating streamside 
roads; vegetating disturbed areas; reducing hazardous fuel loads; 
upgrading road surfaces to reduce sediment runoff; and removing dams. 
Implementation of these four components has resulted in improved 
watershed conditions in many watersheds.
    The recently released monitoring report's objective was to evaluate 
whether the NWFP Aquatic Conservation Strategy is achieving the goal of 
maintaining and restoring the condition of watersheds throughout the 
region covered by the NWFP. The report evaluated two subject areas: 
Upslope riparian areas for all watersheds with at least 5% federal 
ownership, and in-channel stream data (e.g., temperature, sediment, and 
macroinvertebrates). The report compares the effectiveness of 
management practices under the aquatic conservation strategy direction 
for two periods: 1993 and 2012 for upslope riparian assessment, and 
rotational sampling between 2002-2009 and 2010-2013 for in-channel 
stream assessment. These monitoring data were used to detect trends and 
evaluate stream and upslope riparian conditions for 1,974 watersheds in 
the Pacific Northwest.
    The report signified that there has been a slight positive shift in 
upslope riparian condition. Sediment scores were generally very high, 
indicating a low risk of roads delivering sediment to streams. Sharp 
declines in assessment scores were mainly driven by large wildfires, 
and were offset by moderate, broad-scale improvements in vegetation, 
and focused improvements related to road decommissioning.
    BLM also uses technical tools for evaluation, planning, and 
assessment of water quality. BLM is applying the USFS GRAIP tool, as 
well as others, in western Oregon watersheds to assess the 
effectiveness of road decommissioning and in sediment load reduction 
plans for waters listed as impaired under the CWA. These tools will 
also be used to prioritize the backlog of deferred maintenance needs 
that are later identified in the western Oregon Final Environmental 
Impact Statement, Chapter 3, Trails and Travel Management.
    Outside of western Oregon, BLM is involved with various state, 
regional, and national water quality monitoring efforts to assess 
management effectiveness including indirect effectiveness of BMPs 
related to forest management and roads. For example, BLM cooperates 
with the Montana State Environmental Quality Council to monitor how 
forest practices are affecting watersheds in Montana. Montana conducts 
BMP field reviews on state, federal, and private industrial and non-
industrial forest lands to monitor BMP implementation and 
effectiveness. Montana's 2014 BMP review concluded that 96% of BMP 
practices were effective on federal lands (Montana DNRC, 2014).
    BLM has conducted a number of successful watershed restoration 
efforts to improve water quality on BLM lands. One example is the BLM 
Headwaters Forest Reserve Road Restoration Project in California. Since 
2000, BLM has worked with the Pacific Coast Fish, Wildlife and Wetlands 
Restoration Association to decommission and restore 26 miles of old 
logging roads throughout headwaters. An additional 5 miles of 
decommissioning is planned for the next several years.\18\
---------------------------------------------------------------------------

    \18\ https://blm.gov/ca/st/en/prog/nlcs/Headwaters_ForestReserve/restoration.html.
---------------------------------------------------------------------------

3. Federal Programs Are Evolving and Improving
    Both the USFS and BLM have improved their programs that address 
water quality and stormwater from forest roads over the last several 
years. As noted above, the USFS launched a new National BMP program in 
2012 and is currently monitoring the program for results. In addition, 
the USFS has enhanced its Road Preconstruction Handbook on Design \19\ 
as well as the Transportation Structures Handbook on Hydraulics and 
Watershed Protection \20\ to include design considerations for the 
construction and reconstruction of forest roads which minimize road and 
drainage impacts to the watershed. USFS Technology and Development 
Centers have created a number of publications to assist designers when 
addressing road/water interactions.\21\ BLM has taken extensive efforts 
to improve its protection and restoration efforts of watersheds by 
addressing key resource areas and improving resource management plans. 
Even with limited resources, federal programs are using new technology 
to target highest priority problems in watersheds to mitigate water 
quality impacts and monitor watershed health and project effectiveness. 
Improved resource management plans and technology will

[[Page 43503]]

likely continue to evolve and lead to greater improvements.
---------------------------------------------------------------------------

    \19\ See FSH 7709.56 Chapter 40 at https://www.fs.fed.us/dirindexhome/dughtml/fsh_1.html.
    \20\ See FSH 7709.56b Chapter 60 at https://www.fs.fed.us/dirindexhome/dughtml/fsh_1.html.
    \21\ https://www.fs.fed.us/eng/pubs/.
---------------------------------------------------------------------------

C. Third-Party Certification BMP-Based Programs

    In addition to state and federal forest road BMP programs, 
participation in third party forest certification programs has been 
increasing rapidly in the U.S. Forest management certification arose to 
foster an improved stewardship of working forestlands. Programs such as 
certifications, which provide information and disclosure to consumers, 
can generate significant beneficial impacts on the environment while 
imposing fewer costs on industries and producers than direct regulatory 
programs.\22\ Requirements to disclose information to citizens and 
consumers can lead to beneficial change without specific behavioral 
mandates. Certification provides a market incentive to encourage 
landowner commitment to sustainable forest management. It also offers a 
stamp of approval for forest management practices that meet standards 
considered to be environmentally appropriate, socially beneficial, and 
economically viable.
---------------------------------------------------------------------------

    \22\ From Thaler, R., & Sustein, C. (2009). Nudge.
---------------------------------------------------------------------------

    The three largest forestry certification programs in the U.S. are 
the Forest Stewardship Council (FSC), the Sustainable Forestry 
Initiative (SFI), and the American Tree Farm System (ATFS). These 
programs promote higher rates of BMP implementation by mandating 
compliance with applicable state and local laws and applicable BMPs, 
whether regulatory or voluntary. They promote training/education 
(including continuing education) and the use of trained loggers, 
promote monitoring of forestry BMP implementation, and include 
mechanisms for addressing instances where BMP nonconformance is 
observed. FSC requires expanded protection for waterbodies where it 
deems state programs or existing guidelines insufficient to protect 
water quality.
    EPA received comments from state forestry agencies highlighting the 
large areas of state forested land under one of the third-party 
certifications identified above. For example, the Idaho Department of 
Lands notes that over 1.5 million acres of forest lands in Idaho are 
privately held or owned and managed by industries that maintain third-
party certification through SFI, FSC or ATFS (EPA-HQ-OW-2015-0668-
0072). Maine has almost 8 million acres of forest land which is third-
party certified (EPA-HQ-OW-2015-0668-0058); and in Mississippi almost 
470,000 acres of public forest land is certified through the ATFS and 
audited annually to ensure proper BMP implementation (EPA-HQ-OW-2015-
0668-0081).
    The discussion below provides a brief description of the three 
major programs in the U.S., focusing on how they promote management 
practices for mitigating water quality impacts resulting from 
stormwater discharges from forest roads.
1. Forest Stewardship Council (FSC)
    FSC is an independent group with open membership that first 
convened in 1993 to improve forest practices internationally through a 
voluntary, market-based approach. FSC's program places an emphasis on 
whole-forest conservation, including protecting water resources from 
effects of stormwater discharges from forest roads. FSC is the only 
standard that prohibits the use of certain pesticides and herbicides in 
the timber industry and prohibits large clearcuts where they threaten 
the ecological integrity of the forest.
    FSC's program includes a series of overarching principles and more 
specific performance criteria. An example forest management 
certification criterion is Forest Management Standard Criterion C6.5, 
which states, ``[w]ritten guidelines shall be prepared and implemented 
to: control erosion; minimize forest damage during harvesting, road 
construction, and all other mechanical disturbances; and protect water 
resources.'' One ``indicator'' of this criterion provides that 
``[f]orest operations meet or exceed BMPs that address components of 
the Criterion where the operation takes place.'' Another provides,

[t]he transportation system, including design and placement of 
permanent and temporary haul roads, skid trails, recreational 
trails, water crossings and landings, is designed, constructed, 
maintained, and/or reconstructed to reduce short and long-term 
environmental impacts, habitat fragmentation, soil and water 
disturbance and cumulative adverse effects, while allowing for 
customary uses and use rights. This includes: access to all roads 
and trails (temporary and permanent), including recreational trails, 
and off-road travel, is controlled, as possible, to minimize 
ecological impacts; road density is minimized; erosion is minimized; 
sediment discharge to streams is minimized; there is free upstream 
and downstream passage for aquatic organisms; impacts of 
transportation systems on wildlife habitat and migration corridors 
are minimized; area converted to roads, landings and skid trails is 
minimized; habitat fragmentation is minimized; unneeded roads are 
closed and rehabilitated.

    Yet another indicator requires that, ``[a] monitoring program is in 
place to assess the condition and environmental impacts of the forest-
road system.'' Certifiers are independent of FSC itself and the 
companies they audit.
2. Sustainable Forestry Initiative (SFI)
    SFI is an independent, nonprofit organization that is responsible 
for maintaining, overseeing, and improving the SFI certification 
program. Across the U.S. and Canada, more than 280 million acres are 
certified to the SFI Forest Management Standard and additional acres 
are influenced by SFI Fiber Sourcing. SFI administers standards that 
address forest sustainability broadly and water quality specifically. 
The SFI 2015-2019 Forest Management Standard applies to any 
participating organization in the U.S. or Canada that owns or has 
management authority for forestlands and consists of measures designed 
to protect water quality, biodiversity, wildlife habitat, species at 
risk, and forests with exceptional conservation value. The measures 
require developing a program for certification and compliance that 
include monitoring BMPs during all phases of forestry activities, 
mapping of water resources, and recordkeeping. For example, Objective 3 
in the Standard addresses ``Protection and Maintenance of Water 
Resources--To protect the water quality of rivers, streams, lakes, 
wetlands, and other water bodies through meeting or exceeding best 
management practices.'' Under Objective 3, Performance Measure 3.1 
provides that ``Program Participants shall meet or exceed all 
applicable federal, provincial, state and local water quality laws, and 
meet or exceed best management practices developed under Canadian or 
EPA-approved water quality programs.'' Performance Measure 3.2 further 
provides, ``Program Participants shall implement water, wetland, and 
riparian protection measures based on soil type, terrain, vegetation, 
ecological function, harvesting system, state (BMPs), provincial 
guidelines and other applicable factors.'' Objective 11 addresses 
``Training and Education'' and Performance Measure 11.1 provides that 
``Program Participants shall require appropriate training of personnel 
and contractors so that they are competent to fulfill their 
responsibilities under the SFI 2015-2019 Forest Management Standard.''
    SFI noted in its comments that 95% of the fiber delivered to SFI 
Program Participant mills is delivered by harvesting professionals who 
have been trained in sustainable forestry practices (EPA-HQ-OW-2015-
0668-0099). Additional Forest Management

[[Page 43504]]

Standard Objectives address Forest Management Planning (Objective 1) 
and Legal and Regulatory Compliance (Objective 9).
3. American Tree Farm System (ATFS)
    ATFS is a program of the American Forest Foundation, and has a 
forest certification standard that applies to small landowners in the 
U.S. In 2009, ATFS had certified more than 25 million acres of 
privately owned forestland managed by over 90,000 family forest 
landowners. To become certified, ATFS landowners must own at least 10 
acres of forestland and implement a written forest management plan; and 
follow ATFS and AFF's 2015-2020 Standards of Sustainability for Forest 
Certification for Private Forestlands. Tree farms are inspected and 
certified to assure proper forest management that includes the 
conservation of soil, water and wildlife. Standard 4: Air, Water, and 
Soil Protection provides that ``[f]orest-management practices maintain 
or enhance the environment and ecosystems, including air, water, soil, 
and site quality.'' Performance Measure 4.1 provides that each 
``[l]andowner shall meet or exceed practices prescribed by state 
forestry BMPs that are applicable to the property.''
4. Third-Party Certification Programs Are Regularly Updated
    All three certification programs described above continue to update 
standards on a regular basis. FSC has continually revised its 
Principles and Criteria since 1994, with the most recent revision in 
2012. FSC also developed a U.S. Forest Management Standard in July 
2010, which was updated in September 2012. SFI revises its standards 
every five years, and has most recently updated them in January, 2015. 
ATFS is required to review its standards every five years as part of 
its conditions for endorsement by the Programme for Endorsement of 
Forest Certification, an umbrella organization that works with national 
certification programs to promote sustainable forest management.\23\ 
All programs include opportunities for public and other stakeholder 
input through public comment periods, webinars, and surveys.
---------------------------------------------------------------------------

    \23\ https://www.pefc.org/.
---------------------------------------------------------------------------

D. Existing EPA Tools That Address Stormwater Discharges From Forest 
Roads

    In addition to the state, federal, and third-party BMP-based 
programs described above, EPA administers other programs under the CWA 
that address forest road discharges. Stormwater point source discharges 
from forest roads have traditionally been treated similarly to nonpoint 
sources of pollution under the CWA. EPA has addressed these discharges 
under Sections 303, 305, and 319 of the CWA, and for the coastal areas, 
under Section 6217 of the Coastal Nonpoint Source Pollution Control 
Program under the Coastal Zone Act and Reauthorization Amendments 
(CZARA).\24\
---------------------------------------------------------------------------

    \24\ 16 U.S.C. 1455b.
---------------------------------------------------------------------------

1. Section 319 of the CWA
    Under Section 319 of the CWA, EPA provides technical and financial 
support to states in their administration of programs that address 
pollution from nonpoint sources and activities that are not required to 
be regulated by NPDES permits. Many state nonpoint source management 
programs, which include components for the implementation of forestry-
related BMPs, were initiated and continue to be supported, in part, 
through the use of Section 319 grant funds. According to EPA's 2011 
National Evaluation of the Section 319 Program of the CWA, at least 15 
state programs (AL, AR, CA, GA, KY, LA, MT, NC, OK, OR, SC, TX, VA, WV, 
WY) administer state-wide forestry nonpoint source management programs 
aimed at addressing problems associated with forest harvesting 
operations. At least ten of these states (AL, AR, GA, KY, LA, NC, OK, 
SC, VA, WV) rely on Section 319 grant funding through the relevant 
state forestry agency to support water pollution controls associated 
with forestry activities. In many of these states, the state nonpoint 
source management control agency has a formal relationship with the 
state forestry commission (or agency or department) to jointly 
implement the forestry program. EPA guidance provides that states are 
expected to revise and update their programs every 5 years as part of 
ensuring eligibility for continued funding. (Nonpoint Source Program 
and Grants Guidelines for States and Territories, 2013).
    States have flexibility under the Section 319 program to address 
problems not addressed by the NPDES program. State Section 319 programs 
may encompass watershed or water quality-based approaches aimed at 
meeting water quality standards directly; iterative, technology-based 
approaches based on best management practices or measures, applied on 
either a categorical or site-specific basis; or a mix of these 
approaches. State forestry BMP-based programs apply these approaches 
using forestry BMP prescriptions and monitoring to address water 
quality impairments including forest road runoff, and EPA approves 
these programs as part of the Agency's review of state nonpoint source 
programs.
    EPA has developed a Grants Reporting and Tracking System (GRTS) to 
track projects that receive Section 319 grant funding. It also enables 
EPA and the states to characterize the types of projects funded with 
the use of Section 319(h) grant funds. A sample GRTS query of projects 
shows that a number of Section 319(h) grants have been provided to 
address forest roads, such as road construction and maintenance 
projects, across the country. (Grants Reporting and Tracking System 
Forestry Data Pull, 2016). Section 319 funding remains available to 
address forest roads impacts in those states which have prioritized 
this as an issue in their nonpoint source management plans.
    EPA has published various guidance documents to assist forest 
owners in protecting waters from forestry related runoff, and to help 
states to implement their Section 319 control program. For example, EPA 
published the National Management Measures to Control Nonpoint Source 
Pollution from Forestry (EPA, 2005) which includes BMPs for road 
construction, reconstruction, and management. In 2007, EPA also 
provided funding assistance to the Pennsylvania Department of 
Transportation to develop a manual which provides national guidance on 
effective and efficient practices to apply on dirt and gravel roads to 
reduce erosion, sediment, and dust pollution.\25\
---------------------------------------------------------------------------

    \25\ https://www.epa.gov/polluted-runoff-nonpoint-source-pollution/environmentally-sensitive-maintenance-dirt-and-gravel.
---------------------------------------------------------------------------

2. Section 6217 of CZARA
    Section 6217 of CZARA addresses enhancements to state Coastal Zone 
Management Act (CZMA) programs through development and implementation 
of management measures for nonpoint source pollution control to restore 
and protect coastal waters. This program, which is administered jointly 
by EPA and the National Oceanic and Atmospheric Administration (NOAA), 
directs states and territories with approved CZMA programs to provide 
for implementation of management measures for controlling runoff from 
activities within six categories of nonpoint source activities, 
including forestry. Each coastal state or territory administering a 
CZMA program (approved by NOAA) is required to

[[Page 43505]]

describe its program to implement nonpoint source pollution controls, 
known as management measures, in conformity with a guidance published 
by EPA under CZARA Section 6217(g). The guidance describes ten 
management measures for forestry, including management measures for 
planning, road construction/reconstruction, and road management. As 
implemented under a state's CZMA program, CZARA requires enforceable 
policies and mechanisms, as well as monitoring and tracking of 
management measure implementation. NOAA and EPA are required to review 
and approve coastal nonpoint programs of state and territorial CZMA 
programs, and state authorities are responsible for implementing these 
programs. In all, EPA and NOAA have reviewed the programs submitted by 
33 states and territories and, in many cases, approved such submissions 
with conditions. Over time, affected states and territories took action 
to address the program conditions incrementally. Since the federal 
agencies' initial approvals with conditions, all but 10 states have now 
met all of the outstanding conditions.\26\
---------------------------------------------------------------------------

    \26\ https://coast.noaa.gov/czm/pollutioncontrol/.
---------------------------------------------------------------------------

3. Sections 305(b) and 303(b) of the CWA
    Under Section 305(b) of the CWA, states are required to assess the 
quality of their surface waters and report this information to EPA. In 
addition, every 2 years Section 303(d) requires states to identify on 
their Section 303(d) lists, which they submit to EPA for approval, 
those waters that are not attaining water quality standards, referred 
to as ``impaired waters,'' and waters not expected to attain water 
quality standards by the next two-year listing cycle, referred to as 
``threatened waters.'' 33 U.S.C. 1313(d)(1)(A); 40 CFR 130.7(b). States 
must also establish a priority ranking for establishing total maximum 
daily loads (TMDLs) of pollutants for those waters. Id. TMDLs are 
``pollution budgets'' that calculate how much of a given pollutant a 
waterbody can assimilate, including a margin of safety, without 
exceeding its applicable water quality standards. 33 U.S.C. 
1313(d)(1)(C). TMDLs also allocate shares of the waterbody's 
assimilative capacity for that pollutant to all of its point and 
nonpoint sources. 40 CFR 130.2(i). Pollutant allocations may be 
assigned to individual sources or aggregated to sectors such as forest 
roads. Like Section 303(d) lists, states submit TMDLs to EPA for 
approval.
    Impaired waters lists and TMDLs established for those impaired 
waters are ``informational tools,'' Pronsolino v. Nastri, 291 F.3d 
1123, 1129 (9th Cir. 2002), that help states evaluate the significance 
of pollutant sources like forest roads in contributing to water quality 
impairments in the U.S and guide implementation of measures to address 
those impairments. Nationally, pathogens, mercury, other metals, 
sediment, nutrients, and organic enrichment/oxygen depletion are 
identified as the leading causes of impairment of all assessed water 
bodies, based on state electronic data submissions from 2004 through 
2010.
    While TMDLs at their core are pollutant loading calculations and 
allocations, they also can provide a ``comprehensive framework'' for 
pollution reduction in a body of water that fails to meet state water 
quality standards. Amer. Farm Bureau Fed'n v. EPA, 792 F.3d 281, 287-
288 (3rd Cir. 2015). While approving or establishing a TMDL, EPA 
requires ``reasonable assurance'' from the states that their TMDL 
implementation plans will meet their stated goals, i.e., achieve the 
TMDL's allocations and implement the applicable water quality 
standards. Id. at 300. In support of EPA's recently revised TMDL for 
Lake Champlain, for example, Vermont detailed specific actions it would 
take to reduce the flow of sediment into Lake Champlain, including 
enhancing its forest roads forest management practices to reduce 
erosion (EPA Region 1, 2016).
    EPA considered national TMDL data to determine whether forest roads 
have been identified as sources of water quality impairment and 
addressed in TMDL load allocations designed to help meet water quality 
standards.\27\ For example, Endicott (2008) indicates that in 
California TMDLs were required for 10 river basins where silviculture 
was identified as a potential source. EPA reviewed three of these TMDLs 
(Upper Main Eel River and Tributaries TMDL, 2004; Mad River TMDL, 2007; 
Redwood Creek TMDL, 2011) and found that roads and road related 
landslides were the leading anthropogenic cause of sediment loading in 
these watersheds. While EPA is unable to develop national-level summary 
data to describe the degree of impairments from forest roads, EPA notes 
that these and other TMDLs serve as existing CWA planning tools that 
guide silviculture-related pollutant reduction activities on a 
watershed-specific basis. See also Pronsolino v. Nastri supra at 1129, 
where the Ninth Circuit upheld an EPA-established TMDL addressing 
sediment pollution to the Garcia River caused by roads, timber-
harvesting, road surfaces, and road and skid trail crossings.
---------------------------------------------------------------------------

    \27\ Unfortunately, EPA's national-level TMDL data does not 
contain detailed information on specific impairment sources such as 
forest roads. See, for example, the state report ``2012 Pennsylvania 
Integrated Water Quality Monitoring and Assessment Report,'' which 
identifies silviculture as responsible for 19 miles of impairments 
on state waters. Even with state-level data such as this report 
(which still does not make an explicit connection between forest 
roads and impairments), EPA found it exceedingly difficult to gather 
and assess this type of data.
---------------------------------------------------------------------------

VII. Rationale for EPA's Determination Not To Establish New Regulatory 
Requirements for Forest Roads Discharges

    As discussed above, many rigorous programs exist at every level of 
government as well as in the private sector to address stormwater 
discharges from forest roads in the United States. The programs are 
regularly updated to reflect new technology and research findings, are 
specifically tailored for the locations in which they are implemented, 
and have high implementation rates. While these programs have 
limitations and may vary in their effectiveness, EPA has concluded that 
providing support for further improvement to these programs will be 
more effective in further addressing discharges from forest roads than 
would the establishment of a new federal regulatory program under CWA 
Section 402(p)(6).
    A number of practical considerations also militate against the 
establishment of a new federal regulatory program for forest roads. 
These include the site-specific nature of the environmental problem, 
the complex ownership arrangements of forest roads, and the limited 
financial resources and legal tools for addressing these roads, all 
discussed further below. A new program could require the expenditure of 
substantial resources while duplicating or displacing existing 
programs, with limited incremental environmental results. EPA has 
determined that the theoretical benefits of creating a ``federal 
floor'' do not outweigh its certain implementation problems, high 
costs, and potential duplication or displacement of longstanding and 
maturing federal, state, and private initiatives to address stormwater 
discharges from forest roads.
    A primary difficulty in establishing a new, nationwide regulatory 
regime is the variability in water quality impacts from forest roads 
across the country. Many factors affect the extent to which BMPs are 
needed and those best suited to particular locations, including 
physical and meteorological factors (e.g., climate, topography, soil 
type), which affect the nature of erosion and

[[Page 43506]]

sedimentation; the intensity of timber operations; and localized 
scientific research and water quality data. A national regulation 
addressing such site-specific issues would likely be either too general 
or too complicated to be successful. The current multi-faceted, multi-
layered landscape best supports the site-and region-specific nature of 
effective BMPs.
    The options laid out in Section 402(p)(6) of the CWA, the authority 
pursuant to which EPA could have designated stormwater discharges from 
forest roads for regulation, resemble the existing universe of forest 
roads control programs in the U.S. The types of regulatory actions that 
EPA could hypothetically take under Section 402(p)(6) are similar to 
the types of requirements and programs that states and other entities 
across the U.S. have already established, as described above. Section 
402(p)(6) authorizes EPA to: ``establish priorities, establish 
requirements for state stormwater management programs, and establish 
expeditious deadlines'' which may include ``performance standards, 
guidelines, guidance, and management practices and treatment 
requirements, as appropriate.'' 33 U.S.C. Sec.  1342(p)(6). Many 
``state stormwater management programs'' already exist and address 
discharges from forest roads in a manner specifically tailored to 
conditions in each state. See Decker v. Nw. Envtl. Def. Ctr., 133 S. Ct 
1326, 1338 (2013) (``Indeed, Congress has given express instructions to 
the EPA to work `in consultation with State and local officials' to 
alleviate stormwater pollution by developing the precise kind of best 
management practices Oregon has established here. 33 U. S. C. Sec.  
1342(p)(6)''). In addition, states, agencies and organizations, 
including the USFS and EPA, have published ``guidelines'' and 
``guidance'' discussing ``management practices.'' Every state and state 
organization that submitted comments to inform EPA's determination 
strongly opposed additional federal regulations. EPA has decided to 
help states strengthen their programs rather than supplant them, 
consistent with the CWA's policy to ``recognize, preserve, and protect 
the primary responsibilities and rights of States to prevent, reduce, 
and eliminate pollution'' and to plan the ``use . . . of land and water 
resources.'' 33 U.S.C. 1251(b).
    Supporting rather than duplicating state programs is also 
consistent with the CWA's policy of fostering governmental efficiency: 
to ``encourage the drastic minimization of paperwork and interagency 
decision procedures, and the best use of available manpower and funds, 
so as to prevent needless duplication and unnecessary delays at all 
levels of government.'' 33 U.S.C. 1251(f). An EPA program would add 
another layer of bureaucracy for both regulators and the private 
sector, sow confusion about program requirements and responsibilities, 
and lead to an inefficient use of already thin management resources, 
all for potentially limited environmental benefit.
    While Section 402(p)(6) could otherwise generally allow for 
regulation through some sort of permitting, Congress has specifically 
foreclosed that option for discharges ``resulting from the conduct of 
the following silviculture activities conducted in accordance with 
standard industry practice: nursery operations, site preparation, 
reforestation and subsequent cultural treatment, thinning, prescribed 
burning, pest and fire control, harvesting operations, surface 
drainage, or road construction and maintenance.'' 33 U.S.C. 1342(l). 
Congress has also precluded third-party citizen suits to enforce any 
non-permitting program established under Section 402(p)(6) or any other 
limitations applied to silviculture activities. In the absence of these 
implementation and enforcement mechanisms, it would be difficult to 
provide for effective federal implementation and compliance assurance 
for a new set of national forest road discharges.
    Some commenters urged EPA to establish mandatory requirements 
pursuant to Section 402(p)(6), including prioritization of forest 
management areas, requiring road inventories, and monitoring for water 
quality standards. Many of these elements are part of state programs 
already. Requiring all forest landowners in the country to submit data 
to EPA about roads on their properties would necessitate a resource-
intensive outreach operation. The large number of private family forest 
owners in the U.S. and Internet broadband limitations in rural areas, 
among many other factors, would make it difficult to ensure that forest 
road owners and operators are aware of and comply with such this 
requirements; legacy roads with no apparent owner would present even 
greater challenges. Additionally, as one commenter pointed out, many 
programs are targeted at certain impacted watersheds or aquatic 
species. An inventory of all forest roads, many of which do not cause 
water quality problems, does not necessarily provide information needed 
to address these particular impacts. Obtaining forest roads inventory 
information would likely be easier where large areas of forest are 
managed by a single entity, such as the USFS, but those entities are 
the ones most likely to already be engaging in inventory efforts (as 
described in section VI.B.1 of this document). Given these challenges, 
EPA does not believe that creating a new federal inventory of forest 
roads is a cost-effective use of EPA's limited resources.
    Requiring water quality monitoring poses another distinct set of 
problems. Water quality monitoring is in-situ (ambient water) sampling 
for one or a selected set of environmental indicators. These metrics 
can be biological (e.g., macroinvertebrates or fish community health), 
chemical (e.g., pollutant concentrations), or physical (e.g., 
geomorphology). This approach is not typically used to assess one or a 
few BMPs because in-situ water quality is influenced by multiple local 
and upstream factors/sources, and statistical distinctions between 
these factors and determining relative contributions may be impossible. 
Endicott (2008) reported findings ``that the biotic and chemical 
`noise' in larger streams renders the water quality effects of forestry 
activities using BMPs undetectable.''
    EPA recognizes that existing forest road BMP programs have 
limitations, including limited funding. Resource constraints are a 
primary difficulty facing both state and federal programs, limiting 
their abilities to implement and monitor BMPs. Yet a new set of 
requirements from EPA would not address the funding gap. Indeed, 
another federal program could divert resources from on-the-ground 
stream protection efforts to bureaucratic reshuffling. EPA has decided 
not to expend resources on creating, implementing, and enforcing a new 
national program that may not tangibly improve water quality.

VIII. Facilitating Continuous Improvement of Forest Road Programs

    As discussed above, programs at the state, federal, and local 
levels, as well as within the private sector, have demonstrated 
positive momentum in strengthening efforts to address stormwater 
discharges from forest roads. EPA seeks to further facilitate 
continuing improvements in working to address water quality impacts 
from forest roads. Thus, rather than superimposing additional EPA-
regulatory programs over existing programs, EPA plans to help 
strengthen these existing programs by forming an ongoing dialogue with 
all relevant stakeholders (including industry, environmental groups, 
academics, and

[[Page 43507]]

government agencies at the federal, state, tribal, and local levels) on 
program improvements, technical and policy issues, research results, 
state of the art technologies, success stories, and solutions to 
problem areas. This forum could provide an opportunity for stakeholders 
to exchange information and expertise. EPA envisions that a major part 
of these discussions will focus on specific problems and solutions to 
forest roads, such as existing/legacy roads or stream crossings as well 
as particularly effective forest road programs and best practices. 
Working with stakeholders collaboratively, the forum could develop a 
national compendium of highly effective components of private or 
governmental forest roads programs to serve as a resource for states, 
tribes, federal agencies, local government, and industry. The 
compendium could serve as an indicator of expectations for development, 
implementation, and/or revisions of forest road programs by 
highlighting existing robust efforts and the latest developments of 
evolving strong programs.

IX. Response to Key Comments on Existing BMP-Based Programs

    The discussion below responds to significant issues commenters 
raised with regard to the effectiveness of existing BMP-based programs.
    Some commenters expressed concerns about the effectiveness of BMPs. 
In response, EPA makes an important distinction between the well 
documented ability of properly implemented BMPs to adequately control 
the discharge of pollutants, and situations where BMPs are improperly 
implemented or maintained (see multiple studies discussed in Part V). 
As these studies generally conclude, most BMPs are highly effective 
when appropriately designed and implemented; this includes choosing the 
right practice for particular situations and ensuring proper operation 
and maintenance. BMPs are ineffective or perform sub-optimally when not 
properly sited, installed, or maintained. These paradigms hold true for 
all water quality control technologies, not just BMPs, and underscore 
the importance of vigilant operation and maintenance rather than a 
conclusion that BMPs are not effective at protecting water quality. For 
example, Wisconsin DNR (2013) found that when BMPs were applied 
correctly no adverse impacts to water quality were found 99% of the 
time, and Montana DNRC (2014) reported that Montana's forestry BMPs 
were effective in protecting soil and water resources 98% of the time. 
In addition, as with most technologies, it is important to note that 
BMP science continues to evolve and improve.
    One commenter mentioned a study of two watersheds in the U.S. 
Pacific Northwest region, which found that 44% of 80 sediment debris 
slides were associated with roads, even though roads comprised only 
3.1% of the area. However, the authors of the study concluded that 
standard BMPs were the best approach to reducing erosion and sediment 
delivery rates. This is the approach that states and others are already 
pursuing in that region.
    Another commenter pointed to low BMP efficiency data in Edwards and 
Williard (2010, as cited in Nolan et al., 2015) but the cited article 
examined the efficiency of forest harvesting BMPs in reducing sediment, 
not BMPs related to forest roads in particular. EPA also recognizes 
that state BMP-based programs have limitations, including that they may 
not be fully implemented, that their effectiveness differs based on 
numerous variables, and the difficulty in measuring quantitative 
results.\28\ A new federal regulatory program under CWA Section 
402(p)(6), however, would not necessarily improve implementation rates, 
especially given the new limitations in CWA Section 402(l), which 
preclude the use of permits to implement any such program or of citizen 
suits to enforce any new federal requirements.
---------------------------------------------------------------------------

    \28\ For example, Virginia has an implementation rate of 78% for 
forest road BMPs (SGSF BMP Report, 2012). In addition, the following 
states report lower than the national average of 86.7% for BMP 
implementation rates of stream crossing BMPs: Vermont, 68%; North 
Carolina, 72%, Ohio, 78%, Maryland, 67%, and Oregon, 71%. (NASF, 
2015).
---------------------------------------------------------------------------

    A few commenters discussed specific state forest road programs, 
such as Oregon's and Washington's. One commenter stated that Oregon's 
forest roads program is too flexible and is not adequately enforced. 
The commenter specifically identified the approval/rejection process 
for written plans as not being sufficiently stringent because there is 
no requirement to approve or deny a plan. With regard to Oregon (and 
other states), given the nature and scope of the concerns posed by 
forest road runoff, a reasonable degree of flexibility is valuable, as 
it allows for a tailored approach to addressing forest road discharges. 
See Decker v. NEDC, (``Oregon has invested substantial time and money 
in establishing these practices. In addition, the development, siting, 
maintenance, and regulation of roads--and in particular of state forest 
roads--are areas in which Oregon has considerable expertise'').
    Another commenter stated that, in addition to requiring BMPs, 
Washington State also requires water quality-based numeric criteria for 
turbidity and has rules for antidegradation, and that this should be 
required of all states. With regard to Washington State, EPA recognizes 
that states currently have various approaches to addressing 
sedimentation concerns (e.g., numeric and narrative turbidity 
standards, dissolved oxygen standards, temperature standards, etc.) as 
part of their water quality standards programs. EPA agrees that 
applying numeric standards can be extremely effective in protecting 
water quality. However, states are well situated to understand the 
scope and nature of environmental concerns posed by forest road runoff 
in their states and apply state water program requirements to those 
concerns accordingly.
    Some commenters, urged EPA to implement a national water quality-
based monitoring program for forest roads. Requiring water quality 
monitoring for stormwater discharges from forest roads is infeasible 
for the reasons discussed in Section VII. Examining forest road BMP 
implementation on existing roads indicates whether existing programs 
are taking available and reasonable steps to address water quality 
concerns. EPA recognizes that most evaluations and determinations of 
BMP implementation are qualitative, but nonetheless, that information 
constitutes the best available information for EPA to make its 
decision. Extreme storms can pose challenges to the use and performance 
of BMPs, but BMPs can be tailored to some degree in areas subject to 
such events. A federal regulation would not alleviate risks posed by 
extreme storms because it would not be fair or reasonable to impose 
BMPs in all extreme storm events.\29\
---------------------------------------------------------------------------

    \29\ NPDES Bypass and Upset provisions at 40 CFR Sections 
122.41(m) and (n) providing relief in certain circumstances to NPDES 
dischargers.
---------------------------------------------------------------------------

    One commenter stated that forest road BMP programs tend to focus on 
construction of new roads and fail to address older roads, often built 
before BMPs were in place (i.e., they are either ``grandfathered in'' 
or subject to requirements only when brought back into use, 
reconstructed, or at risk of significant failure). The commenter 
observed that older roads can be significant sources of sediment since 
they may be poorly located and built with few if any features to 
control erosion (citing Endicott 2008, which includes some studies that 
identify legacy roads as sources but do not

[[Page 43508]]

provide data regarding sediment discharged by legacy roads). EPA 
recognizes that legacy roads present a challenge and a potential source 
of sediment. Legacy roads are also the most challenging types of roads 
to address through regulation, however. Legacy roads are often no 
longer in use, so there may not be an ongoing silvicultural operation 
to fund BMPs. They may have non-forest uses, also complicating 
responsibility and liability assignment, or they may not be used for a 
period of time while timber is growing and then they may be placed back 
into use when it is ready for harvest. Legacy roads may also be so 
overgrown with vegetation that their presence is no longer detectable.
    Nonetheless, several state programs require older roads to be 
upgraded to current BMP standards if they are brought back into 
service. Endicott (2008) indicates that 24 states had forest road BMPs 
that address road closure. A more recent review indicates that 34 
states have BMPs that address forest road retirement (State Program 
Summary, 2016). Comments indicate that California, Washington, and 
Oregon are among those states having programs addressing legacy road 
issues.
    A few commenters stated that stream crossings for forest roads are 
especially vulnerable locations that can lead to significant erosion. 
One commenter stated that 5% of truck road stream crossings in the 
southern Piedmont region of Virginia were not meeting the relevant 
stream crossing BMPs (Nolan et al., 2015) and that failure to meet BMPs 
in these areas will have a disproportionately negative impact on water 
quality as compared to upland BMP violations. Another layer of 
regulations from EPA, however, would not guarantee that the remaining 
5% of stream crossings would incorporate appropriate BMPs. While stream 
crossings are indeed a high risk area for forest road runoff, a recent 
EPA analysis of state programs showed that 46 states (92%) have 
developed BMPs for stream crossings. (State Program Summary, 2016). 
Additionally, BMP guidance documents addressing road placement make 
clear that roads should avoid or minimize stream crossings and riparian 
areas. Thus, a BMP based approach reduces the incidence of road-stream 
crossings and, when deemed unavoidable, BMPs have been developed to 
install stream crossings while minimizing erosion.
    A commenter also stated that some states do not consider the 
effects of diversion and natural disturbances when designing BMPs for 
stream crossings. These are important factors to consider. They are 
not, however, the only variables considered in a stream crossing 
design; stream flow and volume, soil type, volume and type of vehicle 
traffic, climate, and many other factors also play a role in 
determining the optimal design for a stream crossing. Effective stream 
crossing BMPs depend on site-specific conditions, reflecting the 
difficulty of setting one-size-fits-all federal requirements. In one 
study, researchers examined the effects of upgrading poorly designed 
stream crossings and concluded that the enhanced stream crossings 
produced little sediment and that improved stream crossings could 
significantly reduce sediment contributions from forest roads (Nolan et 
al., 2015). One commenter spoke favorably of several BMPs developed by 
the USFS for use at stream crossings and recommended that EPA adopt 
them nationally. EPA encourages state programs to consider USFS stream 
crossing BMPs for their menus of BMPs.
    EPA also received several comments regarding the compliance and 
monitoring aspects of state programs. One commenter stated that BMP 
effectiveness rates are overstated and suggested that the appropriate 
baseline for comparison should be forests in their natural conditions 
with no roads, whereas most studies compare forest roads with BMPs to 
forest roads with no BMPs. The commenter also asserted that, based on 
three studies, the actual efficiency of forest road BMPs is 53-94%. EPA 
notes in response that forest roads play a critical role in 
silviculture, recreation, fire suppression, and other uses. EPA does 
not expect forest roads to be absent from the landscape and therefore 
does not think that virgin forest must always necessarily serve as the 
baseline for measuring BMP effectiveness.
    A commenter also pointed out that most BMP monitoring \30\ is 
conducted during dry periods, when effectiveness at preventing 
stormwater runoff may be more difficult to discern. The commenter noted 
that variability in BMP performance monitoring can be as high as 50-
100%, which would require frequent sampling to distinguish sediment 
derived from forest roads versus other sources. A number of BMP 
performance studies are conducted under wet weather conditions, 
including most of those cited in Section V of this document. However, 
BMP effectiveness also can be assessed to a large extent in dry 
weather, as evidence of soil movement is often visible for a 
significant time period after rainfall events. For example, gullying or 
landslides will be clearly visible while sediment deposition in low 
areas or waterbodies will also be visible.
---------------------------------------------------------------------------

    \30\ BMP monitoring refers in this case to assessment of BMP 
performance effectiveness, which includes verifying that the 
structure/measures are in place and functioning. BMP monitoring is 
different from water quality monitoring, which involves monitoring a 
waterbody for particular environmental indicators.
---------------------------------------------------------------------------

    Another commenter stated that standardizing BMP compliance 
assessments and reporting protocols is necessary. They add that most 
monitoring focuses on whether a BMP has been implemented, rather than 
monitoring water quality for compliance with water quality standards. 
The commenter cited data from Virginia that noted a 32% non-compliance 
rate for stream crossing BMPs. EPA recognizes that states have used a 
variety of monitoring and reporting mechanisms over time and that this 
can inhibit broader analyses about BMP compliance. However, as 
discussed in Section VI.A.2 of this document, two large groups of 
states have adopted regional standardized monitoring protocols to 
promote consistency in compliance assessment and reporting.
    First, the SGSF has been implementing a broad monitoring program in 
13 southeastern states for nearly a decade. Second, the joint effort 
between USFS and NAASF developed a similar standardized protocol for 
evaluating BMP implementation and effectiveness. These two protocols 
have spread a standardized monitoring process to a significant number 
of states with active forestry programs. Such standardization efforts 
are examples of the type of intra-state consistency that a federal EPA 
program could theoretically institute; their spread in the absence of 
EPA regulations provides an example in which a new EPA program would be 
duplicative.
    Some commenters stated the lack of a national BMP program leads to 
inconsistent BMP application and insufficient water quality 
protections. EPA sees the range of designs in BMP programs as an 
appropriate response to the diversity of conditions these programs are 
intended to address. State or regional timber operations vary in 
intensity, as do the types of forest management programs states or 
other oversight agencies implement. BMPs used at a site will differ 
depending on the factors above, as well as others, such as localized 
scientific research that determines the most effective approaches to 
managing stormwater. Within different state frameworks, certain aspects 
of BMP programs are largely consistent. For example, state BMP 
categories typically encompass

[[Page 43509]]

forest road location/design/construction; road maintenance; stream 
crossings; stream management zones/bank stabilization/buffer strips; 
and many states address forest road retirement and wet weather/winter 
use.
    Many states are taking the lead in enhancing their programs to 
encompass newly developed methods to reduce water quality impacts from 
forest roads. For example, CA's ``Road Rules, 2013'', which was first 
implemented in January 2015, requires that all forest roads used as 
part of an approved plan be hydrologically disconnected from waters 
(EPA-HQ-OW-2015-0668-0055). In the Southern region, the Southern Group 
of State Foresters Silviculture Best Management Practices 
Implementation Monitoring framework requires all southern states to 
include in their implementation monitoring reports counts of water 
quality risks. Finally, while ``traditionally a problem area within all 
states, compliance with stream crossing BMPs continues to improve as a 
result of increased education of landowners and managers as well as 
increased acreage of certified forestland in the region (Schilling et 
al., 2009).'' [Ice et al., 2010.]
    One commenter stated, ``Congress has failed to adequately invest in 
the National Forest System roads budget. Annual spending has declined 
from over $236 million to less than $159 million in the last six fiscal 
years, when adjusted for inflation.'' This has helped to contribute to 
the development of a more than $5 billion deferred maintenance backlog 
on the National Forest System. This commenter also suggested that, 
``[r]egulating stormwater discharges from USFS roads will do nothing to 
address either the forest health crisis or the disinvestment in 
maintaining the existing Forest Road system'' (Id.). EPA acknowledges 
that both the USFS and BLM face resource constraints, often must 
address higher priority issues such as fire suppression to protect 
lives, and confront other challenges that limit the ability to fully 
address all issues arising from forest road activity when it comes to 
maintaining their transportation networks. Another layer of EPA 
regulations, in addition to existing federal programs addressing water 
resources protection and restoration, would not address these resources 
constraints and would likely do little to enhance water quality.
    In conclusion, none of these comments alters EPA's determination 
not to establish a new regulatory program for discharges from forest 
roads under CWA Section 402(p)(6). While EPA recognizes that discharges 
from forest roads have significant impacts on water quality in many 
parts of the country, the Agency has concluded that the most effective 
way to make further progress in addressing these issues is to support 
existing state, tribal, federal, and third-party programs. Given the 
diversity of forest roads programs in this country, some programs will 
necessarily be more rigorous than others. EPA has considered this 
variability, but concluded that any consistency that a national 
regulation could theoretically achieve is far outweighed by the 
challenges of its implementation.

X. References

Anderson, C.J., & Lockaby, B.G. (2011). The effectiveness of 
forestry best management practices for sediment control in the 
southeastern United States: A literature review. Southern Journal of 
Applied Forestry, 35(4), 170-177.
Appelboom, T.W., Chescheir, G.M., Skaggs, R.W., & Hesterberg, D.L. 
(2002). Management practices for sediment reduction from forest 
roads in the coastal plains. Transactions of the ASAE, 45(2), 337.
BLM. (2005). Land Use Planning Handbook; BLM Handbook H-1601-1.
BLM. (2011). Contract for the Sale of Timber and Other Wood Products 
Lump Sum Sale.
BLM. (2015). Public Land Statistics 2014. Volume 199.
BLM. (2016). Appendix J--Best Management Practices. BLM RMPWO Vol. 
3.
Bureau of Indian Affairs. (2009). FY2009: Quarter 4 Catalog of 
Forest Acres.
Butler, B., Hewes, J.H., Dickinson, B.J., Andrejczyk, K., Butler, 
S.M., & Markowski-Lindsay, M. (2016). USDA Forest Service National 
Woodland Owner Survey: A technical document supporting the Forest 
Service update of the 2010 RPA assessment. USFS.
Cissel, R., Black, T.A., Nelson, N., & Luce, C.H. (2014). Monitoring 
the Hydrologic and Geomorphic Effects of Forest Road Decommissioning 
and Road Improvements. USFS.
Cissel, R., Black, T.A., Nelson, N., & Luce, C.H. (2014). Southwest 
Crown of the Continent GRAIP roads assessment. US Department of 
Agriculture, Forest Service, Rocky Mountain Research Station, Fort 
Collins, Colorado.
Clarkin, K., Conner, A., Furniss, M.J., Gibernick, B., Love, M., 
Moynan, K., & Wilson, S. (2005). National inventory and assessment 
procedure for identifying barriers to aquatic organism passage at 
road-stream crossings. USFS.
Cristan, R., Aust, W.M., Bolding, M.C., Barrett, S.M., Munsell, 
J.F., & Schilling, E. (2016). Effectiveness of forestry best 
management practices in the United States: Literature review. Forest 
Ecology and Management, 360, 133-151.
Decker v. Northwest Environmental Defense Center, 133 S. Ct. 1326, 
568 U.S., 185 L. Ed. 2d 447 (2013).
Dub[eacute], K., Shelly, A., Black, J., & Kuzis, K. (2010). 
Washington Road Sub-Basin Scale Effectiveness Monitoring First 
Sampling Event (2006-2008) Report. Department of Natural Resources, 
State of Washington, 102.
Edwards, P.J., & Williard, K.W. (2010). Efficiencies of forestry 
best management practices for reducing sediment and nutrient losses 
in the eastern United States. Journal of Forestry, 108(5), 245-249.
EPA. (2004). Upper Main Eel River and Tributaries (including Tomki 
Creek, Outlet Creek and Lake Pillsbury) Total Maximum Daily Loads 
for Temperature and Sediment.
EPA. (2005). National Management Measures to Control Nonpoint Source 
Pollution from Forestry.
EPA. (2007). Mad River Total Maximum Daily Loads for Sediment and 
Turbidity.
EPA. (2013). Nonpoint Source Program and Grants Guidelines for 
States and Territories.
EPA. (2016). Grants Reporting and Tracking System Forestry Data 
Pull.
EPA Region 1. (2016). Phosphorus TMDLs for Vermont Segments of Lake 
Champlain.
Great Lakes Environmental Center, & Endicott, D. (2008). National 
Level Assessment of Water Quality Impairments Related to Forest 
Roads and Their Prevention by Best Management Practices. Final 
Report. Report prepared for US Environmental Protection Agency, 
Office of Water. Contract No. EP-C-05-066, Task Order, 2, 250.
Ice, G. (2004). History of innovative best management practice 
development and its role in addressing water quality limited 
waterbodies. Journal of Environmental Engineering, 130(6), 684-689.
Ice, G. & Schilling, E. (2012). Assessing the effectiveness of 
contemporary forestry best management practices (BMPs): Focus on 
roads. NCASI. Special report No. 12-01.
Ice, G.G., Schilling, E., & Vowell, J. (2010). Trends for forestry 
best management practices implementation. Journal of Forestry, 
108(6), 267-273.
Idaho Department of Environmental Quality. (2015). Idaho Nonpoint 
Source Management Plan.
Megahan, W.F., & King, J.G. (2004). Erosion, sedimentation, and 
cumulative effects in the northern Rocky Mountains.
Miller, S.A., Gordon, S.N., Eldred, P., Beloin, R.M., Wilcox, S., 
Raggon, M., . . . & Muldoon, A. (2015). Northwest Forest Plan the 
First 20 Years (1994-2013): Watershed Condition Status and Trend.
Montana Dept. of Natural Resources & Conservation. (2014). Forestry 
Best Management Practice (BMP) 2014 Monitoring Report Executive 
Summary.
Montana Dept. of Natural Resources & Conservation. (2015). Montana 
Forestry Best Management Practices.
NASF. (2015). Protecting Water Quality through State Forestry Best 
Management Practices.
NCASI Forest Watershed Task Group. (2001). Forest roads and aquatic 
ecosystems: a review of causes, effects, and management practices.
Nolan, L., Aust, W.M., Barrett, S.M., Bolding, M.C., Brown, K., & 
McGuire, K. (2015). Estimating costs and effectiveness of

[[Page 43510]]

upgrades in forestry best management practices for stream crossings. 
Water, 7(12), 6946-6966.
North Carolina Forest Service. (2006). North Carolina Forestry Best 
Management Practices Manual to Protect Water Quality.
Northwest Environmental Defense Center v. Brown, 640 F.3d 1063 (9th 
Cir. 2011).
Olszewski and Jackson. (2006). A Primer on the Top Ten Forest 
Environmental and Sustainability Issues in the Southern United 
States. NCASI. Special report No. 06-06.
Oregon Department of Forestry. (2015). Board of Forestry Streamside 
Buffer (Riparian) Rule Analysis Decision.
Redwood National and State Parks. (2011). Redwood Creek--Progress 
Report on Erosion Control Work and Sediment TMDL.
Schilling, E. (2009). Compendium of forestry best management 
practices for controlling nonpoint source pollution in North 
America. NCASI. Technical bulletin No. 966.
SFI. (2015). Report on the Status of Logger Training and Education 
(LT&E) Programs in 34 Forested U.S. States & 6 Canadian Provinces.
SGSF. (2012). Implementation of Forestry Best Management Practices: 
2012 Southern Region Report.
SGSF. (2007). Silviculture Best Management Practices Implementation 
Monitoring: A Framework for State Forestry Agencies.
Skaugset, A., & Allen, M.M. (1998). Forest Road Sediment and 
Drainage Monitoring Project Report for Private and State Lands in 
Western Oregon.
Sugden, B.D., Ethridge, R., Mathieus, G., Heffernan, P.E., Frank, 
G., & Sanders, G. (2012). Montana's forestry Best Management 
Practices Program: 20 years of continuous improvement. Journal of 
Forestry, 110(6), 328-336.
Tetra Tech Inc. (2016). Updated Summary of State Forest Road BMP 
Program Information.
USFS. (1988). Soil and water conservation practices handbook.
USFS. (2007). Best Management Practices (BMP) Manual-Desk Reference: 
Implementation and Effectiveness for Protection of Water Resources.
USFS. (2007). Best Management Practices (BMP) Monitoring Manual-
Field Guide: Implementation and Effectiveness for Protection of 
Water Resources.
USFS. (2012). National Best Management Practices for Water Quality 
Management on National Forest System Lands Volume 1: National Core 
BMP Technical Guide.
USFS. (2014). USDA Forest Service Update March 2014 Subject: Aquatic 
Organism Passage.
USFS. (2015). National Best Management Practices Monitoring Summary 
Report Program Phase-In Period Fiscal Years 2013-2014.
USFS. (2015). USDA Forest Service Strategic Plan: FY 2015-2020.
Wisconsin DNR. (2013). Wisconsin's Forestry Best Management 
Practices (BMPs) for Water Quality 2013 BMP Monitoring Report.


    Dated: June 27, 2016.
Joel Beauvais,
Deputy Assistant Administrator, Office of Water.
[FR Doc. 2016-15844 Filed 7-1-16; 8:45 am]
 BILLING CODE 6560-50-P