Forestry impacts on freshwater habitat of anadromous salmonids in the Pacific Northwest and Alaska requirements for protection and restoration

[From the U.S. Government Printing Office, www.gpo.gov]

Science for Solutions

""I't OF
NOAA COASTAL OCEAN PROGRAM
Decision Analysis Series No. 7

FORESTRY IMPACTS ON FRESHWATER
HABITAT OF ANADROMOUS SALMONIDS
IN THE PACIFIC NORTHWEST AND ALASKA--
REQUIREMENTS FOR PROTECTION
AND RESTORATION

Michael L. Murphy
October 1995

V-J

@QL638.52 U.S. DEPARTMENT OF COMMERCE
M87 National Oceanic and Atmospheric Administration
1995 Coastal Ocean Office

DECISION ANALYSIS SERIES

The Decision Analysis Series has been
established by NOAA's Coastal Ocean
Program (COP) to present documents for
coastal resource decision makers which
contain analytical treatments of major
issues or topics. The issues, topics, and
principal investigators have been selected
through an extensive peer review
process. To learn more about the COP or
the Decision Analysis Series, please
write:

NOAA
Coastal Ocean Office
1315 East West Highway
Silver Spring, MD 20910

phone: 301-713-3338
fax: 301-713-4044

Cover photo: Landscape view of private timberlands in
southeast Alaska showing a typical mosaic of harvested
areas interspersed with patches of uncut forest and buffer
zones along streams. (Photo by K Koski, NMFS).

Science for Solutions

NOAA COASTAL OCEAN PROGRAM
Decision Analysis Series No. 7

FORESTRY IMPACTS ON FRESHWATER
HABITAT OF ANADROMOUS SALMONIDS
IN THE PACIFIC NORTHWEST AND ALASKA--
REQUIREMENTS FOR PROTECTION
AND RESTORATION

Michael L. Murphy

NOAA National Marine Fisheries Service
Alaska Fisheries Science Center
Auke Bay Laboratory

October 1995

U.S. DEPARTMENT OF COMMERCE
Ronald If Brown, Secretary
National Oceanic and Atmospheric Administration
D. James Baker, Under Secretary
Coastal Ocean Office
410N
/4

Donald Scavia, Director

This I?ublication should be cited as:

Murphy, Michael L. 1995. Forestry Impacts on Freshwater Habitat of Anadromous Salmonids in
the Pacific Northwest and Alaska--Requirements for Protection and Restoration. NOAA Coastal
Ocean Program Decision Analysis Series No. 7. NOAA Coastal Ocean Office, Silver Spring, MD.
156 pp.

This publication does not constitute an endorsement of any commercial product or
intend to be an opinion beyond scientific or other results obtained by the National
Oceanic and Atmospheric Administration (NOAA). No reference shall be made
to NOAA, or this publication furnished by NOAA, in any advertising or sales
promotion which would indicate or imply that NOAA recommends or endorses any
proprietary product mentioned herein, or which has as its purpose an interest to
cause directly or indirectly the advertised product to be used or purchased because
of this publication.

Note to Readers

Forestry Impacts on Freshwater Habitat ofAnadromous Salmonids in the Pacific Northwest and
A laska--Requirements for Protection and Restoration was developed by Michael L. Murphy of
the NOAA National Marine Fisheries Service's Alaska Fisheries Science Center with funding
from the NOAA Coastal Ocean Program (COP). The document presents a science overview of
the major forest management issues involved in the recovery of anadromous salmonids affected
by timber harvest in the Pacific Northwest and Alaska. The synthesis reviews salmonid habitat
requirements and potential effects of logging, describes the technical foundation of forest
practices and restoration, analyzes current federal and non-federal forest practices, and
recommends required elements of comprehensive watershed management for recovery of
anadromous salmonids.

COP provides a focal point through which NOAA, together with other organizations with
responsibilities for the coastal environment and its resources, can make significant strides toward
finding solutions to critical problems. By working together toward these solutions, we can
ensure the sustainability of these coastal resources and allow for compatible economic
development that will enhance the well-being of the Nation now and in future generations. The
goals of the program parallel those of the NOAA Strategic Plan.

A specific objective of COP is to provide the highest quality scientific information to coastal
managers in time for critical decision making and in a format useful for these decisions. To help
achieve this, COP inaugurated a program of developing documents that would synthesize
information on issues that were of high priority to coastal managers. A three-step process was
used to develop such documents: 1) to compile a list of critical topics in the coastal ocean
through a survey of coastal resource managers and to prioritize and select those suitable for the
document series through the use of a panel of multidisciplinary technical experts; 2) to, solicit
proposals to do research on these topics and select principal investigators through a rigorous
peer-review process; and 3) to develop peer-reviewed documents based on the winning
proposals. Seven topics were selected in the initial round, but the series is expanding because of
the suitability of findings from other COP-funded research to appear in this synthesis format.
The documents already published are listed on the inside back cover.

As with all of its products, COP is very interested in ascertaining the utility of the Decision
Analysis Series particularly in regard to its application to the management decision process.
Therefore, we encourage you to write, fax, call, or E-mail us with your comments. Please be
assured that we will appreciate these comments, either positive or negative, and that they will
help us direct our future efforts. Our address and telephone and fax numbers are on'the inside
front cover. My Internet address is [email protected].

Donald Scavia
Director
@W_@ __

NOAA Coastal Ocean Program

Table of Contents

ACKNOWLEDGMENTS xiii

EXECUTIVE SUMMARY xv
Habitat Requirements and Effects of Logging xv
Objectives of Forest Practices Regulations xvii
Analysis of Current Forest Practices xviii
Habitat Restoration xx
Conclusions and Recommendations xx

CHAPTER 1: INTRODUCTION 1

CHAPTER 2: METHODS 3

CHAPTER 3: HISTORICAL BACKGROUND 5
The Decline of Anadromous Salmonids 5
Legacy of Past Logging 9
Present-day Timberlands 10

CHAPTER 4: SALMONID HABITAT REQUIREMENTS 13
Upstream Migration of Adults 14
Spawning 15
Incubation 16
Freshwater Rearing 18
Food availability
Temperature 21
Streamf low 22
Dissolved oxygen and turbidity 22
Channel morphology 23
Cover 24
Seaward Migration 26
Limiting Factors 27

iv Contents

CHAPTER 5: THE POTENTIAL EFFECTS OF LOGGING 29
Stream Temperature 33
Sediment 35
Streamf low 42
Large Woody Debris 43
Migration Habitat 45
Food Availability 46
Cumulative Effects 48
Salvage Logging 52

CHAPTER 6: TECHNICAL FOUNDATION OF FOREST PRACTICES 55
Buffer Zones 55
Buff er design 58
Factors affecting buffer effectiveness 59
Best Management Practices 64
BMP programs 65
BMP monitoring 65
Cumulative Effects Management 67
Watershed analysis 68
Habitat conservation plans 69

CHAPTER 7: CURRENT FOREST PRACTICES 73
Federall-ands 73
Northwest Forest Plan and PACFISH 73
Tongass Land Management Plan 76
State and Private Lands 76
Alaska 76
California 78
Idaho 80
Oregon 82
Washington 86
Other States 88

CHAPTER 8: ANALYSIS OF CURRENT FOREST PRACTICES 89
Buff er Designs 89
Buffer Effectiveness 92
Shade protection 93
LWD recruitment 93
Sediment control 98
Best Management Practices 99
Cumulative Effects Management 100

Contents v

CHAPTER 9: HABITAT RESTORATION 103
Restoration Procedures 103
Upland restoration 104'
Riparian restoration 104
Instrearn restoration 107
Restoration planning and evaluation ill
Current Restoration Programs 113
Restoration on national forests 113
California's habitat restoration program 114
Oregon's basal area credits 115
The Role of Restoration in Watershed Management 117

CHAPTER 10: CONCLUSIONS AND RECOMMENDATIONS 119
Buff er Zones 119
Best Management Practices 120
Watershed Analysis 121
Restoration 121
Community Outreach 122

GLOSSARY 123

ACRONYMS 129

REFERENCES 131

List of Figures and Tables

Figures

Figure ES.1. Major elements of a strategy for comprehensive,
ecosystem-based watershed management. xvi
Figure 3.1. Distribution of native anadromous salm onids in North
America. 5
'Figure 3.3. Diagram of a watershed's drainage network, showing
stream orders according to the Strahler (1957) classification
system. 8
Figure 3.4. Cross-section of a woodland stream, showing
association with riparian vegetation and flood plain. 8
Figure 3.6. Lumber production in states within the range of
Pacific anadromous salmonids. 11

Figure 3.7. Ownership of timberlands in the Pacific Northwest
and Alaska. 12

Figure 4.1. Generalized life cycle of anadromous salmonids,
involving several distinct freshwater life stages and a period
of ocean residence. 13

Figure 4.2. Egg-to-fry survival in relatio.n to fine sediment
(<2-6.4 mm) in spawning gravel. 17
Figure 4.3. Lethal threshold temperature for juvenile salmonids in
relation to acclimation temperature. 21
Figure 4.4. Formation of various pool habitats by large woody
debris. 24
Figure 4.5. Winter density of coho salmon in relation to large
woody debris in southeast Alaska. 25

Eigures vii

Figure 4.6. Habitat use by juvenile steelhead in winter in relation
to embeddedness of the streambed by fine sediment. 25
Figure 4.7. Illustration of "bottlenecks" restricting salmonid smolt
production. 27
Figure 5.1. Connections between timber harvest, road
construction, and other timber management activities in a
watershed and effects on salmonid habitat via changes in
watershed processes and structures. 30
Figure 5.2. Typical daily stream temperature in clearcut and
forested streams d,uring clear weather in Oregon's Coast
Range. 83
Figure 5.3. Comparison of the rate of landsliding (A) and mass
soil erosion (B) associated with forests, clearcuts, and roads
as measured by six long-term studies of more than 10
years. 36
Figure 5.4. Surface runoff being intercepted by a logging road
and flowing down the road's ditch after heavy rain in
southeast Alaska. 37

Figure 5.5. The failure probability of culverts designed for 25-
year and 1 00-year flood events. 37
Figure 5.6. Fine sediment produced by logging truck in southeast
Alaska. 38

Figure 5.7. Soil disturbance caused by different yarding methods.
Each point represents findings of a separate study. 39
Figure 5.8. Diagram showing storage of bedload sediment
""wedges" behind large woody debris. 40
Figure 5.9. Disturbance and loss of channel structures caused
downcutting and export of stored sediment from a
headwater stream in southeast Alaska. 41

Figure 5.10. Diminishing increase in water yield after timber
harvest in southwest Oregon. 42
Figure 5.11. Change in August water yield from a western
Oregon watershed after logging. 43

viii Figures

Figure 5.12. A model of the changes in large woody debris after
clearcut logging without a riparian buffer zone in
southeastern Alaska. 45

Figure 5.13. Changes in pool frequency between 1935 and 1992
in Oregon and Washington. 46
Figure 5.14. Relationship between density of coho salmon fry
and algae biomass in old-growth, buffered, and clearcut
reaches of streams in southeast Alaska in summer. 47
Figure 5.15. Fine sediment derived from upstream timber harvest
being carried by a headwater stream into downstream fish
habitat in foreground. 49
Figure 5.16. The relationship between amount of fine sediment in
pools and level of timber harvest and road construction in
several northern California watersheds. 49
Figure 5.17. The relationship between peak strearnflow and the
area of a watershed affected by roads and soil compaction
in western Oregon. 50
Figure 5.18. Frequency of pools associated with-large woody
debris in ten Oregon coastal streams with different levels of
logging. 50
Figure 5.19. A stream in southeast Alaska 15 years after timber
harvest without a buffer zone, showing extreme reduction in
pools and habitat complexity due to loss of large woody
debris. 51

Figure 5.20. Salmonid species diversity (inverse of the Berger-
Parker index) in relation to level of timber harvest (low =
< 25% logged; high = > 25% logged) in 14 coastal Oregon
watersheds of different size, rock type, and geographic
location. 52

Figure 6.1. A riparian buffer zone along an anadromous fish
stream in the Tongass National Forest, southeast Alaska. 55
Figure 6.2. The cumulative effectiveness of various functions of
riparian vegetation in relationto distance from the
streambank in western Oregon. 57

Figures ix

Figure .6.3. The cumulative effectiveness of various functions of
forest vegetation in relation to distance from the edge of
adjacent clearcuts in western Oregon. 58
Figure 6.4. The relationship between canopy density and buffer
width (uncut buffers) in western Oregoh. 60
Figure 6.5. Source distance for large woody debr is (LWD) in
southeast Alaska. 61

Figure 6.6. The diameter of stable large woody debris as a
function of channel width. 62
Figure 8.1. Predicted sources of conifer LWD in buffer zones at
mid-rotation -for representative fish-bearing streams as a
percentage of LWD sources present in mature conifer
stands. 97

Figure 9.1. Unused and unneeded roads should be "put to bed"
by removing culverts and outsloping road surfaces to drain
properly without maintenance. 105
Figure 9.2. Increases in number of outmigrant coho salmon
smolts after experimental addition of large woody debris to
debris-poor streams in the Alsea River basin (A) and
Nestucca R. basin (B) in western Oregon. 109
Figure 9.3. Diagram showing variety of instrearn projects that
shape and stabilize the stream channel, store sediment,
create pools, dissipate stream energy, and provide diverse
habitat for either spawning or rearing. 110
Figure 9.5. Distribution of nearly 3,000 restoration projects
administered by California Department of Fish and Game
done to control erosion, improve fish passage, stabilize
stream banks, and improve instrearn habitat. 115

x Tables

Tables

Table 4. 1. Average redd area and area per spawning pair. 16
Table 4.2. Number of*coho salmon at different-fite stages in.
Sashin Creek, Alaska, for two brood years with a large
difference in spawner escapement. 18
Table 5.1. Influences of timber harvest on physical
characteristics of stream environments, potential changesin
habitat quality, and resultant consequences for salmonid
growth and survival. 31
Table 7.1. California's requirements for width of Watercourse
and Lake Protection Zones (WLPZs) by slope class and
stream class. 79

Table 7.2. Idaho's requirements for width of Stream Protection
Zones (SPZs) and leave trees within SPZs per 1,000 ft (304
m) along each side of streams by diameter breast height
(dbh). 81
Table 7.3. Oregon's requirements for width (in ft) of riparian
management areas by stream size. 83
Table 7.4. Oregon's minimum requirements for conifer leave
trees [in number per 1,000 ft (304 m) along each side of the
stream] in riparian management areas by stream size. 84
Table 7.5. Oregon's standard targets for conifer basal area in
riparian management areas each side of Type F streams by
stream size class in selected regions. 85
Table 7.6. Washington's requirements for width and leave trees
per 1,000 ft (304 m) in Riparian Management Zones (RMZs)
in western Washington under the 1992 forest practices
rules. 87

Table 8.1. Definitions of stream classes for federal and private
lands in five states. 90

Table 8.2. Width requirements (in feet) for riparian buff er zones
on federal [Northwest Forest,Plan (NFP) and Tongass Land
Management Plan (TLMP) as amended by the Tongass

Tables A

Timber Reform Act] and private lands in five states for the
three common stream classes. 91
Table 8.3. Required leave trees [per 1,000 ft (304 m), each side
of stream] within riparian buffer zones on federal and private
lands in five states. 94

Table 8.4. Comparison of minimum level of protection for conifer
MID sources for representative anadromous fish streams in
federal (NFP) and private lands in five states. 96

Acknowledgments

The information in this synthesis reflects the combined input from many professionals in the
field of forestry/fisheries interactions. Much was gained through discussions with Chip Andrus,
Rowan Baker, Jerry Bames, Rich Bettis, Cara- Berman, Bob Bilby, Pete Bisson, Jerry Boberg,
Buck Bryant, Lisa Burton, Pete Cafferata, Jeff Cederholm, Sam Chan, Jim Colla, Tim Curtis,
Lowell Diller, Scott Downie, Richard Everett, Elizabeth Gaar, Lynn Hood, Sharon Kramer,
Stan Gregory, Rick Harris, Brian Hoelscher, Jim Hopelain, Bob House, Hiram Li, Jeff
Lockwood, Ken McDonald, Tom Merritt, Wayne Minshall, Tharon O'Dell, Pete Owston,
Gordon Reeves, Tom Robison, Terry Roelofs, Jeff Schinike, Jeanette Smith, Mario Solazzi, Jim
Steele, and Russ Strach.

Special thanks are due to many people and companies for providing tours of forest practices and
restoration:

In Alaska: Thanks to Frank Rue and Lana Shea, Alaska Department of Fish and Game, and to
Rick Harris of Sealaska Corporation for inviting my participation on field inspections of forest
practices on Sealaska lands. Thanks to Marty Wellboum and Bruce Johnson of Alaska
Department of Natural Resources for affanging for field discussions with Area Foresters.

In California: Thanks to Jeff Schimke of the California Department of Forestry and Fire
Protection for leading a tour of forest practices issues on private lands in Califomia. Thanks
to Rich Bettis and Henry Alden of The Pacific Lumber Company, and Tharon O'Dell and
Lowell Diller of Simpson Timber Company. Thanks to Jim Hopelain and Scott Downie of
California Department of Fish and Game and Rich Bettis for an informative tour of watershed
restoration projects on Pacific Lumber Company lands. Thanks to Jerry Bames, USDA Forest
Service (FS), for explaining stream restoration projects on the Six Rivers National Forest and
providing several,photographs of restoration work.

In Idaho: Thanks to Jim Colla for arranging a field tour of forest practices on private lands,
which unfortunately had to be cancelled because of forest fire.

In Oregon: Thanks to Fred Robinson, Ted Lorensen, and others of the Oregon Department of
Forestry for arranging tours of forestry practices and restoration programs on private lands in
western Oregon. Thanks to Sam Chan and Dan Majkowski, FS Pacific Northwest Research
Station, for demonstrating research on riparian restoration in coastal Oregon. Thanks to Mario
Solazzi, Oregon Department of Fish and Wildlife, for showing research on stream restoration

Xiv Acknowledgments

in western Oregon. Thanks to Lynn Hood and Lisa Burton, FS, for explaining ongoing stream
restoration work. Thanks to Willamette Industries, Lone Rock Timber Company, Starker
Forests, and Weyerhaeuser Company for providing tours of forestry practices, restoration
projects, and monitoring activities on their lands.

In Washington: Thanks to Pete Bisson of Weyerhaeuser Company for arranging a tour of
logging practices and restoration projects in the Tolt River watershed. Thanks to Jeff
Cederholm, Washington Department of Natural Resources, for showing ongoing experiments
in stream restoration.

Sharon Kramer, K Koski, Jeff Lockwood, and Matt Longenbaugh reviewed an early draft and
gave helpful comments. I also appreciate thorough reviews by Rowan Baker, Pete Bisson,
Tamra Faris, Bill Peterson, Gordon Reeves, Lawrence Six, and Laurie Sullivan. K Koski
conceived of,the idea for this project and guided it through its early development. Funds for
this project were provided by the NOAA Coastal Ocean Program.

Executive Summary

This synthesis presents a science overview of the major forest management issues involved in
the recovery of anadromous salmonids affected by timber harvest in the Pacific Northwest and
Alaska. The issues involve the components of ecosystem-based watershed management and how
best to implement them, including how to:

+ Design buffer zones to protect fish habitat while enabling economic timber production;'

* Implement effective Best Management Practices (BMPs) to prevent nonpoint-source
pollution;

* Develop watershed-level procedures across property boundaries to prevent cumulative
impacts;

* Develop restoration procedures to contribute to recovery of ecosystem processes; and

* Enlist support of private landowners in watershed planning, protection, and restoration.

Buffer zones, BMPs, cumulative impact prevention, and restoration are essential elements of
what must be a comprehensive approach to habitat protection and restoration applied at the
watershed level within a larger context of resource concerns in the river basin, species status
under the Endangered Species Act (ESA), and regional environmental and economic issues (Fig.
ESA).

This synthesis 1) reviews salmonid habitat requirements and potential effects of logging; 2)
describes the technical foundation of forest practices and restoration; 3) analyzes current federal
and non-federal forest practices; and 4) recommends required elements of comprehensive
watershed management for recovery of anadromous salmonids.

HABITAT REQUIREMENTS AND EFFECTS OF LOGGING

The life cycle of anadromous salmonids has several stages, each with its own habitat needs.
Among other things, adults returning to spawn require access to spawning gravel; their eggs
need cool, oxygenated water; juveniles need adequate food, cover, and temperature; and smolts

xvi Executive Surnmarv

Ecosystem-Based
Watershed Management

Habitat Habitat
Protection F_ Restoration

Forest Practices Restoration Procedures
Best Management Practices Uplands
Buffer Zones Riparian
Cumulative Effects Management Instrearn

Laraer,Context
River Basin Condition
Species Status
Regional Issues

Conservation and Recovery
of Anadromous Salmonids

Figure ES. 1. Major elements of a strategy for comprehensive, ecosystem-based
watershed management.

migrating to sea need adequate streamflow. Because of these diverse habitat requirements,
salmonid streams must provide appropriate diverse habitat conditions from the headwaters to the
estuary.

Freshwater habitats for anadromous salmonids are created by physical and biological processes
affecting the flow of water, sediment, nutrients, and organic matter through the watershed,
modified by features such as large woody debris (LWD), and periodically "reset" by natural
disturbances.

Small streams are the "backbone" of salmonid habitat. Even when not used because of barriers
or steep gradient, small, even intermittent streams are critical to downstream fish habitats
I

because they transport water, sediment, and woody debris from the upper watershed.
Intermittent stream channels account for over one-half of the total length of stream channels in

Executive Summary Xvii

many watersheds. Small streams are easily affected by logging and other land uses because they
are sensitive to changes in riparian vegetation and condition of the surrounding watershed.
Forest practices that alter erosion, runoff, or riparian vegetation can have major impacts on
small streams and the rivers into which they flow.

Impacts from over 100 years of logging and other land uses are still evident in streams of the
Pacific Northwest and other regions. The most pervasive effect has been reduced habitat
complexity due to loss of LWD, causing a widespread reduction in salmonid abundance and
diversity. Despite improvements over the last 20 years, logging activities can still have multiple
impacts. Effects of timber harvest, road construction, and other activities anywhere in a
watershed can be transmitted through hydrologic and erosional processes to affect salmonid
habitat. The most important impacts result from changes in sediment, strearnflow, temperature,
and LWD.

OBJECTIVES OF FOREST PRACTICES REGULATIONS

Forest practices must be designed to protect fish and wildlife habitat while enabling economic
timber production. Agencies regulate forest practices through buffer zones, Best Management
Practices (BMPs), and cumulative effects management.

Buffer zones are administratively defined areas along streams or erosion hazard areas in which
aquatic resources are given highest management priority. The function of buffer zones is to
protect streams and riparian areas from disturbance; filter sediment from uplands; and supply
food, cover, shade, and LWD. Regulations determine both the width of buffers and activities
within them. Buffer zones are not necessarily "lock-out" zones; trees often can be harvested,
but with restrictions to protect aquatic resources. Restrictions are generally tighter on public
than on private lands. Under the federal Northwest Forest Plan, for example, buffers can be
modified only if watershed analysis demonstrates that a modification is needed to attain
ecosystem management objectives (USDA and USDI 1994a).

The appropriate design for buffer zones depends on management objectives. The widest buffers
with greatest restrictions on activities are used along fish-bearing streams to meet a full range
of objectives for fish habitat, as well as for other wildlife (e.g., owls and amphibians) (USDA
and USDI 1994a). Leaving large conifers in a sufficiently wide buffer (at least as wide as the
height of a mature tree) is particularly important for providing LWD for fish-bearing streams.
Narrower, selectively harvested buffers can be used along non-fish streams specifically to protect
water quality and prevent downstream impacts.

BMPs are specific rules (e.g., waterba'rring skid trails), to prevent nonpoint-source pollution,
particularly from fine sediment. The Clean Water Act gives states authority to certify their
forest practices rules as approved BMPs, and to certify BMPs of federal agencies for streams
under federal jurisdiction. States with regulatory BMP programs impose requirements on forest
practices and assess penalties for noncompliance.

Xviii Executive Summary

Monitoring is conducted primarily by federal and state agencies. Implementation monitoring to
determine whether BMPs are applied as specified is the most common type of monitoring.
Effectiveness monitoring to determine whether BMPs achieve their intent is important in
improving BMP performance. Comprehensive monitoring programs should also determine
whether habitat problems are being recognized and appropriate practices specified, determine
whether the combined system of BMPs protects water quality for particular projects, and provide
for public review for improving BMPs.

Recent assessments indicate that forestry BMPs can protect water quality if they are carefully
developed and implemented (Brown and Binkley 1994). Current problems often result from
poor BMP implementation, which is generally worse on small private parcels than on public or
large industrial holdings. Most state BMPs, furthermore, do not carefully protect small non-fish
streams, and BMPs for protecting unstable slopes still need to be developed.

Cumulative effects managgment is a form of planning for preventing impacts from nonpoint-
source pollution that could be overlooked at the project level. Evaluations of potential
cumulative effects consider watershed erosion potential, slope stability, current and past
disturbances from timber harvest and other land uses (e.g., grazing, agriculture, mining),
recovery rate after disturbance, and project area relative . to the total watershed. The most
comprehensive procedure for analyzing cumulative effects at the watershed scale is "watershed
analysis" (Washington Forest Practices Board 1993), a systematic process to describe current
watershed conditions and develop prescriptions to prevent undesirable cumulative impacts. The
assumption is that undesirable cumulative impacts can be avoided by managing sensitive areas
appropriately and applying standard practices in non-sensitive areas.

Planning at the basin, regional, and even larger scales is also necessary for managing cumulative
effects on anadromous salmonids because of their wide-ranging migrations. The NMFS
Proposed Recovery Plan for Snake River Salmon (USDC 1995) is a good example of the
comprehensive planning required to address all potential factors that cumulatively affect salmon
populations. In this plan, watershed uses, including timber harvest, are just one of five planning
components that also include main-stem and estuarine habitat, fisheries harvest management,
hatchery propagation, and changes in institutional structure to improve decision making. These
other four components are beyond the scope of this synthesis, but forestry-fisheries issues should
properly be considered in this context.

ANALYSIS OF CURRENT FOREST PRACTICES

Federal land management agencies and the five western states with.anadromous salmonids
(Alaska, California, Idaho, Oregon, and Washington) have recently revised their rules to
increase habitat protection. On federal lands, principal direction is given by one of three
sources: the Northwest Forest Plan (NFP) within the range of the northern spotted owl (USDA
and USDI 1994a); PACHSH, an interim strategy until Environmental Impact Statements can be
completed for non-NFP areas (USDA and USDI, 1994b); and the Tongass Land Management
Plan (TLMP) as supplemented by the Tongass Timber Reform Act (TTRA) in Alaska. On

Executive -Summ"r Ax

private lands, forest practices follow their state's administrative rules. All these programs
provide examples of "state-of-the-art" management for protection of salmonid habitat.

Forest 'management under NFP1 PACFISH, TLMP, and the five states have many common
elements, including buffer zones and regulatory BMPs. They differ mainly in how they manage
buffer zones and cumulative effects.. All classify streams by "beneficial use" (i.e., fish streams
vs. non-fish streams) and give fish-bearing streams more protection through buffers and BMPs
than small non-fish streams. Buffers along anadromous fish streams range in minimum width
from 25 ft (8 m) on private lands in Washington to 300 ft (91 m) on federal NFP and PACFISH
lands. Buffer width for perennial non-fish streams can range from 0 ft in Alaska, Oregon, and
Washington to 150 ft (46 m) under NFP and PACFISH. Intermittent stream channels routinely
have buffers only on NFP and PACFISH lands and in Idaho, but some states may use buffers
when warranted by site conditions.

For fish-bearing streams, harvest restrictions within buffers are designed to protect most riparian
functions, particularly shade, channel stability, and LWD. Four of the five states require a
specific number of "leave trees" in combination with other vegetation requirements for LWD
sources. These requirements, however, do not fully provide for future LWD sources, and result
in leaving only an estimated 23% to 58% of potential conifer LWD sources compared.to the
sources present in mature forest.

Non-fish streams and intermittent channels are managed primarily to prevent sediment pollution
and downstream impacts. Buffers on these stream channels on private lands are often narrow
and heavily harvested, and stream protection relies heavily on BMPs. Three BMPs are
particularly important in protecting small non-fish streams from disturbance. They determine
1) whether trees may be felled into stream channels and limbed there, 2) whether cable yarding
may, cross streams with full or partial log suspension, and 3) whether tractors may operate within
streams or their buffer zones.

The. states' BMPs for these activities carefully protect fish-bearing streams, but small non-fish
streams are not as carefully protected. All states require that trees be felled away from fish-
bearing streams, but several allow felling into small non-fish streams. Cable yarding across fish-
bearing streams must have full suspension and prior approval in Oregon and Washington, but
not for small non-fish streams. Tractor yarding is not allowed across fish-bearing streams except
at constructed temporary crossings, but the states do allow tractors ir. some intermittent non-fish
streams.

The agencies differ in how they assess and manage potential cumulative effects. Watershed
analysis is a major component of the strategy for preventing cumulative effects on federal lands
under NFP and PACFISH, but not currently under TLMR California, Idaho, and Washington
have a formal process for I evaluating cumulative effects, but Alaska and Oregon do not.
Applying watershed Analysis on private lands is hindered because of the difficulty in coordinating
resource assessment and management across property boundaries.

XX Executive Summary

HABITAT RESTORATION

Habitat restoration has an important role in the recovery of anadromous salmonids as one
element in a comprehensive program of watershed management emphasizing habitat protection.
Habitat restoration is used to stabilize deteriorating conditions and speed recovery in key
watersheds. Restoration should be regarded as an interim measure until degraded watersheds
recover under effective management, not as an exemption from stream protection. Before
initiating restoration, land uses that have caused the degradation must be modified to end adverse
effects.

Effective habitat restoration has a watershed perspective based on hydrologic principles and is
preceded by careful analysis to assess habitat problems and evaluate restoration potential.
Habitat restoration includes three components: 1) upland restoration to control erosion, stabilize
roads, upgrade culverts, and manage watershed uses; 2) riparian restoration to modify riparian
vegetation to provide shade, LWD recruitment, and other functions; and 3) instream restoration
using boulders, LWD, or other structures to provide missing habitat features and increase habitat
'complexity. ' I

The need for habitat restoration is great. More than two-thirds of the riparian areas and one-half
of all streams in the Pacific Northwest are degraded. Restoration of key watersheds, those with
the best remaining habitat or greatest restoration potential, comprising one-third of federal lands
in Oregon, Washington, and northern California, would cost $720 million over 10
,,_years (Pacific
Rivers Council 1993a). Although restoration costs are high, the investment return would be
considerable because it would generate many jobs, and the recovery of salmon and watershed
functions would have many social and economic benefits. Considering the costs, habitat
protection is obviously preferable to allowing habitat to degrade to the point of needing
restoration.

CONCLUSIONS AND RECOMMENDATIONS

A comprehensive watershed-level approach to habitat protection and restoration is essential for
maintaining or restoring salmonid habitat because the watershed is a fundamental unit for both
ecological processes and land management. The main technical elements of watershed
management are buffer zones, BMPs, watershed analysis, and restoration.

To maintain or restore optimal habitat in fish-bearing streams, buffer zones should be at least
as wide as the height of a mature tree, usually 30-40 m, and be managed to attain characteristics
of mature native forest. Narrower buffers may not maintain adequate LWD over the long term,
and selective harvest within buffers further reduces LWD sources. No-harvest buffers are most
appropriate along fish-bearing streams - with mature forest, most common in Alaska and in
national forests. On private lands in other states, the number and size of leave trees should be
increased where additional large conifers are available.

Many previously logged areas have degraded vegetation consisting mostly of hardwoods and
brush and lacking large conifers. Restricting harvest would not necessarily improve habitat

Executive Summarv Xxi

protection nor help restore riparian functions. Active management of these riparian areas is
needed to meet habitat requirements of fish. Selective harvest within these buffers could be used
to improve riparian vegetation (i.e., by thinning and conifer planting). Forest practices rules
can include incentives for timber operators to actively. manage degraded riparian stands to
reestablish mature conifers or other appropriate vegetation. Reestablishing conifer forest in
riparian areas would benefit both fisheries and timber because trees could be selectively
harvested where shown not to harm fish habitat

Buffeizones are also needed along small non-fish streams that affect salmonid habitat. The
usually minimal buffers on these streams on private lands means that their protection must rely
on BMPs which do not always protect them from disturbance. Monitoring studies have not yet
shown that BMPs for non-fish streams are effective in preventing downstream impacts. Buffer
width and harvest prescriptions for these areas can be developed specifically to protect headwater
sources of temperature control, sediment, and debris for downstream fish habitat.

The BMPs for activities near small non-fish streams need to be closely monitored to ensure they
are effective. This is essential because small non-fish streams are particularly important for
preventing sediment pollution and because buffer zones along them on private lands are usually
narrow and heavily -harvested. Effective BMPs may be the only practical means of protecting
the numerous non-fish headwater streams in managed timberlands while other activities continue.,

Watershed analysis is an important tool for assessing cumulative effects. In mixed-ownership
watersheds, agencies can organize and lead landowners in cooperative watershed management
across property boundaries. Ultimately, basin-wide planning efforts are needed that include all
public and private land managers.

Habitat restoration should have a watershed-level approach and include measures to control
erosion, reestablish riparian conifers, and improve instream structure. A priority is to stabilize
existing roads to control erosion. Instream projects should be used only as part of a
comprehensive watershed management program. Due to limited ftinds, most degraded habitat
must rely on slow natural recovery under effective management. Because of the current
depressed condition of many salmonid stocks, as much as possible should be done to speed
recovery in key watersheds. The goal is to secure, expand, and link key watersheds in a system
of refugia connected by intact migration corridors.

Habitat restoration is not a panacea for recovery of anadromous salmonids. There must also be
changes in land and water uses to improve habitat protection and changes in fisheries
management to ensure sufficient escapement. Habitat restoration and protection, however, are
critical because even with fisheries closures, depressed stocks cannot recover without habitat.

Any conservation strategy will probably fail without. community support. Comprehensive
watershed management must also include outreach programs to recruit support from landowners
and local communities. Tax credits and cost-sharing programs can be expanded to compensate
landowners for measures taken to protect public aquatic resources.

xxii Executive Summary

The focus of this synthesis is on effects of forest management activities on anadromous fish
habitat, but many land and water uses besides forestry have contributed to the decline of
anadromous; salmonids and therefore must also contribute to their recovery (USDC 1995).
Improving forest practices and restoring fish habitat will not, by themselves, guarantee recovery
of anadromous salmonids. However, these things are needed if the populations are to recover.

Chapter 1
Introduction

Across much of their range in the Pacific Northwest, anadromous salmonids have declined to
the point that many stocks are depleted, federally listed as threatened or endangered, or extinct.
Once-productive fisheries have been drastically curtailed or closed. Although habitat damage
from timber harvest is not the only cause, it is an important factor in the decline of many stocks
(Nehlsen et al. 1991; Botkin et al. 1994). To help reverse this decline, land managers have
come to recognize the need for increased protection of stream habitat in areas managed for
timber production and increased efforts to restore streams degraded by past timber harvest.

Several trends converged during the last two decades to focus concern on protecting and
restoring fish habitats affected by timber management. The great value of fish and wildlife has
become apparent, and the public demand has increased for recreational use of the forest (Meehan
1991). The listing of salmonid stocks and other wildlife species under the Endangered Species
Act (ESA), with its strong regulatory measures, has brought many forest managers to see the
need for a new approach to managing land and water resources. Forest managers of today are
moving away from maximizing timber production by high-yield forestry toward managing
resource complexes including fish (Meehan 1991). "Ecosystem management" is becoming the
forest manager's paradigm for providing long-term maintenance of multi-species biological
communities (Franklin 1992; Reeves and Sedell 1992; FEMAT 1993).

Habitat protection and restoration are two key elements in the recovery of anadromous salmonid
stocks. Protection of fish habitat should be among every land manager's goals because fish
habitat is influenced by uses throughout entire watersheds. Restoration is considered a "band-
aid" approach to bridge the interim until habitat recovers enough under good watershed
management to contribute to the recovery and sustained natural reproduction of anadromous
salmonids. Without restoration, natural recovery of many impaired fish habitats would take
decades or centuries (Rhodes and McCullough, in press); salmonid stocks near extinction may
not survive that long. Protection of good existing habitats should have the highest priority, but
many streams have been damaged and need to be restored (Meehan 1991).

Fortunately, much scientific knowledge is available to help guide resource managers. Salmonid
habitat requirements and the effects of timber harvest on fish habitat are basically understood,
and the fundamentals of restoration are known. Enough is known to implement land-use
practices that prevent further habitat degradation (Chamberlin et al. 1991) and to begin restoring
habitats previously degraded (Koski 1992).

2 Introduction

In the past, habitat restoration and forest practices were generally treated as separate activities
and not related to the watershed as a whole. In the modem paradigm of ecosystem management,
however, they are recognized as essential elements of what must be a comprehensive approach
to habitat protection and restoration applied at the watershed level (FEMAT 1993; Jensen and
Bourgeron 1994). This approach is necessary to account for the way physical and biological
processes function and interact in watersheds. A comprehensive watershed-level approach to
habitat protection and restoration is essential for maintaining or restoring salmonid habitat.

Numerous publications have reviewed separate issues, such as the effects of timber harvest on
fish habitat (e.g., Macdonald et al. 1998; Hicks et al. 1991a), buffer zones (e.g., Belt et al.
1992; Johnson and Ryba 1992), and habitat restoration (Reeves et al. 1991; Koski 1992).
However, there are few analyses of all the elements essential to a watershed-level program of
forest practices and restoration.

The purpose of this synthesis is to provide an overview of the important management issues
involved in a watershed approach to forest practices and habitat restoration relating to protection
and recovery of anadromous salmonids. Specific objectives are to

1) review habitat requirements of anadromous salmonids and potential effects of logging on
salmonid habitat;

2) describe the function and technical foundation of forest management practices and fish habitat
restoration;

3) analyze current forest management practices as examples of "state-of-the-art" watershed
management strategies; and

4) recommend required elements of comprehensive watershed management for protection and
recovery of anadromous salmonids.

The geographic focus of this synthesis is on forest practices within the range of anadromous
salmonids in the Pacific Northwest and Alaska, but the principles also apply to other areas where
timber harvest and other land uses affect aquatic habitats.

Chapter 2
Methods

Information for this synthesis was gathered primarily from review of the, literature (published
and unpublished) and consultations, with scientific and technical experts in the forestry and
fisheries fields.

Several interdisciplinary site visits were used to review current forest practices and restoration
programs in California, Oregon, Washington, and Alaska. Emphasis of these.visits was on
viewing activities on private lands because forest practices on private lands vary among the
states, whereas practices on federal lands are more consistent across the region.

In California, a 2-day tour of ongoing forestry activities was conducted by the California
Department of Forestry and Fire Protection (CDF). This tour visited several industrial timber
operat ions, as well as smaller non-industrial landowners. Focus of the tour was on current forest
practices regulations and their administration by the CDE Site visits to habitat restoration
activities on private lands were provided by the California Department of Fish and Game
(CDFG). Tours of habitat restoration projects on the Six Rivers National Forest were organized
by the USDA Forest Service (FS).

In Oregon, the Oregon Department of Forestry (ODF) provided a 2-day tour offorestry
activities on private industrial timberlands. Site visits were also used to review restoration
research programs conducted by the Oregon Department of Fish and Wildlife (ODFW), Oregon
State University, and the FS Pacific Northwest Research Station. Ongoing research programs
are investigating both instream and riparian restoration activities.

In Washington, forest practices and restoration on private lands were reviewed during a site visit
conducted by Weyerhaeuser Company. Restoration research conducted by the Washington
Department of Natural Resources (WDNR) was also reviewed during a site visit.

In Alaska, reviews of forest practices on private lands were jointly conducted by the Alaska
Department of Fish and Game (ADFG) and the Department of Natural Resources. Other
information on forest practices and related research was provided by the Alaska Working Group
on Cooperative Forestry/Fisheries Research.

Chapter 3
Historical Background

THE DECLINE OF ANADROMOUS SALIVIONIDS

In North America, the native anadromous salmonids occur from mid-California to the Arctic
Ocean in the west, and from Connecticut to northern Newfoundland in the east (Fig. 3.1;
Meehan and Bjornn 1991). In the west, anadromous salmonids occur in five states: California,
Oregon, Washington, Idaho, and Alaska. Anadromous species include five species of Pacific
salmon (Oncorhynchus spp.), steelhead (Oncorhynchus mykiss), sea-run cutthroat trout (0.
clarki), and Dolly Varden (Salvelinus malma). In the east, only Atlantic salmon (Salmo salar)
and to some degree brook trout (Salvelinus fontinalis) are anadromous. Non-anadromous
populations of salmonids occur throughout the U.S. and Canada.

Atlantic
Pacific
Salmon
Salmon

Figure 3.1. Distribution of native anadromous salmonids in North America. (After
Meehan and Bjornn 1991.)

6 Historical Background

Because of their fidelity in homing to natal streams in which they were spawned and reared,
anadromous, salmonids have become reproductively isolated into hundreds, of locally adapted
populations, or stocks (Ricker 1972). Any given stream with appropriate habitat can harbor
several coexisting species and several different stocks of each species (Fig. 3.2).

Stories of.the original abundance of anadromous salmonids in the United States (Netboy 1974)
seem like fiction from today's perspective. With the exception of Alaska, existing populations
of native anadromous salmonids on both coasts are mere remnants. On the East Coast, Atlantic
salmon have been severely depleted for over 100 years. On the West Coast, at least 106 major
stocks are extinct, and another 101 are at high risk of extinction (Nehlsen et al. 1991). In
California, for example, coho salmon (Oncorhynchus kisutch) now occur in only one-half of
historic natal streams (CDFG 1994).

Salmon are still the nation's most
valuable fisheries (U. S. Dept.
Commerce 1990), with anaverage
(1985-89) commercial ex-vessel
value in the northeastern Pacific of
$773 million (Talley 1990). Value
for sport, subsistence, and other non-
market amenities are also substantial
(Huppert and Fight 1991). Because
of the depleted condition of the
stocks, most fisheries off California,
Oregon, and Washington were closed
in 1994. The Pacific Coast
Federation of Fishermen's
Associations estimates that 98% of
the jobs , dependent on salmon
fisheries in the Pacific Northwest
have been lost since 1988, and the
economic impact of lost salmon
fishing is $1.25 billion. Considering
the depleted condition of most native
stocks, the potential value of the
resource if restored would be

enormous.

Three principal factors caused the
decline of anadromous salmonids:
1) loss of habitat due to habitat Figure 3.2. Salmonid habitat in a small
destruction, inadequate passage at forested stream in southeast Alaska, showing
dams, and inadequate streamflow; its complexity and abundance of pools,
2) overfishing of weaker stocks spawning gravels, and cover provided by large
in mixed-stock fisheries; and 3) woody debris. (Photo by K Koski, NMFS.)

Historical Background 7

negative interactions with non-native fish, especially from hatcheries (Nehisen et al. 1991).
Belief that hatcheries could mitigate habitat loss also provided an excuse for taking habitat for
development. Several factors usually operated in concert. For example, the now-extinct run
of coho salmon in Oregon's Grande Ronde River was first reduced by habitat destruction from
logging, grazing, and agriculture; then further reduced by dams on the Snake River in the
1960s-1970s; And finally overfished when managem6nt agencies decided the stock was too weak
to warrant protection in mixed-stock fisheries (Nehlsen et al. 1991). A new paradigm that
advances habitat restoration and ecosystem function rather than hatchery production is needed
for salmonid stocks to survive and prosper (Nehlsen et al. 199-1).

'Habitat destruction from logging, mining, grazing, agriculture, and urban development was the
factor most commonly associated with the decline of anadromous salmonids (Nehlsen et al.
1991). Unregulated clearcut logging damaged numerous streams, rivers, and estuaries (Sedell
et al. 1991). Hydraulic gold mining in California created a giant wave of sediment that
progressed from mountain streams to San Francisco Bay (Ritter 1978). Dredge mining for gold
in the 1800s dug up streambeds in California and Idaho, leaving tailings piles visible today
(McIntosh et al. 1994). Overgrazing by sheep and cattle caused chronic erosion and altered
riparian areas, and Agriculture converted forests to erosion-prone croplands and withdrew water
for irrigation (Wissmar et al. 1994). Construction of housing and highways accompanying
explosive population growth in many areas of the West severely impacted numerous watersheds
(Netboy 1974).

As of August 1995, the National Marine Fisheries Service (NMFS) has listed four stocks as
threatened or endangered under ESA: Snake River spring/summer chinook, Snake River fall
chinook, Snake River sockeye, and Sacramento River winter chinook. NMFS has also proposed
to, list Umpqua River cutthroat trout, Klamath Mountain Province steethead, and three distinct
evolutionarily significant units (ESUs) of coho salmon on the Oregon and California coasts. All
other stocks of anadromous salmonids outside Alaska are undergoing status review.

For the first stocks listed under ESA, concerns included big-river problems with fish passage
and water management in the Columbia River and Sacramento River Basins, as well as fishing
mortality and land management 'activities, especially in Idaho and eastern Oregon and
Washington. The NMFS Proposed Recovery Plan for Snake River Salmon (USDC 1995) has
five major planning areas including tributary ecosystems, main-stem river and estuarine
ecosystems, fisheries harvest management, hatchery propagation, and changes in institutional
structure to improve decision making. As the status of coho salmon, steelhead, and cutthroat
trout came under review, concern expanded to a broader geographic area and more emphasis on
impacts of land uses and hydro projects on small streams (L. Sullivan, NMFS, pers. comm.
1995).

Small streams are the "backbone" of salmonid habitat. Salmonids occupy streams ranging from
tiny first-order tributaries to the main-stem Columbia River (ninth-order; Leopold et al. 1964),
but most spawning and rearing, especially for coho salmon, steelhead, Dolly Varden, and
cutthroat trout, take place in second- to fourth-order streams (Figs. 3. 1 . and 3.3). Large rivers
can provide important spawning, rearing, and migratory habitat for all species, but small streams
greatly outnumber the higher-order rivers. Even when small streams are, not used by salmonids

8 Historical Background

because of barriers or steep gradient, they are important to the quality of downstream habitats
because they carry water, sediment, nutrients, and woody debris from the upper watershed to
downstream reaches.

Small streams are also easily affected by land uses (Meehan 1991). They are intimately
associated with their riparian zones and flood plains (Fig. 3.4), and they are highly responsive
to changes in riparian vegetation and the condition of the surrounding watershed. Land uses that
increase erosion, modify runoff, or alter riparian vegetation have greater effects on small streams
than on larger streams.

First
order 2
2 3 4 3

Fourth 4
order

Figure 3.3. Diagram of a watershed's drainage network, showing' stream orders
according to the Strahler (1957) classification system. First-order streams are
headwater streams without tributaries; second-order streams are formed by the
confluence of two first-order streams; third-order streams are formed by the
confluence of two second-order streams; and so on.

Floodplain Channel I loodplainj

Gravel Bar Right
Bank
Streambed

Figure 3.4. Cross-section of a woodland stream, showing association with riparian
vegetation and flood plain. (From Sullivan et al. 1987; reprinted with permission from
University of Washington, Institute of Forest Resources.)

Historical Backaround 9

LEGACY OF PAST LOGGING

Logging and clearing of land historically impacted vast areas of U.S. forests and their salmonid
habitats. Over I million km' of the original forest has been converted to agriculture and other
uses (Powell et al. 1993). The greatest rate of forest clearing was between 1850 and 1910,
when American fanners cleared 758,000 km', averaging 35 km' per day for 60 years. Logging
began in New England in the 1700s, in the Great Lakes and Gulf Plain in the early 1800s, in
the Pacific Northwest in the mid 1800s (FEMAT 1993), and in Alaska about 1950 (Gibbons et
al. 1987).

In the early years, streams were used to move logs downstream to accumulation sites. Every
stream of sufficient size in western Oregon and Washington was cleared of obstructions for log
drives during high water (Sedell et al. 1991). On streams too small for log drives, splash dams
of log cribbing were used to raise a head of water for sluicing logs (Fig. 3.5; Sedell and
Luchessa 1982). Repeated splash damming caused major long-term damage to fish habitat as
torrents of water and logs severely scoured many streams, leaving barren bedrock. By about
1900, over 300 major splash dams and numerous undocumented smaller dams operated in
Oregon and Washington.

Even where splash dams were not
used, stream channels served as
transportation corridors for logging.
Railroads were built along the larger
drainages, and then logs were yarded
down the smaller tributaries to the
railbed. In this way, impacts
extended to the tiniest intermittent
channels. Whole watersheds were
logged as convenience dictated,
beginning in the lower watershed and
progressing upstream until all
valuable timber was taken. Logs
were yarded downhill, scraping
debris and sediment into stream
channels.

Logging practices after this early
period were improved but still
affected salmonid habitat. Streams
were protected from being used for
yarding in the 1950s. Clearcutting
to the streambank, however, was
normal practice until the 1980s.
Riparian buffer zones were not used
much until the late 1980s, and most Figure 3.5. Finishing up a log drive after
of these buffers contained only splash damming in western Washington in
minimal trees (Phinney et al. 1989). 1927. (Photo courtesy of J. Seddil, FS.)

10 Historical Background

Before regulations limited logging slash entering streams, timber harvest often completely buried
streams in accumulated limbs and tops (Gibbons and Salo 1973). The small unstable debris
often washed downstream within a few years, destabilizing natural debris jams (Bryant 1980).
To address this problem, forest practices rules in the 1970s began requiring removal of logging
slash from streams after timber harvest. Timber operators did their best to thoroughly clean
streams after yarding, often removing every piece of debris from the channel, including
beneficial natural debris (Bilby and Ward 1989).

Lack of knowledge about fish habitat requirements also caused mistakes in stream management.
In response to the days when logging slash was a major problem, fisheries managers from the
1950s to the 1970s focused on removing large woody debris (LWD) from streams. Woody
debris was seen as a detriment to salmon migration, and all along the Pacific coast, logjams
were removed with the intention of opening new reaches of stream (Narver 1971; Hall and
Baker 1982).

Impacts from over 100 years of logging and debris removal are still evident in most streams of
the Pacific Northwest and other areas of the nation (]Ko�ki 1992). One of the most damaging
long-term effects has been a drastic reduction in LWD (Bisson et al. 1987), extending from the
headwaters to the estuary (Sedell et al. 1991). Past logging practices have reduced large, stable
LWD in streams and depleted future LWD sources in riparian zones. Depending on geology
(Hicks 1990), a typical coastal stream in second-growth forests of the Pacific Northwest has
greatly simplified habitat consisting of a single, cobble-bed channel lacking pools and LWD.
In other areas, streams may have bedrock channels lacking gravel, woody debris, and other
channel features (McIntosh et al. 1994).

Today, forest management is evolving to provide greater protection for salmonid habitat.
Improved knowledge of habitat requirements and increased concern for habitat quality by
recreational and commercial fishermen, environmental organizations, and the general public have
led to this greater emphasis in providing for fish habitat in forest management activities (Brouha
1991).

PRESENT-DAY TIMBERLANDS

Forests in the US, now amount to 70% of the area that was forested in 1600 (Powell et al.
1993). Today's landscape, however, generally consists of small patches of forest of mostly
middle seral stages interspersed with recently harvested areas. At lower elevations, forests are
intermingled with farms and towns, and fragmented by highways and residential developments.
Inventories in the 1980s showed only 18% of the original old-growth forest in California,
Oregon, and Washington still existed; 85% was on public land and higher elevation (Bolsinger
and Waddell 1993). Much has been cut since this last inventory. The more valuable forest
types have been the most heavily logged. The ponderosa pine forests in national forests on the
east slope of the Cascade Mountains, for example, have only 2-8 % climax. old growth remaining
(Eastside Forests Scientific Society Panel 1993). Generally, forest land at lower elevations is
privately owned and has been logged at least once.

Historical Background 11

About 20% (2 million kM2) of the nation's total land area is classed as "timberland," capable
of producing more than 20 cubic ft of industrial wood per acre (1.4 m 3/ha) per year and not
withdrawn from timber harvest in parks and other reserves (Powell et al. 1993). Most of the
nation's highly productive forest lands, capable of producing more than 120 cubic ft per acre
(8.4 m'/ha) per year are in the South and in the Pacific Northwest. Areas that produce both
timber and anadromous salmonids coincide over much of western North America and mostly in
Maine in the eastern United States. Composition of U.S. forests is diverse, ranging from pure
stands of Ponderosa pine (Pinus ponderosa) in the semiarid West to multi-species hardwoods in
the Northeast. The most productive timberlands are the loblolly-shortleaf pine forests of the
South and the Douglas fir (Pseudotsuga menziesii) and redwood (Sequot .a sempervirens) forests
of the Pacific Northwest.

U.S. forests within the range of Pacific anadromous salmonids produce about 17 billion board-
feet of sawlogs and other wood products per year-one-quarter of the total U.S. timber
production (Powell et al. 1993; Warren 1993; Fig. 3.6). Private lands and national forests
account for most timber harvest, with smaller contributions from USDI Bureau of Land
Management (BiM) lands in Oregon and state lands in Washington. Most U.S. timber
production, however, comes from the South, outside the range of. anadromous salmonids.
British Columbia, Canada, is also a major producer, about the same as the Pacific Northwest.

These timberlands are owned primarily by the federal goverm-nent and private firms and
individuals (Fig. 3.7). Most federal timberlands are national forests; other public lands include
mostly BLM lands in Oregon and state lands in Washington and Alaska. Private I ,ands are about
equally divided between the forest products industry and private non-industrial groups and
individuals. As the amount of available timber on federal lands declines, more pressure is put
on private lands. Conversely, as timber declines on private lands, pressure is put on public
lands to provide timber to maintain local economies.

7,000

6,000

4? 5,000 -
D Private
CU
o4, 000 0 USFS
QBLM
0 3,000 State
C/)
C Mother
0 _J
2,000

1,000 -
0 ...... .
Oregon Washington California Idaho Alaska

Figure 3.6. Lumber production in states within the range of Pacific anadromous
salmonids. (Data are from Warren 1993.)

12 Historical Background

100,000
0 Private forest industry
0 Private non-industry
0 USFS
0 BLM
80,000
0 State
0 Other
C: 60,000

...... 77
0 ...........7

40,000 .......

............
Cz

20,000

...........
0 L
Oregon Washington California Idaho Alaska

Figure 3.7. Ownership of timberlands in the Pacific Northwest and Alaska. (Data are
from Powell et al. 1993.)

Chapter 4
Salmonid Habitat Requirements

The typical life cycle of anadromous salmonids consists of several stages that encompass marine,
estuarine, and freshwater environments (Fig. 4. 1; Groot and Margolis 1991). Each life stage,
has. different habitat requirements. Adults returning from the sea require access to spawning
gravel and cover from predators; eggs and alevins incubating in the streambed require stable,
permeable gravel and cool, oxygenated water; juveniles rearing in the stream require food,
suitable temperature, and cover; and smolts migrating to sea require adequate strearnflow.
These multiple factors operate simultaneously and vary seasonally and annually. Their role in
salmonid population dynamics also changes with the different stages of the life cycle.

Adult
Spawning

Smolt
Migration
Juvenile Ocean Egg
Rearing Residence Incubation

Fry
Emergence

Figure 4. 1. Generalized life cycle of anadrornous salmonids, involving several distinct
freshwater life stages and a period of ocean residence.

14 Habitat ReQuirements

Habitat requirements of anadromous salmonids have been thoroughly reviewed (Bjomn and
Reiser 1991). This chapter presents the salient habitat requirements that pertain to effects of
logging. It follows the life cycle of salmonids, beginning with the habitat requirements of adults
as they return to spawn, and ending with the migration of smolts as they leave for the sea.

Although descriptions of habitat requirements are. generally species-specific, almost all aquatic
habitats used by anadromous fish accommodate complex assemblages of stocks rather than a
single stock. Because the various species, stocks, and life stages of salmonids have different
habitat requirements, the streams they inhabit must provide appropriate diverse hydraulic and
geomorphic conditions. from the headwaters to the river mouth.

UPSTREAM MIGRATION OF ADULTS

Adult anadromous salmonids must reach the spawning grounds at the proper time and with
enough energy left to spawn (Bjornn and Reiser 1991). Native stocks have migration schedules
that are flexible enough to accommodate delays during normal periods of unsuitable conditions.
Stocks that migrate up long river systems often arrive in the spawning area several months
before spawning, whereas those migrating up short coastal streams often do not move into
streams until just before spawning. Some stocks enter streams in fall when strearnflow is high
and do not spawn until the following spring.

Although the migration schedule is flexible, it has evolved to meet the specific flow and
temperature regimes of the natal streams. Temperature during incubation is probably the
primary determinant of spawning time (Heggberget 1988). Any given stock spawns mostly
during a 3- to 4-week period determined by the temperature regime where the eggs incubate
(Meehan and Bjornn 1991). Spawning is timed to allow for egg incubation so that fry emerge
at the right time to feed and grow.

For successful migration, conditions must be suitable for at least part of the migration season.
Strearnflow should be at least 30-70% of the mean annual flow, and water depth should be at
least 18-24 cm, depending on the species (Bjornn and Reiser 1991). High turbidity can cause
fish to delay migration, but turbidity generally does not affect homing. Fish easily negotiate
their way up highly turbid glacial rivers (Eiler et al. 1992). . Water temperature can be in a
broad range between 3 and 20'C. If strearnflow is too high, water velocity may be too fast.
Adult salmon can swim at a sustained speed of more than 3 m/s and dart over 6 m/s, but the
maximum negotiable water velocity is under 3 m/s in critical stream reaches, usually shallow
riffles.

Low dissolved oxygen (DO) can impede migration and even cause fish kills in some situations.
Swimming performance declines sharply when DO concentration drops below 7 mg/L, and fish
may stop migrating when DO is below 4.5 mg/L (BJornn and Reiser 1991). Pre-spawn die-offs
of adult pink (0. gorbuscha) and chum salmon,(O. keta) can occur during temporary summer
droughts when fish become crowded in pools and deplete oxygen (Murphy 1985).

Habitat Reguirements 15

Large pools and abundant cover are important components of migration habitat. Adult salmon
often hold for several weeks in large pools as they ascend a stream to spawn (Burger et al. 1985;
Thedinga et al. 1993), and some fall-run stocks spend months in fresh water before spawning
(Meehan and Bjornn 1991). During this time, the adults are vulnerable to predation and
disturbance, and must conserve energy for further upstream migration and spawning. Bears can
take a large toll in some areas, but are not significant where their number has been reduced.
Human disturbance can be important where salmon holding areas are near recreation sites.
Large pools and abundant LWD provide resting habitat and cover needed for successful upstream
migration.

Perhaps most important for adult migration is freedom from barriers. Water falls, debris jams,
and other obstacles can block upstream passage. Many such barriers are only temporary.
Debris jams and beaver dams may be barriers at low flow but become passable at higher flow
(Bryant 1984). Salmon and steelhead can jump obstacles up to 3 m high if the pool below the
falls is deep enough (Bjornn and Reiser 1991). Salmonids can get past many obstacles that
appear to be barriers, given suitable streamflow and obstacle configuration.

SPAWNING

The usable spawning habitat in a stream depends on the stream's size, depth, velocity,
temperature, amount of proper-size gravel, and configuration of pools and riffles (Bjornn and
Reiser 1991). Streamflow regulates the amount of spawning area in a stream by controlling the
area covered by water and its depth and velocity. For each stream, there is an optimum flow
that provides the maximum usable spawning area. Salmonids can spawn when water temperature
is as low as I'C and as high as 20'C, but the favorable range is between 4 and 17'C,
depending on the stock (Bjornn and Reiser 1991). Suitable gravel substrate is between 1 and
10 cm diameter, depending on fish size, but upwelling groundwater can make substrates finer
than 1 cm suitable for spawning (Lorenz and Eiler 1989).

The number of possible redds in a stream depends on the area required per spawning pair and
the area of suitable spawning habitat (13jornn. and Reiser 1991). The average redd size of
anadromous salmonids ranges from 0.6 n12 for pink salmon to 10 M2 for chinook salmon, and
the area needed per spawning pair 'is from 0.6 to 20 M2 (Table 4. 1). The number of possible
redds in a stream is roughly equal to the gravel area divided by 4 times the average redd size
(Bjornn. and Reiser 1991). Suitable spawning habitat, however, is often much less than the area
of gravel because suitable habitat also requires proper channel configuration.

The best channel configuration for spawning is often at transition areas between pools and
riffles. These areas often have abundant intragravel flow which brings oxygen to the eggs and
removes metabolic wastes (Bjornn and Reiser 1991). Downwelling currents at the tails of pools
force water into the gravel, and upwelling currents at downstream ends of riffles force water up
from below. Structural features of the channel, such as LWD, are important for spawning
habitat because they help create these pool-riffle transition areas. In gravel-poor streams, LWD
also forms spawning habitat by trapping gravels (Everest and Meehan 1981) and in sediment-rich
streams, by scouring silt out of spawning beds (Sedell and Swanson 1984).

16 Habitat Reguirements

Table 4. 1. Average redd area and area per spawning pair. (After Bjornn and Reiser
1991.)
Species Redd area (M2) Area per pair (M2)

Chinook salmon 3-10 13-20
Steelhead 4-5
Coho salmon 2.8 12
Chum salmon 2.3 9
Sockeye salmon 1.8 7
Pink salmon 0.6-0.9 6
Cutthroat trout 0.1-0.9

As in upstream migration, cover is also important during spawning, and nearness to cover often
determines spawning sites (Bjornn and Reiser 1991). Overhanging and submerged vegetation,
undercut banks, deep water, turbulence, and LWD provide shade and protect fish from
disturbance.

INCUBATION

Successful incubation depends on numerous variables, such as strearnflow, channel stability, and
temperature, but usually the most important factor is substrate permeability. Permeable substrate
is needed so that water can move freely to bring oxygen to the embryos and carry waste away.

Fine sediment (usually < 0. 8 mm diameter) is detrimental to embryo survival because it reduces
substrate permeability. Permeability of gravel decreases sharply as fine sediment increases from
5 to 20% of the -substrate, and embryo survival declines similarly (Fig. 4.2; Chapman 1988).
Low DO can kill embryos or cause them to emerge in poor condition (Koski 1975). Fine
sediment can also interfere with emergence by blocking interstices in the gravel. Coarser
sediment (1-2 mm diameter) can cap the substrate and trap alevins within the gravel.

The spawning activities of adult salmonids can remove fine sediment from the redd and
surrounding area (Chapman 1988)., New redds contain less fine sediment, especially fine organic
particles which are light and easily moved. Where numerous spawners use the same area every
year, they help keep the gravel in good. condition. If the population declines, the spawning
habitat can deteriorate.

Shifting of spawning gravel during periods of high strearnflow ca'n cause heavy mortality during
incubation (McNeil 1964). Losses can be especially high where stream bedload is excessive
because of mass wasting. For example, in the Queen Charlotte Islands, British Columbia, 45-
86% of salmon eggs were destroyed by scour in watersheds with severe mass wasting, compared
to only 0-14% in more stable areas (Tripp and Poulin 1986).

Habitat Requirements 17

1 P0

1-10
0
CaD 80

60-
E ......
a)
0 40-

'5 20-

0
0 10 20 30 40 50 60
Percent fine sediment

Figure 4.2. Egg-to-fry survival in relation to fine sediment (<2-6.4 mm) in spawning
gravel. The stippled area includes data from four separate laboratory studies. (After
l3jornn and Reiser 1991.)

The time required for embryos to hatch and for alevins to emerge is sensitive to water
temperature and varies by species (Bjornn and Reiser 1991). Steelhead eggs hatch after 85 days
at VC and after 26 days at 120C; Pacific salmon eggs hatch after 115-150 days at 4'C, and
after 35-60 days at 12'C. After hatching, alevins remain in the gravel for about twice the time
it takes eggs to hatch.

The favorable temperature range for incubation is between 4 and WC (Bjornn and Reiser
1991). Many salmonid streams are colder than VC in winter, but this is generally not a
problem if initial development occurs within the suitable range. Because spawning and egg
development seldom occur in summer, intragravel water temperature is usually well below lethal
or sublethal levels. Eggs of spring spawners, however, may be exposed to high water
temperature in late spring.

In northern and high-elevation regions, freezing can cause high mortality during incubation.
Direct mortality from freezing is one of the three most important mortality factors (besides low
DO and gravel shifting) for- pink salmon embryos in southeast Alaska (McNeil 1964). Low
temperature also has indirect effects due to formation of anchor ice on the substrate, which can

18 Habitat Requirements

block interchange of water into the redd. Ice dams can destabilize the stream and scour the
spawning gravel.

FRESHWATER REARING

The different species and'stocks of anadromous salmonids spend different periods of time in
fresh water. Pink and chum salmon spend the least time, usually migrating to sea immediately
after emergence. Most sockeye salmon (0. nerka) spend up to 3 years in lakes, but juveniles
of some stocks inhabit riverine sloughs and migrate to sea their first spring or surnmer (Heifetz
et al. 1986; Eiler et al. 1992). Juvenile chinook salmon (0. tshaKyucha) occur as two general
types: "stream type," which inhabit streams for I or 2 years (e.g., spring runs), and "ocean
type," which migrate to sea their first summer (e.g., fall runs) (Healey 1983). Juvenile coho
salmon spend I to 3 years, and juvenile steelhead, Atlantic salmon, and cutthroat trout spend
2 or more years in streams.

Stream habitat is obviously more important for species that have extended rearing in streams.
For these species, the stream's habitat puts an upper limit, or "carrying capacity," on the
number of juveniles (Bjornn and Reiser' 1991. At low levels of spawning and fry abundance,
carrying capacity does not restrict abundance of juveniles. In such cases, older juvenile
populations are directly related to spawner abundance but less than carrying capacity. At higher
levels, biological and physical factors cause density-dependent mortality and emigration so that,
by late summer, the juvenile population approaches equilibrium with the stream's carrying
capacity.

If spawning escapement is adequate, sufficient fry are usually produced to exceed carrying
capacity (e.g., Crone and Bond 1976). In two years in Sashin Creek, Alaska, for example, co'ho
spawners and potential egg deposition increased more than fivefold, but the resulting number of
yearling juveniles was nearly identical (Table 4.2). This is probably an extreme example
because the number of juvenile fish in a stream are unlikely to vary so little. It does, however,
illustrate the principle of how rearing habitat can limit populations of coho salmon. Reduced
spawning success in such populations would not necessarily reduce smolt yield, as long as

Table 4.2. Number of coho salmon at different life stages in Sashin Creek, Alaska, for
two brood years with a large difference in spawner escapement. (Data from Crone
and Bond 1976.)
Brood Spawning Potential Pre-emerged
year adults eggs' fry Fry Yearlings

1963 916 1,460,000 214,000 51,852 4,581

1964 162 260,000 58,000 20,355 4,546

'Potential number of eggs estimated from number of females and mean fecundity.

Habitat Requirements 19

sufficient fry continue to be produced to fully seed available habitat (Everest et al. 1987a).
However, if spawning escapement is reduced by overfishing or other causes so that the reservoir
of surplus juveniles is lacking, the number of smolts is more sensitive to forestry-related impacts
(Cederholm and Reid 1987).

Theoretically, carrying capacity of summer habitat sets a density-dependent limit on the juvenile
population, which then suffers mortality in proportion to the severity of winter conditions and
quality of winter habitat. Unlike in summer, mortality of juvenile salmonids in winter results
mainly from density-independent factors (Hunt 1969). Winter mortality in streams is substantial:
46-94 % of late-summer populations (Murphy et al. 1984; Hartman et al. 1987), and can usually
be attributed to hazardous conditions during floods, stranding by ice dams, and physiological
stress from low temperature, oxygen depletion, or progressive starvation (Bryant 1984; Murphy
et al. 1984; Harding 1993). Survival factors become more important as latitude increases,
because winters are more severe and juveniles stay longer in fresh water. South of Alaska, for
example, coho salmon generally spend only one winter in fresh water, whereas in Alaska, most
spend two winters (Gray et al. 1981).

The ultimate production of salmon smolts is determined by numerous in teracting environmental
factors, including food availability, temperature, water quality, strearnflow, and cover (Bjornn
and Reiser 1991). The importance of each factor varies seasonally: food availability,
temperature, and strearnflow are most important in summer; cover is key in winter.

Food Availability

In many streams, summer carrying capacity is more related to food supply than to physical
factors. Juvenile salmonids will not stay in a stream if the food level is not adequate
(Konopacky 1984; Wilzbach 1985). The proof that wild salmon populations in summer are often
food-limited is that their number can be increased by supplemental feeding (Mason 1976).
Abundant food can increase carrying capacity because more fish can occupy a given area and
fewer emigrate (Mason and Chapman 1965). Salmonids can adjust the size of their territories
according to food abundance by altering their aggressive behavior (Dill et al. 1981). When food
becomes more plentiful, aggression subsides and territories contract.

The period of greatest mortality of juvenile anadromous salmonids is usually during spring and
early summer, when newly emerged fry colonize habitat and establish territories, and the
population adjusts to carrying capacity. Adults usually spawn in limited parts of a drainage, but
the juveniles spread out and colonize most accessible and suitable areas (Leider et al. 1986).
Where fry are numerous, their territories cover all suitable parts of the stream, and surplus fry
are displaced and emigrate. As fish grow, they require more space, and the population must
shrink to accommodate needs.

Stream-rearing salmonids predominantly eat invertebrates that drift downstream (Elliott 1973).
Because salmonids are generalist predators, all groups of invertebrates are potential prey, but
certain types of immature aquatic insects are eaten most because of their propensity to drift.
Common food items include the midges (Chironomidae), mayflies (Ephemeroptera), blackflies
(Simuliidae), and net-spinning caddisflies (Hydropsychidae) (Murphy and Meehan 1991). These

20 Habitat Requirements

insects generally are small, and their populations turn over rapidly so that production is often
2-10 times their standing biomass.

Salmonids are usually territorial, and the amount of food they get depends on their territory size
and location. Suitable territories have one or more stations with slack water yet close to fast
water so that fish can wait for drift while saving energy (Fausch 1984). Such stations are
usually limited, so salmonids habitually use only part (often < 15 %) of the available habitat
(Bachman 1984).

To understand food production in streams, one must understand how stream ecosystems operate
(Murphy and Meehan 1991). Stream ecosystems have two energy sources: in-stream primary
production from aquatic plants (e.g., diatoms) and out-of-stream sources of organic matter (e.g.,
leaves from riparian plants). The amount of these energy sources depends mainly on riparian
vegetation and stream size (Vannote et al. 1980).

Riparian vegetation influences energy sources for the stream by providing shade and dropping
organic matter into the stream and flood plain (Meehan et al. 1977). Small streams are often
so heavily shaded that aquatic primary production is limited by dim light, and most energy
come s from decomposition of the leaves and other organic matter that fall from streamside
plants. Besides leaves and other litter, insects falling from riparian vegetation supplement the
salmonid diet.

As streams get larger, the influence of riparian vegetation diminishes (Vannote et al. 1980). In
the headwaters, small trees and shrubs can provide effective shade, but farther downstream, even
large trees may be insignificant. Small streams receive more of their organic matter from local
vegetation than do larger streams, although secondary channels of rivers can function like small
streams and receive matter directly from local vegetation (Sedell and Froggatt 1984).

The amount of organic matter in a stream is not, by itself, a good measure of stream
productivity; more important is food quality (Murphy and Meehan 1991). Almost all forms of
organic matter have too high a carbon:nitrogen ratio to be utilized directly by invertebrates, and
must first be colonized by fungi and bacteria to enhance food quality. At the low end of the
food-quality spectrum are woody debris and the fine particulate residues of already-processdd
detritus; at the high end are periphyton and deciduous leaves. A large reservoir of decay-
resistant organic matter usually is stored in the stream channel throughout the year. Between
the fresh inputs from leaf fall and algal blooms, most labile detritus is metabolized, whereas the
refractory portions accumulate and steadily decline in nutritional quality.

Large woody debris has a key role in the retention of organic detritus for processing by the
stream ecosystem (Bisson et al. 1987). Debris increases the diversity of stream morphology,
creating pools, multiple channels, sloughs, and backwaters. Debris dams trap branches that
retain leaves and even carcasses of anadromous salmonids (Cederholm and Peterson 1985).
Because large logs resist decay and transport, they provide stable storage sites in small streams,
where one-half of the streambed may consist of woody debris and detri'tus. The role of woody
.debris changes with stream size. In small streams, debris remains where it falls and affects most

Habitat Requirements 21

of the stream. In larger streams, debris is clumped in logjams or floated onto the banks but is
still important in creating habitat.

Thus, stream productivity depends on complex interactions between the stream and riparian
vegetation. Although energy flow is a key determinant of carrying capacity for juvenile
salmonids, the energy must flow within the context of suitable physical habitat.

Temperature

Because salmonids are cold-blooded, water temperature directly influences their physiology and
activity. The preferred temperature of juvenile Pacific salmon is 12-14 C, but juvenile
salmonids in fresh water have broad tolerance for temperature extremes and readily acclimate
to the ambient temperature (Bjornn and Reiser 1991). They seem especially tolerant of high
temperature, many degrees higher than they are likely to encounter (Beschta et al. 1987).

Lethal temperature determined in the laboratory depends on the temperature to which the fish
is accustomed. Fish acclimated to 5'C survive up to about 22'C; fish acclimated to 15 C
survive up to about 25 C (Fig. 4.3; Beschta. et al. 1987). The application of such laboratory

25- Chinook
Sockeye
Coho
Lethal

threshold Pink
24
temperature ( C)

23- Chum

22-

0 5 10 15 20 25
Acclimation temperature ( C)

Figure 4.3. Lethal threshold temperature for juvenile salmonids in relation to
acclimation temperature. (After Beschta et al. 1987.)

22 Habitat Requirements

data to natural situations, however, is tenuous because natural streams are complex temperature
environments, and wild fish have behavioral and physiological strategies to resist temperature
0
stress. They can seek out localized cool-water sources, such as deep stratified pools and areas
with upwelling groundwater (Bilby 1984a), or they can become inactive (Beschta et al. 1987).
1
The optimum temperature for growth depends on food availability. Fish will not grow until
their metabolic requirements are met, and their requirements increase with temperature.
Laboratory studies show that growth decreases in situations with high temperature and limited
food (Beschta et al. 1987). Growth can be maintained at higher temperature if the food supply
increases enough to compensate for increased metabolic needs. Except when fish are starving,
growth and activity increase with temperature up to some optimum and then decrease as the
optimum is exceeded.

Cold temperature in fall and winter causes juvenile salmonids to become less active, feed less,
and hide in the substrate, woody debris, and other cover (Bjornn and Rieser 1991). Hiding
behavior begins when temperature drops below about 5'C. If cover is not available as
temperature drops, fish may move long distances before finding suitable winter habitat.

Streamf low

Streamflow is a basic habitat determinant for salmonids. The flow regime varies regionally,
depending on climate, topography, and vegetation (Swanston 1991). Coastal streams have high
flow in fall and winter because of heavy rain and snow, and the largest floods happen during
rain-on-snow events. In the interior, flow is high in spring during snowmelt, but rain-on-snow
floods can happen in winter. Minimum flow generally occurs in late summer, and in northern
and interior regions, a second minimum occurs during winter freeze-up.

The relationship between streamflow and carrying capacity depends on surrounding land forms.
It differs, for example, between streams in V-shaped canyons, where flood flows are contained,
and streams on broad valleys, where floods can spread out over a wide flood plain. In general.,
carrying capacity increases as streaniflow increases up to a point, and then levels off or declines
if water velocity becomes excessive (Bjornn and Reiser 1991). Minimum strearnflow in summer
can limit carrying capacity on a broad scale. Total commercial catch of coho salmon off
Washington and Oregon is directly related to the amount of summer streamflow when the
juveniles were in streams 2 years before (Smoker 1955; Mathews and Olson 1980).

Dissolved Oxygen and Turbidity

Dissolved oxygen is normally adequate in most natural streams because turbulence keeps DO
near saturation (about 10 mg1L at 10'C). Oxygen can be depleted, however, when strearnflow
is low and temperature high, and where fine sediment blocks interchange with interstitial water.
Local DO depletion to as low as 2 mg/l, can occur where upwelling subsurface water enters the
stream (Beschta et al. 1987). Juvenile salmonids can survive to as low as 2 mg/L, but growth
is reduced at less than 5 mg/L, and swimming performance declines below 8 mg/L (Bjornn and
Rieser 1991).

Habitat Requirements 23

Streams periodically become turbid with suspended sediment during storms and intense
snowmelt. Turbidity is an optical property of water wherein suspended materials, such as clay,
silt, fine organic matter, and plankton, scatter light. Nephelometric techniques measure this
"cloudiness" of water in terms of nephelometric turbidity units (NTU). At about 20 NTU,
water first appears cloudy; at 40 NTU, turbidity is highly noticeable; and at 100 NTU, visibility
is reduced to a few centimeters. Juvenile salmonids can tolerate moderate, short-term turbidity,
and may even benefit from the added cover from predators (Gregory 1993a). High turbidity that
occurs in watersheds with excessive erosion, however, can disrupt fish behavior and reduce
growth. Juvenile coho salmon stop feeding when turbidity exceeds 60, NTU (Berg and Northcote
1985) and avoid water with 70 NTU (Bisson and Bilby 1982). The turbidity response depends
on the species. Juvenile coho salmon avoid glacial rivers with high turbidity, whereas chinook
and sockeye salmon feed and grow in glacial rivers with turbidity as high as 200 NTU (Murphy
et al. 1989; Brownlee 1990).

Channel Morphology

Natural streams contain a diverse mixture of habitats differing in depth, velocity, and cover,
arranged in repeating habitat units of pools, riffles, and glides (Bisson et al. 1987).
Classification of stream area according to habitat units (Bisson et al. 1982) forms the basis for
today's quantitative analysis of salmonid habitat (McCain et al. 1990; Dolloff et al. 1993).

A stream's channel morphology is determined mainly by associated hill slopes and riparian
vegetation (Sullivan et al. 1987). The principal factors controlling channel morphology are
water discharge, sediment load, bank characteristics, and solid structures, such as LWD,
bedrock, and boulders. Stream channels are shaped primarily during storms, when flow is high
enough to move sediment lining the channel bed. Stable channels are in dynamic equilibrium
where the influx of sediment is balanced by the stream's capacity to carry sediment away. Pools
and riffles may change location, but their average balance is maintained.

The physical consequences of LWD are particularly important to salmonids (Sullivan et al.
1987). Debris creates pools and undercut banks, deflects streamflow, retains sediment, and
stabilizes the stream channel. Debris not only provides cover directly, but also forms 80-90%
of pools in typical valley-bottom streams (Fig. 4.4; Heifetz et al. 1986). By forming pools and
retaining sediment, LWD also helps maintain water levels in small streams during periods of low
streamflow (Lisle 1986). Debris increases the hydraulic complexity of streams, and the slack
water around debris offers good opportunities for feeding close to the faster water carrying insect
drift.

Water velocity is an important factor determining habitat use for juvenile salmonids (Bjornn and
Reiser 1991). The velocity range used- by salmonids differs among species and generally
increases with fish size. Coho generally use slower water than steelhead and chinook.
Complexity of natural stream habitat helps accommodate the diverse needs of the various
salmonid species and age groups.

Most stream-dwelling salmonids inhabit pools because of low current velocity. Drifting food,
however, is usually more abundant where velocity is high. Although the flow carries drift, the

24 Habitat Requirements

0.,

Plunge pool

Dammed pool

Lateral-scour pool

Figure 4.4. Formation of various pool habitats by large woody debris. (From Bisson
et al. 1987; reproduced with permission from University of Washington, Institute of
Forest Resources.)

metabolic cost of maintaining position in riffles is too high. Riffles have few suitable feeding
stations, usually located behind boulders and debris, and these provide limited vision of passing
food (Bisson et al. 1987). Coho salmon avoid riffles almost entirely, preferring pools with
ample cover where they can feed on drift as it enters the pool from upstream riffles (Nickelson
et al. 1992a).

Cover

Abundant cover can increase a stream's carrying capacity for juvenile salmonids by providing
security from predation and floods. Abundant cover also visually isolates fish, reducing
aggression and territory size (Dolloff 1986). Cover can include woody debris, overhanging
vegetation, undercut banks, cobble and boulder substrate, water depth and turbulence, and
aquatic vegetation. The need.for cover varies diurnally, seasonally, by species, and by fish size
(Bjomn and Reiser 1991).

In winter, the need for cover overrides all other habitat needs. As winter approaches, salmonids
feed less and seek cover, particularly in ponds and stream reaches with abundant woody debris

Habitat Requirements 25

or cobble substrate (Peterson 1982; Murphy et, al. 1984; Johnson et al. 1986). Pools with LWD
and coarse substrate are critical winter habitats of juvenile salmonids (Fig. 4.5). Fine sediriient,
can reduce amount of cover available in the streambed by filling interstitial spaces (Fig. 4.6).

0.25
4-0
E
.0.20
Cz

0.15

CL
C: 0.10
0
E
CU
0 0.05
0

0
L)
0 .00 0-49 50-99 100-199 201-400 >500
Cubic meters of large woody debris per 1 00-m reach

Figure 4.5. Winter density of coho salmon in relation to large woody debris in
southeast Alaska. (After Murphy et al. 1984.)

I V
E
OR 8
ca

Cr 6
CD

(D
CL 4

Cz
(D 2 -

CD
Q) 0
-&-0
U) 0 20 40 60 80 100
Percent embedded by fine sediment

Figure 4.6. Habitat use by juvenile steelhead in winter in relation to embeddedness
of the streambed by fine sediment. (After Bjornn and Reiser 1991.)

26 Habitat Requirements

SEAWARD MIGRATION

The timing of seaward migration of salmonids that spend extended periods in streams is
regulated primarily by photoperiod, and is modified by temperature, strearnflow, and fish growth
(Groot and Margolis 1991). The timing of outmigration is often nearly,the same each year and
usually peaks during a spring freshet. Migrations of native stocks are timed so that the smolts
arrive in the estuary when food is plentiful. Rapid growth during the early period in the estuary
is critical to survival because of high size-dependent mortality from predation (Murphy et al.
1988).

The parr-smolt transition is often incomplete when fish begin migration, especially in larger
watersheds (Rodgers et al. 1987). "Presmolts" may spend several weeks in the lower parts of
rivers until the transition is complete.' Stocks of ocean-type chinook exhibit slow "rearing
migrations," in which the fish feed and grow while migrating toward lower parts of the drainage
system in summer (Johnson et al. 1992). In short coastal streams, migrations tend to be quick,
and the parr-smolt transition is nearly complete before migration starts (e.g., Thedinga et al.
1994).

Habitat requirements during seaward migration are similar to those of rearing juveniles, except
that smolts tend to be more fragile (Thedinga et al. 1994). They easily lose scales if handled,
and are less tolerant of low DO and high temperature. Because most smolt migrations occur
during freshets in spring, low oxygen and high temperature are not usually a problem. High
strearnflow aids their migration by flushing them downstream and reducing their vulnerability
to predators.

Migrating smolts are particularly vulnerable to predation because they are concentrated and
moving through areas of reduced cover where predators congregate (Larsson 1985). Most
predation occurs in certain areas that are predation "hot spots" (Fast et al. 1991). Mortality
during seaward migration can exceed 50% and represent a major loss in salmon production
(Larsson 1985; Thedinga et al. 1994). In small streams, numerous predators, such as
mergansers, otters, and mink, may be drawn to streams during the smolt migration and can take
a heavy toll on migrating fish. Woody debris, undercut banks, and moderate turbidity provide
important cover for migrating smolts.

Fish can also be serious predators in larger rivers and reservoirs. In the Columbia River, for
example, northern squawfish (Ptychocheilus oregonensis), walleye (Stizostedion vitreum), and
smallmouth bass (Micropterus dolomieu) took 14% of all migrating salmonids that entered the
John Day Reservoir (nearly 3 million fish); 21 % of this loss was in a small area at the head of
the reservoir (Rieman et al. 1991). The NMFS Proposed Recovery Plan for Snake River
Salmon (USDC 1995) proposes actions to control predation by squawfish, birds, marine
mammals, and non-native fishes to increase survival during downriver migration.

Habitat Requirements

LIMITING FACTORS

Most analyses of habitat requirements examine the role of specific habitat components in
isolation. To assess overall effects of habitat change on salmonids, specific factors that may
limit their abundance must be considered in the context of all other factors over the entire
freshwater period. This is the province of the analysis of limiting factors.

The objective of limiting factor analysis is to identify the "bottleneck" that restricts overall smolt
yield (Fig. 4.7). In this analysis, the - salmonid life cycle is divided into different life stages,
each dominated by different habitat components (Reeves et al. 1989). The factor that lirnits
overall smolt production from a stream can affect the population at any life stage. For example,
the stream may be underseeded and in need of greater escapement of adults. Spawning may be
adequate, but summer carrying capacity may be restricted by high temperature or low food
production. In many streams damaged by past logging, lack of winter cover is the main
"bottleneck" that restricts smolt production.

For anadromous salmonids, river systems should be viewed as interconnected networks which
contain habitat for spawning, rearing, and migration in different areas within the network. To
provide for long-term maintenance of habitat for all life stages, drainage networks need to be
managed on a landscape scale (Naiman 1992; Reeves et al. in press).

SPRI"M SUMMEA FALL W4W.WF1

SPRIHO sukumn FALL .14TIER

A. W L ---------
ra

To CWA^m

C FALL

Figure 4.7. Illustration of "bottlenecks" restricting salmonid smolt production.
Bottlenecks in winter (A) may be caused by poor cover from flooding and icing;
bottlenecks in summer (B) may be caused by poor food production. Panel C illustrates
(EN
KKED

how poor winter cover nullifies increased food availability in summer. (From Reeves
et al. 1991; reproduced with permission from the American Fisheries Society.)

Chapter 5
The Potential Eff ects of Logging

Logging and associated activities can have multiple effects on salmonid habitat. Salmonid
habitat is a product of interactions among the stream, floodplain, riparian area, and uplands-in
short, the entire watershed. Effects of timber harvest, road construction, and other activities
anywhere in the watershed can be transmitted through changes in hydrologic and erosional
processes to modify habitat for salmonids (Fig. 5. 1; Chamberlin et al. 199 1).

Effects of logging on anadromous salmonids have been studied intensively since the 1950s, and
have been reviewed comprehensively (Gibbons and Salo 1973; Salo and Cundy 1987; Meehan
1991). Although many details about logging impacts are still unknown, their causes and
mechanisms are understood. Impacts on a specific site are usually not predictable quantitatively
because of the many interacting factors involved, including random weather events (Sullivan et
al. 1987; Chamberlin et al. 1991). However, given information on the logging activity,
watershed characteristics, riparian vegetation, stream, and fish populations, one can specify the
probable direction and magnitude of habitat changes and effects on salmonid populations.

Studies of effects of logging typically have examined specific habitat components and
consequences for different parts of the salmonid life cycle. Most studies attempt to isolate
effects of one variable from others that also change after logging. Fewer studies have examined
overall cumulative effects of logging and the integrated population response of salmonids in
whole basins. Thus, much is known about the potential effects of timber harvest on various
habitat components and the response of various salmonid life stages (Table 5.1). Integrating
these separate comp onents into an ecosystem model of habitat effects and population response
is more theoretical (Hicks et al. 1991a).

In addition to these effects of actual timber harvest and roads, other timber management
activities also can affect salmonids adversely. Use of forest chemicals (fertilizers, pesticides,
and fire retardants) can affect salmonids directly and indirectly (Norris et al, 1991). The risk
of direct toxic effects can be reduced when buffer zones along streams are left untreated and
applicators are careful to prevent drift and avoid direct application to surface water.
Recreational use of streams and riparian areas usually will have minor negative effects on fish
habitats, but recreational activities are highly variable and must be evaluated locally (Clark and
Gibbons 1991). Intensive recreational use can damage riparian vegetation and streambanks or
disturb spawning adults.

30 Effects of Logging

Forest Management Activities
Ar@ '*)k
Process Water balance
changes .4 Energy balance
Sediments
Nutrients

Structural Soil structure/stability
so. Vegetation and debris
changes .44 Drainage network
Channelshape

Water velocity/oepth
Water quality
Habitat Banks
elements
Bed composition
Cover
Riparian vegetation
Migration barriers

Number
Fish ow Growth
Survival
Distribution
Species

Figure 5. 1. Connections between timber harvest, road construction, and other timber
management activities in a watershed and effects on salmonid habitat via changes in
watershed processes and structures. (After Chamberlin et al. 199 1.)

Effects of Lo gging 31

Table 5. 1. Influences of timber harvest on physical characteristics of stream
environments, potential changes in habitat quality, and resultant consequences for
salmonid growth and survival.* (After Hicks et al. 1991 a.)

Forest Potential Potential change in Potential effects on
practice change salmonid habitat salmonid populations

Timber Decreased Increased summer Changes in growth,
harvest in shade temperature; more age at smolting; early
riparian zones light, more algae, more emergence; increased
food production summer carrying
capacity

Decreased Reduced cover, pool Decreased winter
supply of habitat, gravel and survival, spawning
large woody organic matter storage, success, and species
debris hydraulic complexity, diversity; increased
and food production predation

Addition of Increased oxygen Reduced spawning
slash (bark, demand, organic success; short-term
branches) matter, food, and increase in growth
cover; decreased
channel stability

Streambank Reduced cover, stream Increased carrying
erosion depth capacity for fry,
reduced for older fish;
increased preddtion

Increased fine sediment Reduced spawning
in streambed; reduced success; slower
food supply growth

Timber Altered Temporarily increased Temporarily increased
harvest on streamf low summer base flow survival
hillslopes;
forest roads

Increased peak flows Increased embryo
and bedload shift mortality

32 'Effects of Logging

Table 5. 1. Continued.

Forest Potential Potential change in Potential effects on
practice change salmonid habitat salmonid populations

Increased Increased fine sediment Reduced spawning
erosion in stream gravels; success, growth, and
reduced food carrying capacity;
production and cover increased winter
mortality

Increased supply of Increased or
coarse sediment decreased rearing
capacity

Increased debris Migration blockages;
torrents; less cover in poor survival in
torrent track; more torrent track; better in
debris jams debris jams

Increased Increased food Increased summer
nutrients production carrying capacity

Stream Obstructions in stream Restricted upstream
crossings channel; sediment movement; reduced
inputs spawning success

Scarif ication Increased Short-term increase in Temporarily increased
and slash nutrients nutrients and food growth and summer
burning production carrying capacity

Inputs of fine Increased fine. Reduced spawning
organic and sediment; temporarily success
inorganic increased oxygen
sediment demand

Effects of logging 33

STREAM TEMPERATURE

The detrimental increase in summer temperature was one of the first issues identified as
environmental awareness about the effects of logging developed in the 1960s (Beschta et al.
1987). These concerns led to changes in FS and BLM national policy and development of state
forest practices acts intended to prevent temperature changes in fish-bearing streams.

The principal source of energy for heating streams is direct solar radiation hitting the water
surface, whereas heat from convection and conduction are insignificant (Beschta et al. 1987).
Streams are heated by reductions in shade, not by warm air, nor do they quickly cool after
entering shaded sections. Streams exposed over long reaches do not heat up indefinitely, but
reach equilibrium as evaporation, convection, conduction, and inflow of groundwater balance
the radiation load.

Canopy removal can increase stream temperature as much as 10'C (Fig. 5.2; Beschta et al.
1987). The increase is directly proportional to the area opened to sunlight and inversely
proportional to stream discharge (Beschta et al. 1987). Thus, the increase is greatest on small
streams and diminishes as streams get wider because of the lessening influence of tree canopy.
Specific stream and watershed conditions cause wide variation in temperature response. Stream
gradient, morphology, orientation, latitude, and bed materials are some factors that affect how
much temperature will increase after canopy removal.

22-
011 Clearcut
20-
(D
18-
:3
4"
Cz
16-
CL 14 --
12 Forested

0 6 12 1@ 24

Hours

Figure 5.2. Typical daily stream temperature in clearcut and forested streams during
clear weather in Oregon's Coast Range. (From Beschta et al. 1987; reprinted with
permission from University of Washington, Institute of Forest Resources.)

34 Effects of Logging

Changes in stream temperature are considered harmful to salmonids because stocks are adapted
to their stream's natural regime, and any change can alter development, growth, survival, and
timing of life-history events (Beschta et al. 1987). Increased temperature beyond the preferred
range can cause juveniles to leave or grow slower. High temperature can inhibit upstream
migrations of adults and increase disease. Increased temperature can exacerbate die-offs of adult
pink and chum salmon during temporary surnmer droughts in small coastal streams (Murphy
1985).

Although increased temperature is a major concern, field studies have generally failed to
demonstrate significant temperature impacts on salmonids after clearcut logging (Beschta et al.
1987). On the contrary, streams in clearcuts can have large populations of juvenile salmonids
(Murphy and Meehan 1991). The reason for this may be that tolerance limits determined in the
laboratory may not apply to the complex thermal environments in streams. Local cool-water
sources (e.g., upwelling groundwater) can provide refuge from periodic high temperature (Bilby
1984a). Daily stream temperature in clearcuts fluctuates widely and can briefly exceed the
reported lethal threshold. Salmonids apparently can withstand these short-term exposures
without detrimental impact (Beschta et al. 1987). These field studies, however, have generally
examined streams in small clearcuts where the temperature increase was moderated by upstream
forest. Cumulative increases in temperature from numerous small clearcuts could have major
impacts on downstream habitat.

Another reason for lack of reported 'temperature impacts is that most studies have been
conducted in regions with moderate temperature regimes in the center of the salmonid
distribution (e.g., coastal Oregon, Washington, British Columbia, and Alaska; Hicks et al.
1991a). In other regions with higher ambient temperature, on the margins of their distribution,
streams may become too warm for salmonids because of excessive exposure to sunlight (BJorrm
and Reiser 1991). In these regions, larger streams, which are naturally more open to sunlight,
often become uninhabitable for salmonids in summer. Canopy reductions along these streams
or in their headwaters can extend the time and area that temperature is unsuitable.

Long-term warming of streams can cause increased competition and predation. Salmonids may
be replaced because of competition from warmwater species (Reeves et al. 1987). Elevated
water temperature in the Columbia River Basin allowed introduced smallmouth bass, major
predators on juvenile chinook salmon, to expand their range into salmon rearing areas where
cold water might have excluded them in the past (Li et al. 1987).

Although timber harvest can change stream temperature in summer, it does not greatly affect
stream temperature in winter. Canopy removal can raise winter temperature in low-elevation,
coastal drainages; but it can lower it in northern areas and higher elevations because of lost
insulating cover and increased radiative cooling (Beschta. et al. 1987). Where winter temperature
decreases, ice forms more readily and salmonids may die from freezing and icing hazards.
Although effects on winter temperature may be slight, caution is warranted because even a small
change can affect fish when water temperature is low.

Increased winter temperature can have mixed effects on salmonids. Elevated temperature during
egg incubation can speed development and cause fry to emerge early. Early.emergence can be

Effects of-Logging 35

beneficial by prolonging the growing season, leading to larger size in fall and winter, which
helps overwinter survival (Holtby 1988; Thedinga et al. 1989). Early emergence, however,'also
exposes fry to late-winter freshets, and fry and smolts may migrate to sea before the spring
plankton bloom in the estuary, leading to poor ocean survival (Holtby et al. 1989).

Because of the extensive geographic range of salmonids, potential temperature impacts should
be'viewed with a regional perspective. In some regions of the Pacific Northwest andAlaska,
concern over increased summer temperature may be unwarranted (Beschta et al. 1987). In
southeast Alaska, for example, stream temperature in clearcuts rarely exceeds 26'C, except in
exposed, inten-nittent pools (Sheridan and Bloom 1975). Even in southeast Alaska, however,
high temperature may be a problem for salmonids in some "temperature-sensitive" streams that
are wide, shallow, low gradient, and have lake or muskeg sources (Gibbons et al. 1987). In
other regions with comparatively high ambient temperature, such as southern Oregon, California,
and the interior Columbia River Basin, 'increased temperature may have profound negative
effects on salmonid populations.

SEDIMENT

The term "sediment" commonly refers to fine particles the size of clay and silt, but in the strict
sense, sediment includes all particles from colloids to boulders. Generally, however, it is the
fine sediments that are of concern because of possible detrimental effects on salmonid habitat,
whereas the coarser gravels, cobbles, and boulders help shape channel morphology and provide
substrate for cover and spawning. Logging activities can have major effects on the amount of
sediments delivered to streams and their subsequent routing downstream.

In mountainous terrain, sediments of all sizes are delivered to streams primarily by landslides
(Swanston 1991). These occur as slow-moving slumps and earthflows or as episodic debris
torrents and avalanches which happen during heavy rainfall when saturated soils trigger slope
failures. Undisturbed forest soils normally resist surface erosion because their coarse texture
and thick surface layer of duff and moss prevent overland flow.

Surface erosion in forested sites usually occurs only after the soil is bared by landslides, fire,
overgrazing, or logging (Swanston 1991). Compaction of soils by logging equipment increases
surface erosion by reducing soil infiltration and causing overland flow. Surface erosion is
greatly increased where disturbed or compacted soils are exposed to rainfall. Road surfaces,
landings, skid trails, ditches, and disturbed clearcut areas can contribute large quantities of fine
sediments to streams (Chamberlin et al. 1991). Nearly all forest operations disturb soil to some
degree. Road construction and maintenance, log hauling, tree felling, yarding, slash disposal,
and site preparation for replanting are all potential nonpoint sources of fine sediment pollution.

Construction of roads in steep terrain can substantially increase all types of soil erosion (Furniss
et al. 1991). Landsliding associated with roads can be more than 300 times more frequent than
in undisturbed forest, and because the landslides are relatively large, the amount of sediment
produced from roads greatly exceeds the sediment from forests and clearcuts (Fig. 5.3; Furniss
et al. 1991). The increase in landslides caused by roads depends on soil and bedrock type,

36 Effects of Logging

19 _j 32
(B)
ca 40

co CD
(D EIH. J. Andrews
(D [DAIder Creek
(D EMBlue River
CU 30 ElMapleton I
(D HMapleton 2
6
SCoastal B.C.
(D

CU E
20-
0 4

E
0
Z 10
2
CD

0
0 1 H N' M I 1.1N.-M 01 rl-@
Forest Clearcut Roads Forest Clearcut Roads

Figure 5.3. Comparison of the rate of landsliding (A) and mass soil erosion (B)
associated with forests, clearcuts, and roads as measu *red by six long-term studies of
more than 10 years. (Data are from Furniss et al. 1991: H.J. Andrews, Alder Creek,
and Blue River are in western Cascade Range, Oregon; Mapleton 1 and 2 are in the
Oregon Coast Range.)

steepness, and especially road location. Landsliding can continue for decades after the roads are
built.

Road failure is a concomitant risk of building roads in mountainous terrain (Furniss et al. 1991).
Landslides and severe gullying can result where roads intercept runoff and route it onto
hillslopes. The most common causes of landslides are improper placement of road fills,
inadequate maintenance, inadequate culverts, overly steep hillslopes, improper sidecasting, poor
road location, undercutting of slope support, and interception of surface and subsurface water.
Water intercepted by ditches can carry eroded sediment directly to streams (Fig. 5.4).

Culverts and bridges pose the greatest risk to fish of any road feature (Furniss et al. 1991).
When a culvert becomes plugged by debris or overtopped by high strearnflow, the stream can
be diverted, causing severe sedimentation. The risk of failure of a crossing structure depends
on its size compared to flood events (Fig. 5.5), and culverts should be sized to accommodate

Ef f ects of Loc
iging 37

I t44

Figure 5.4. Surface runoff being intercepted by a logging road and flowing down the
road's ditch after heavy rain in southeast Alaska. (Photo by B. Baker.)

100

Design life
80 -
110
0- 25-year
...... ......
. ..... .....
...... ......
Z, 1 00-year
60
CO ......
...... ......
...... ...... ..... .....
..... ..... ...... ......
...... ...... ..... .....
0
..... ..... ...... ......
...... ...... * * ... .....
..... ..... ...... ......
...... ...... ....
..... ..... ...... ......
...... ...... .....
..... .....
CL 40
...... ......
..... .....

...... ......

...... ...... ......
...... .....
Cz
LL 20 -
...... .....

70MM

0
5 10 15 20 30 50
Years

Figure 5.5. The failure probability of culverts designed for 25-year and 1 00-year flood
events. (Data are from FEMAT 1993.)

38 Effects of Logging

at least a 50-year flood (Furniss et al. 1991). Whatever the design life, any crossing structure
is virtually certain to fail if not maintained or removed when the road is abandoned.

Regular road use can cause chronic sediment inputs to streams nearly as great as during road
construction. The road surface can break down with repeated heavy wheel loads of hauling
trucks, particularly under wet conditions, resulting in a continual source of fine sediment (Fig.
5.6). In the Clearwater River, Washington, for example, the amount of sediment that washed
off roads equaled the amount from landslides (Reid and Dunne 1984).

Although most large landslides are associated with roads, many small landslides originate on
clearcut slopes as the root strength which holds soils together is lost within 2-10 years after trees
are cut (Burroughs and Thomas 1977; Ziemer and Swanston 1977). The increase in landslides
in clearcuts varies widely, depending on slope stability (Fig. 5.3). Because landslides tend to
be small, the amount of soil erosion is much lower than the erosion associated with roads, but
the greater total area of clearcuts makes this a substantial source of sediment (Swanston 1991).

Yarding operations can disturb ground over large areas (Everest et al. 1987a). Cable yarding
and helicopter systems that suspend logs generally cause minimal soil disturbance, but tractor
yarding can disturb and compact soils considerably (Fig. 5.7; Chamberlin et al. 1991). The bare
and compacted soil associated with tractor yarding can cause landsliding and surface erosion.
Effects of compaction and overturn of topsoil are also important because of potential long-term
reduction in soil permeability and productivity (Froehlich and McNabb 1984).

-A

Figure 5.6. Fine sediment produced by logging truck in southeast Alaska. (Photo
by T. R. Merrell, Jr.)

Effects oflog-ging 39

100

0
0- N
60

0 e
Cn
40
(D

0
Helicopter Skyline High lead Ground skid
cable cable tractor

Figure 5.7. Soil disturbance caused by different yarding methods. Each point
represents findings of a separate study. (After Chamberlin et al. 1991.)

For almost all forestry activities, as hillslope gradient increas es, so does the potential for
delivering sediment into streams. Hence, forest practices need to be tailored to site conditions
to minimize effects on slope stability. Unstable areas in watersheds should be identified, and
special precautions should be used.to avoid impacts in these areas.

Once delivered to streams, fine and coarse. sediments move downstream by different routes.
Fine sediment moves suspended in the water column, and coarse sediment moves as "bedload,"
rolling and bouncing along the stream bottom. Suspended sediment generally consists of clay
and silt which move rapidly downstream and deposit in slack-water areas 'and -on flood plains,
or infiltrate coarser substrate of the streambed (Beschta and Jackson 1979).. Bedload sediment
consists of coarse sand, cobble, and larger particles which move sporadically during floods and
are deposited as "wedges" behind 'structural feature's such as LWD (Fig. 5.8) and at channel
bends (Swanston 1991). Coarse sediment is sorted and. arranged by strearnflow to form an
66armor" layer on,the streambed, preventing bedload transport except for only a few days each
year during "flushing" strearnflows (Swanston 1991). Overturn of the upper layers of the
streambed at these times flushes fine sediments, redistributes bedload to form new pools and
riffles, and causes local accumulation of sediment deposits behind obstructions and at points of
reduced gradient.

40 Effects of Logaina

Cross section

Sediment wedge
po 0.
Large woody debris

Plan view

Large woody debris
Sediment wed e

Figure 5.8. Diagram showing storage of bedload sediment "wedges" behind large
woody debris. (After Swanston 1991.)

Undisturbed streams maintain a dynamic equilibrium between sediment delivery and routing, and
have abundant sediment stored in their channels (Everest et al. 1987a). Most streams, however,
do not remain undisturbed, even without human influence, but operate in context of natural long-
term disturbance cycles. Streams periodically receive large pulses of sediment and woody debris
during large storms, wildfire, and other watershed disturbances (Reeves et al. in press). Thus,
sediment delivery, routing, and storage are not static, and streams pass through natural long-term
cycles of excessive as well as depauperate amounts of sediment.

Past forest practices have changed the sediment equilibrium and storage in streams by increasing
hillslope erosion and causing a loss of structural channel features (Everest et al. 1987a). The
loss of structural features reduces storage and accelerates routing of bedload sediment
downstream (Fig. 5.9). Aggraded downstream reaches become wider, shallower, and more
prone to lateral migration and bank erosion (Sullivan et al. 1987).

The response of salmonid populations to increased sediment from logging is often difficult to
assess because of natural variability and the multiple effects of logging on stream ecosystems.
Much of the knowledge about sediment impacts is extrapolated from controlled laboratory
experiments (Everest et al. 1987a) and from field studies examining egg-to-fry survival under
natural conditions (e.g., Koski 1966; Tagart 1976).

Several field studies, however, have demonstrated significant adverse effects of sediment from
logging. In Carnation Creek, British Columbia, increased fine sediment after timber harvest

Effects of Logging 41

7,*- h- - @*",
A"in",
_4

Figure 5.9. Disturbance and loss of channel structures caused downcutting and
export of stored sediment from a headwater stream in southeast Alaska. (Photo by
T. R. Merrell, Jr.)

reduced chum salmon escapement by 25 % (Holtby and Scrivener 1989), and logging sediment
probably contributed to the general decline in chum salmon in the last 40 years in west
Vancouver Island (Scrivener 1991). Sediment from extensive logging in the South Fork Salmon
River basin in Idaho buried spawning and rearing habitats. A logging moratorium was begun
in 1966 which allowed conditions to begin to improve (Platts and Megahan 1975). By 1979,
the percentage of fine sediment in spawning areas had decreased from 30% to 8% and gravel
increased from 32% to 68% (Sullivan et al. 1987), In the Queen Charlotte Islands, British
Columbia, 45-86% of salmon eggs were destroyed by scour in watersheds with severe mass
wasting, compared to only 0-14% in more stable areas (Tripp and Poulin 1986). In a tributary
of the Clearwater River, Washington, sediment from a debris torrent and streamside salvage
logging aggraded the stream channel to the point that the stream dried up in summer because the
water level dropped below the level of deposited sediment; coho smolt yield decreased 60-86%
(Cederholm and Reid 1987).

These examples indicate the range of direct adverse effects that increased fine sediment can have
on salmonid populations. Fine sediment can directly reduce egg-to-fry survival, food
production, summer rearing area, and winter survival (see Chapter 4). Less-direct effects
include changes in stream channel morphology and stability, causing long-term reductions in
carrying capacity and survival.

42 Effects of Logging

STREAMFLOW

Cutting trees and building roads can alter the watershed's water balance and accelerate
movement of water from hillsides to stream channels (Chamberlin et al. 1991). Of greatest
concern are the changes in low flow (base flow) in summer and peak flow during rainstorms and
snowmelt.

Base flow increases after timber harvest because removing trees increases soil moisture and
groundwater as less vegetation results in less transpiration and i 'nterception on foliage. Base
flow usually increases most in summer when transpiration has the greatest effect on soil
moisture. In fail and winter, the soil is usually saturated, and runoff is'similar in both clearcuts
and forests. Increases in base flow are short lived, decreasing as vegetation recovers (Fig.
5. 10). After 10-30 years, base flow may return to normal or decrease below pre-harvest levels
because rapidly growing second-growth hardwoods transpire more water than mature trees (Fig.
5.11; Hicks et al. 1991b). Increased base flow can benefit salmonids by maintaining higher
water levels (Hetherington 1988), but decreased base flow can shrink available habitat, especially
at critical low-flow periods in late summer.

Peak flows increase after logging because water is routed more quickly to stream channels
(McIntosh et al. 1994). Activities that disturb and compact the soil increase surface runoff
which reaches streams faster than subsurface flow. Ditches along roads collect runoff and
intercept subsurface flow and route it quickly to streams (Fig. 5.4). Roads act as first-order
streams and channel more water directly into larger streams (Wemple 1994). More snow
accumulates in clearcuts and melts earlier and faster, causing more severe rain-on-snow events
and higher and.earlier peaks during spring snowmelt (Harr 1986; Golding 1987). Increased peak

600
E
E 500 -

2 400 -

ca J00 -
3:

200 -

as
T) 100 -

0
0 5 10 15 20
Years after clearcut logging and burning

Figure 5.10. Diminishing increase in water yield after timber harvest in southwest
Oregon. (After Harr 1983.)

Effects of Logging 43

E
E, lo - Increased

5-
(D
ca

0-
mr-M

M -5
Decreased
-10 L . . . . . . . . . . . I . . 1
1953 1958 1963 1968 1973 1978 1983 1988
- ------------
Pre-logging Clearcut Hardwoods established
& burned

Figure 5.11. Change in August water yield from a western Oregon watershed after
logging. Water yield initially increased after the watershed was clearcut and burned,
but decreased after dense second-growth hardwoods became established. (After
Hicks et al. 1991 b.)

flow is detrimental for fish habitat because the resulting bedload overturn can scour stream
channels, kill incubating eggs (McNeil 1964), and displace juvenile salmonids from winter cover
(Tschaplinski and Hartman 1983).

LARGE WOODY DEBRIS

Large woody debris is an integral part of streams in forested watersheds, providing structure to
the stream ecosystem and important habitat for salmonids (Bisson et al. 1987). It plays
important roles in controlling stream morphology, regulating storage of sediment and particulate
organic matter, and creating and maintaining fish habitat. Removal of LWD results in
immediate loss of important habitat features and a decline in salmonid abundance (Hicks et al.
1991a). Debris removal destabilizes the stream channel and eliminates pools and cover. The
increased riffles may favor underyearling steelhead and cutthroat trout, which prefer riffle
habitat, but the loss of pools harms coho salmon and older steelhead and cutthroat trout (Bisson
and Sedell 1984; Murphy et al. 1986).

44 Effects of Loggingg

Logging activities can reduce LWD in several ways (Bisson et al. 1987). Existing LWD can
be destabilized during tree felling and yarding, and later exported downstream or onto the
floodplain. Salvage of downed merchantable logs from the stream channel and floodplain
removes LWD and destabilizes what is left. Cleaning of stream channels after yarding also
removes LWD and destabilizes channels (Bilby and Ward 1989). Even if left undisturbed, LWD
declines over time if riparian trees are cut because second-growth vegetation provides insufficient
new conifer debris to replace the key pieces as they decay or wash downstream (Andrus et al.
1988; Murphy and Koski 1989). The "key pieces" of debris that create stable habitat in streams
have been hard hit by past logging. Many streams in second-growth forest become progressively
debris-poor as total LWD declines and changes to mostly small pieces of alder (Alnus spp.).

Woody debris in streams is depleted by decay, fragmentation, and export to downstream reaches
and floodplains (Bisson et al. 1987). The depletion rate depends mostly on size and species of
wood and type of stream. Woody debris from Sitka spruce (Picea sitchensis) and western
hemlock (Tsuga heterophylla) in southeast Alaska is naturally depleted by about 1-3% per year
in valley-bottom streams (Murphy and Koski 1989). Some species, such as redwood and
western red cedar (Thuja plicata), last much longer.

Removal of trees from the riparian area during logging causes a long-term reduction in the
recruitment of new LWD (Bisson et al. 1987). _ After clearcutting in riparian areas, second-
growth trees are the principal source of new woody debris. In young forest stands, inputs of
debris large enough to provide stable habitat is low for the first 50-75 years (Grette 1985;
Andrus et al. 1988; Murphy and Koski 1989), and could remain low much longer if riparian
areas- are converted to non-conifer species (Chan 1993). Accumulations of LWD continue to
decrease (Bryant 1985) as large key pieces are depleted.

Effects of timber harvest in riparian areas can last hundreds of years. If all sources of new
LWD are removed by clearcutting, the key pieces of large LWD in the stream will disappear
over a period of about 250 years (Murphy and Koski 1989). Because new key pieces from
second growth do not begin to enter the stream for, 60-80 years after logging, LWD begins a
long-term decline that may not bottom for nearly 100 years and not recover for more than 250
years. Timber harvest rotations of less than 100 years will permanently eliminate large LWD
unless streams are protected by adequate buffer zones.

A model of LWD input and depletion was used to demonstrate long-term effects of clearcutting
without buffer zones in southeast Alaska (Murphy and Koski 1989). The model showed that if
trees were not left along streams during timber harvest, the LWD in a stream would be reduced
by 70% after 90 years, and would take more than 250 years to recover (Fig. 5.12). An uncut
30-m buffer zone would maintain LWD over the long term, whereas narrower buffers or
partially harvested buffers would cause LWD to decline. This model needs to be adjusted to
apply to regions outside southeast Alaska, but the principle applies elsewhere.

Effects of Logging 45

120- Second-Growth Trees ......I
J

100--l - - -- - -- - -- - -- - -- - --J I-
2 80- Pre-logging LWD li Total LWD
0)
-6
o 60-

0
40- Post-logging LWD
2
20-
......... . ......... . .;1
0
0 50 160 150 200 250
Years after clear-cutting with no buffer strip

Figure '5.12. A model of the changes in large woody debris after clearcut logging
without a riparian buffer zone in southeastern Alaska. (After Murphy and Koski 1989.)

MIGRATION HABITAT

Changes in stream channels after road construction and timber harvest can interfere with fish
migration by blocking passage through culverts at stream cro�sing@, causing logjams, decreasing
cover from predators, decreasing the frequency of large pools used for resting, and adversely
affecting temperature and DO.

Decreases in large pools and cover because of the loss of LWD can expose migrating adults to
predation and deprive them of resting habitat. Suitable large pools are usually in limited supply
along a stream, so that each is important and often will hold large numbers of migrating adults.
Logging activities can decrease the frequency of large pools by decreasing the frequency of
"key" pieces of LWD. In Oregon and Washington, frequency of large pools decreased by
nearly two-thirds between the 1930s and the late 1980s (Fig. 5.13).

Culverts can be a barrier to upstream fish migration, especially if installed above the grade of
a stream (Furniss et al. 1991). Poorly installed culverts not only block migrations of adult
salmon returning to spawn but also impede seasonal movements of juvenile fish between summer
and winter rearing areas within a watershed. Culvert conditions that block fish passage include
too high a water velocity, too shallow a water depth, lack of a resting pool below the culvert,
and too high a jump to the culvert (Furniss et al. 1991). A single poorly installed culvert can
eliminate the fish population of an entire stream system.

46 Effects of Logging

El 1935-1945
a) 5
1987-1992
CD
E
....... .......
0 ... ...
4
....... ...

....... . ....
E
....... .........
CU .......
3 .......

....... ........
CD 2
CL
........ .... .......
C0
75

....... .......
0 ... .......
Cascades Coast Eastern Coast Blue Mtns
Washington Oregon

Figure 5.13. Changes in pool frequency l5etween 1935 and 1992 in Oregon and
Washington. (Data are from FEMAT 1993 and McIntosh et al. 1994.)

FOOD AVAILABILITY

A potential benefit of timber harvest results from the increased light when forest canopy over
the stream is opened up. This can stimulate aquatic primary production and increase food for
fish (Murphy and Meehan 1991). Forest streams are often light limited, and studies from
California to Alaska have shown increased algal production after canopy removal. The increased
algal production results in more abundant benthic invertebrates which juvenile salmonids eat.

Although energy sources change after timber harvest, the dominant macroinvertebrates and
functional feeding groups usually remain unchanged (Hawkins et al. 1982; Duncan and Brusven
1985). Insects that feed by collecting fine detritus particles dominate in both shaded and open
stream reaches. This is because the increased algal production in open reaches is used mostly
as organic detritus after the algae sloughs from rocks (Murphy et al. 1981). The algae-derived
detritus is more nutritious than detritus from forest litter. Thus, canopy removal can increase
the abundance of invertebrates by enhancing the food quality of detritus.

Effects of Lociaing 47

Summer density of fry of some salmonid species often increases during the first 10-15 years
after timber harvest because of increased production of invertebrates (Murphy and Meehan
1991). Where food is limiting and other habitat factors are suitable, density of coho salmon fry
in summer is directly related to the abundance of algae (Fig. 5.14) because increased algal
production increases energy flow through the food web. The higher density of coho salmon fry
probably results from smaller feeding territories (Dill et al. 1981). Other habitat features,
however, must also be suitable for fry density to respond favorably to increased food. Other
salmonids, furthermore, may not respond, the same way as coho. In Carnation Creek, B.C., for
example, juvenile coho increased after logging, but steelhead, cutthroat trout, and chum salmon
decreased (Hartman 1988; Holtby 1988; Scrivener and Brownlee 1989).

Old Growth
30 -
Buffered
E Clearcut
0
25

0
20 -

15 -
70

0 10

0
U 5

0-
0 2 4 6 8 10 12
Algae biomass (mg/m 2

Figure 5.14. Relationship between density of coho salmon fry and algae biomass in
old-growth, buffered, and clearcut reaches of streams in southeast Alaska in summer.
(After Murphy et al. 1994.)

48 Effects of Logging

Timber harvest can affect availability of leaf detritus by altering riparian vegetation and physical
conditions in the stream (Gregory et al. 1987). Effects vary as the riparian plant community
passes through stages of recovery. Litterfall from streamside vegetation decreases by 75 %
immediately after riparian timber harvest, but recovers quickly with the growth of deciduous
shrubs. Loss of debris dams because of decreased LWD reduces the channel's storage capacity
for organic matter, resulting in reduced food resources and habitat for aquatic invertebrates.
Increased temperature accelerates microbial decomposition of organic matter, which can promote
increased invertebrate production and lead to more fish food (Warren et al. 1964).

An important long-term effect of clearcut logging is potential overshading from second-growth
canopy (Murphy and Hall 1981; Sedell and Swanson 1984). Second-growth vegetation produces
a denser shade and lacks the canopy gaps that are common in old-growth forest (BJornn et al.
1992). Thus, increased stream production in the first 20 years after timber harvest may be
followed by a much longer period of depressed production.

CUMULATIVE EFFECTS

Cumulative effects result from the combined effects of separate management activities through
time and space (Burns 1991). Although individual management activities by themselves may not
cause significant harm, incrementally and collectively they may degrade habitat and cause long-
term declines in fish abundance (Bisson et al. 1992). Effects of individual actions, such as
dispersed, separate harvest units and road building, should be considered in the context of all
other previous and ongoing activities in the watershed.

Changes in sediment dynamics, strearnflow, and water temperature are not just local problems
restricted to a particular reach of stream, but problems that can have adverse cumulative effects
throughout the entire downstream basin (Sedell and Swanson 1984; Grant 1988). For example,
increased erosion in headwaters combined with reduced sediment storage capacity in small
streams can overwhelm larger streams with sediment (Bisson et al. 1992; Fig. 5.15). Likewise,
increased water temperature in headwater streams may not harm salmonids there but can make
water too warm downstream (Bjornn and Reiser 1991).

Cumulative effects on sediment and hydrology worsen as the area affected by timber harvest
increases (Rhodes and McCullough, in press). The amount of sediment delivered to streams and
fine sediment in pools increase with increasing timber harvest and road construction (Chen 1992;
Lisle and Hilton 1992; Fig. 5.16). Water yield increases in proportion to the area devegetated
(Harr 1983), and peak flows increase in proportion to roads and soil compaction (Harr et al.
1979; Fig. 5.17). Pool depth and frequency, LWD, and channel complexity decrease with
increased logging (Fig. 5.18; Bisson et al. 1992; Reeves et al. 1993).

Effects of Loqqft 49

-7-7 --r

.... . .. ...

Figure 5.15. Fine sediment derived from upstream timber harvest being carried by a
headwater stream into downstream fish habitat in foreground. (Photo by T. R.
Merrell, Jr.)

E
50 -

(D
u) 40-

30 -
E
7@5 20 -

-0
0- 0
0 20 40 60 80 100
Percent of watershed logged

Figure 5.16. The relationship between amount of fine sediment in pools and level of
.timber harvest and road construction in several northern California watersheds.
Vertical bars are one standard erro r (After Lisle and Hilton 1992.)

50 Ef f ects of Logging

CO 40
CL
C:
30

Cn
Cz
20

C:
,0 10-
0,

0
0 5 10 15 20
Percent of watershed compacted

Figure 5.17. The relationship between peak streamflow and the area of a watershed
affected by roads and soil compaction in western Oregon. (After Harr et al. 1979.)

2.5
E
Cz
CD
2

0
E 1.5
Cn
0 a)
CL 1
C5
IC
0.5

0
0 20 40 60 80 100
Percent of watershed logged

Figure 5.18. Frequency of pools associated with large woody debris in ten Oregon
coastal streams with different levels of logging. (After Bisson et al. 1992.)

Effects of Loggin 51

The most pervasive cumulative effect of past forest practices has been an overall reduction in
habitat complexity (Bisson et al. 1992). Habitat complexity has declined principally because of
Xeduced size and frequency of pools due to filling with sediment and loss of LWD (Fig. 5.19;
Reeves et al. 1993; Ralph et al. 1994). This cumulative habitat simplification has caused a
widespread reduction in salmonid diversity (Fig. 5.20). A few fish species were favored by the
changes in habitat, whereas others declined or disappeared (Reeves et al. 1993). A similar
pattern of decreased diversity of fish communities has been observed in streams altered by other
human activities, such as agriculture (Schlosser 1982; Berkman and Rabini 1987) and
urbanization (Lcidy 1984; Scott et al. 1986).

Figure 5.19. A stream in
southeast Alaska 15 years
after timber harvest
without a buffer zone,
showing extreme reduction
in pools and habitat
complexity due to loss of
large woody debris. (Photo
by M. Murphy, NMFS-)

52 Effects of Logging

Harvest Level
1.8- M LOW
M High
CD
1.6 - ........

cO 1.4 ......
........ ........

0 .........
....... ........
E 1.2
CZ
........ ........
....... ........
CD
F 7 77 - 7 -1
<500 ha >500 ha
Sandstone Sandstone Basalt Basalt
Central Coast South Coast

Figure 5.20. Salmonid species diversity (inverse of the Berger-Parker index) in relation
to level of timber harvest (low = < 25% logged; high = > 25% logged) in 14 coastal
Oregon watersheds of different size, rock type, and geographic location. (After
Reeves et al. 1993.)-)

SALVAGE LOGGING

Salvage logging after catastrophic events, such as wildfire, windthrow, flooding, or insect
damage, is often detrimental for salmonid habitat because of the importance of large woody
debris for fish habitat and the possible harmful effects of disturbing riparian areas. Catastrophic
events are part of the natural disturbance regime which helps maintain ecosystem diversity
(Everett et al. 1994). Research on effects of fire, for example, shows that riparian areas are the
first to recover from catastrophic events (A. Youngblood, FS, Bend, OR, pers. comm. 1994)
and may actually benefit from being burned (W. Minshall, Idaho State University, Pocatello,
ID, pers. comm.'1994).

Salvage logging in riparian areas after fire should usually be avoided because the areas.are then
extremely fragile and can not withstand roading, yarding, and other salvage activities @Minshall
et al. 1989; Minshall et al. 1990; Minshall and Brock 1991). Wildfire dramatically increases
runoff and fine sediment while decreasing shading and cover from undercut banks and woody
vegetation. Salvage logging can exacerbate. these impacts. Under postfire salvage conditions
on the east slope of the Cascade Mountains, traditional logging systems, such as tractor skidding
over bare ground and cable skidding, cause more severe soil disturbance and erosion than
LM

advanced systems, such as skyline, helicopter, and tractor skidding over snow (Klock 1975).

Effects of Loggin 53

A likely impact of timber, salvage after wildfire is an increase in @ water runoff, erosion, and
landslides because of increased snow accumulation and faster snowmelt where trees (even dead
ones) have been removed (Megahan 1983). Factors influencing snow accumulation and melt,
rather than evapotranspiration, dominate the spring hydrologic regime in interior areas
(Swanston 1991). This is because the moisture deficit from'-evapotranspiration is satisfied by
rain in fall and early winter. Salvaging dead trees increases snow accumulation because of
changes in winter snowmelt, reduced interception in standing trees, and especially, aerodynamic
factors affecting snow deposition (Megahan 1983). The increased snow pack results in increased
soil saturation which can trigger landslides, increase runoff intercepted by roads, and exacerbate
the scour of stream channels and downstream sediment transport common after wildfires
(Minshall et al. 1989, 1990).

A study of the effects of helicopter logging (Megahan 1983) shows the potential effects of timber
salvage after wildfire. Megahan (1983) compared two headwater drainages in the Idaho
Batholith: one was clearcut and yarded by helicopter; one was an unharvested control; and both
were burned by a hot wildfire. In the logged-and-burned watershed, snow accumulation
increased 41%, spring melt rate increased 30%, the subsurface flow intercepted by roadcuts
increased 96%, and peak flows increased 27%. None of these effects were detectable in the
burned-only watershed.

The implication of Megahan's (1983) study is that even helicopter salvage of standing dead trees
after wildfire can increase the risk of landslides. Landslide hazard is directly proportional to
the depth of the saturated zone relative to soil depth, and most landslides begin after intense rain
or rapid snowmelt creates a temporary water table and high pore-water pressure in the soil
(Swanston 1991). In Megahan's (1983) study, soil moisture and subsurface flow increased much
more in the logged-and-burned watershed than in the burned-only watershed. Coupled with the
declining cohesive strength of decaying tree roots, increased soil saturation after timber salvage
can seriously increase landsliding (Megahan 1983).

Besides potential problems with sediment production, salvage logging can also retard attainment
of riparian management objectives by removing trees that are sources of LWD for the stream *'
In fire-climax ecosystems on the east slope of the Cascade Mountains, new debris principally
enters the stream in pulses after fire, rather than by slow continuous recruitme 'nt (Minshall et
al. 1990). Cutting and removing trees from the riparian area would leave fewer trees to,replace
the stream's debris as it is depleted by decay, fragmentation, and transport.

New sources of large woody debris are critical to the stream's post-fire recovery (Minshall et
al. 1989, 1990). After fire, existing woody debris in the stream channel is often removed by
high stream discharge and exported downstream or deposited along the floodplain. The export
of woody debris reduces storage capacity of the stream channel for sediment. As a result, stored
sediment is exported and the stream channel becomes deeper and streambed becomes coarser.
The sediment exported downstream comes from both increased hillslope erosion and erosion of
stream channels. Beginning after about 2 years, new woody debris gradually begins to.
accumulate in stream channels from the undercutting and blowdown of fire-killed trees. This
large debris serves as accumulation points for sticks and fine detritus, forms pool habitat, and

54 Effects of Logging

creates new storage sites for sediment, helping to slow the downstream transport of fine
sediment.

Large woody debris from fire-killed trees has important roles in sediment routing, not only in
streams, but also on hillslopes (Wilford 1984). As the fire-killed trees fall or blow down across
the slope, they form cross-slope obstructions. Sediments and small debris from upslope mass
movements are deposited behind these obstructions, forming a series of terraces which delay the
delivery of sediments to stream channels. Salvage of fire-killed trees could reduce the formation
of these beneficial sediment-storage elements on hillslopes, resulting in gully erosion and
transport of previously stored sediments into stream channels.

Although salvage logging can have adverse effects on stream ecosystems, it might be warranted
in some situations. Effects of wildfire and insect outbreaks under current forest conditions can
be more severe than in natural landscapes because of years of fire suppression (Amo and Ottmar
1994; Mason and Wickman 1994). Therefore, some management activities, including salvage
logging, might help to ease the transition to a more natural disturbance regime (S. Chan, FS.
Corvallis, OR, pers. comm. 1994).

Salvage of insect-killed trees in riparian areas can be justified in some situations to protect
integrity of riparian vegetation from further insect damage (Daterman 1994). Removal of
infested trees from riparian areas, however, would probably be unsuccessful in stopping insect
damage because 1) not all infested trees can be found and removed; 2) infested trees are usually
removed afterIthe beetles have emerged in spring; and 3) pest management on a "stand level"
is ineffective because of the beetle's strong flight capability. To improve success in controlling
insect epidemics, a watershed-scale pest management plan for the ecosystem must be,
implemented on a landscape scale (Daterman 1994).

Salvage to reduce fuel loads might also be justified in som'e situations. Fish may be killed when
riparian areas along small streams bum in high-intensity fires (Minshall and Brock 1991).
Salvage of a proportion of insect-killed trees may be beneficial in reducing risk of high-intensity
fires in some riparian areas.

Chapter 6
Technical Foundation of Forest Practices

The challenge for watershed management is to sustain all forest resources, processes, and
ecosystem linkages while enabling economic timber production. Forest practices rules for
protecting fish habitat fall into three basic categories (Belt et al. 1992): 1) buffer zones; 2) Best
Management Practices (BMPs); and 3) cumulative effects management. This chapter examines
these rules and their technical foundation.

BUFFER ZONES

Buffer zones (also called riparian management areas, stream protection zones, etc.) are lands
immediately adjacent to streams or lakes designated to protect aquatic resources (Fig. 6. 1).

Figure 6.1. A riparian buffer zone along an anadromous fish stream in the Tongass
National Forest, southeast Alaska. (Photo by K Koski, NIVIFS.)

56 Te6hnical Foundat on

These areas receive special management consideration, but are not necessarily "lock-out" zones;
timber harvest and road crossings are often permitted but with restrictions to protect aquatic
resources. Restrictions are generally tighter on public than on private lands. Under the federal
Northwest Forest Plan, for example, buffers can only be modified if watershed analysis
demonstrates that a modification is needed to attain ecosystem management objectives (USDA
and USDI 1994a).

Buffer zones are administratively defined as the area within some distance from the stream
channel in which protection of water quality and fish habitat is given highest management
priority. The emphasis in defining riparian buffers for fish habitat is on ecological functions
from the perspective of the stream, not on botanically defined riparian plant communities. Thus,
a riparian buffer zone usually includes both upland forest and distinct riparian vegetation.
Buffers may also be designed to benefit wildlife and other non-fish aquatic species in addition
to anadromous fish.

An understanding of the influence of riparian vegetation on streams is fundamental to
understanding the function and effectiveness of riparian buffer zones. Small streams are
intricately connected physically, chemically, and biologically to their riparian zones (Meehan et
al. 1977; Murphy and Meehan 1991). Roots of streamside vegetation stabilize stream banks,
retard erosion, affect nutrients in groundwater, and create overhanging cover. Vegetation and
downed woody debris dissipate stream energy during floods and obstruct movement of sediment
and organic matter. The canopy provides leaves and other organic matter that are part of the
energy base for the stream ecosystem, and its shade limits algal production and moderates stream
temperature. The trees and other LWD that fall into the stream channel provide the principal
structural features that shape the stream's morphology, linkages to the flood plain, habitat
complexity, streambed materials, and other characteristics (Sullivan et al. 1987; Beschta 1991).

Small perennial and intermittent non-fish streams are especially important in routing water,
sediment, and nutrients to downstream fish habitats (Reid and Ziemer 1994a). Intermittent
streams account for more than one-half of the total channel length in many watersheds in the
Pacific Northwest and Alaska, so they strongly influence the input of materials to the rest of the
channel system. These small channels store large volumes of hillslope materials and release
them over long periods. Much of the sediment eroded from hillslopes during a major storm may
be stored in the smallest channels and released gradually, thereby lessening the harm on
downstream habitats. These sites can beparticularly important as potential sediment sources
because they are often susceptible to gullying and debris flows. Intermittent channels and
unchanneled swales associated with them often are areas of considerable potential instability.

Many functions of riparian vegetation, decrease with increasing distance from the streambank
(FEMAT 1993). The point where the graph of cumulative effectiveness reaches 100% indicates
the distance over which a function operates (Fig. 6.2). A standard way of measuring functional
distance is by considering the height of mature trees growing along the stream. For example,
the contribution of root strength to maintaining streambanks operates within a distance of 0.5
tree height. Inputs of leaves and particulate organic matter come mainly from the area within
0.6 tree height. Shading and large woody debris are derived from a distance of about I tree
height.

Technical Foundation 57

100 Root . . . . . . ..................
-0-0 strength*"@-

a) Shade

Litter
50 fall
Large woody
debris input

0
0 0.215 0.5 0.75 1.0
Distance from channel (tree height)

Figure 6.2. The cumulative effectiveness of various functions of riparian vegetation
in relation to distance from the streambank in western Oregon. (After FEMAT 1993.)

The same types of relationship also determine the buffer width needed to attenuate changes in
microclimate from the adjacent logging unit (Fig. 6.3). The width needed to buffer changes in
solar radiation, for example, is less than 1 tree height, whereas the width needed to attenuate
changes in wind speed and relative humidity is about 3 tree heights. Such changes in
microclimate may be important in determining the long-term viability of the buffer and in
determining the suitability of the buffer for riparian-dependent plants and wildlife (Hibbs et al.
1991; FEMAT 1993).

The concept of "site-potential tree height" (the average maximum height possible given site
conditions) can be used to adjust buffer width for differences in site productivity (FEMAT.
1993). Tree height depends on local growing conditions, and tree height largely determines the
distance over which ecological functions operate (FEMAT 1993). This height can be determined
from the location's site index and silvicultural data on mature forests that develop on that type
of site. The site-potential tree height provides a standard measure of the way many riparian
functions, such as providing woody debris, decrease away from,the stream bank in different
areas.

The area of influence of riparian vegetation also depends on channel constraint and floodplain
development (Sparks et al. 1990). Streams constrained within bedrock channels with minor
flood plains have a restricted zone of interaction with riparian vegetation (Gregory and Ashkenas
1990). In contrast, unconstrained, valley-bottom streams with extensive flood plains interact
with riparian vegetation over a much broader area.

58 Technical Foundation

100
1-0 Radiation
0-
Air temR..-,;;-
CO
CA 75

Wind speed
.50

Relative
humidity
25
Co op
75 J-0
E
=3
0 0
0 1 2 3
Distance from stand edge into forest (tree heights)

Figure 6.3. The cumulative effectiveness of various functions of forest vegetation in
relation to distance from the edge of adjacent clearcuts in western Oregon. (After
FEMAT 1993.)

Buffer Design

Forest practices rules regulate two features of buffer zones: their width and timber harvest within
them. , In general, there is little controversy about using buffers to maintain aquatic resources,
but there is some controversy about how wide buffers should be and how they should be
managed (Johnson and Ryba 1992).

The recommended width for buffer zones depends on management objectives (Johnson and Ryba
1992). If a specific function is targeted for protection, the width can be determined by that
requirement. If several f@nctions are targeted, the function with the widest requirement can
decide buffer width. In practice, the narrowest buffers are used along non-fish streams where
management objectives are primarily to maintain water quality. The widest buffers are used
along fish-bearing streams to not only protect the stream, but also to maintain integrity of the
riparian vegetation (i.e., "to put a buffer on the buffer;" Cederholm 1994).

Buffer width can be fixed or variable (Belt et al. 1992). Fixed-width buffers are more easily
enforced and require fewer specialized staff, whereas variable-width buffers allow tailoring of
forest practices to site-specific conditions (Bisson et al. 1987; Bradley 1988). Fixed
prescriptions may be less effective than management techniques adapted to local topography and
natural disturbance regimes (Naiman et al. 1991). In practice, agencies use a hybrid system of
fixed-width buffers where the required width varies across a small number of categories, based

Technical Foundation 59

on beneficial use, stream width, or hillslope gradient. Some rules (e.g., Oregon's) also allow
buffer width to vary along a given stream, as long as it averages above the required minimum
(ODF 1994).

Prescriptions for timber harvest within buffer zones range from complete no-harvest to complete
harvest. As with buffer width, the amount of timber harvest allowed within buffer zones
depends on management objectives. More harvest is allowed along small, non-fish streams.
managed for water quality than along streams managed for fish habitat. Buffers can be managed
to achieve objectives or left as unmanaged, no-harvest zones. Managed buffers are more flexible
to account for local conditions and better able to implement restoration in degraded areas. No-
harvest unmanaged buffers are more easily administered and less costly in time and personnel.

Factors Affecting Buffer Effectiveness

Effectiveness of buffer zones has been evaluated primarily for four basic functions: 1) filtering
sediment, 2) providing shade, 3) providing LWD, and 4) overall protection of fish habitat.

SEDIMENT FILTERING

Effectiveness of vegetation in filtering sediment has mainly been evaluated for filter strips below
roads, which are generally the largest source of sediment (Belt et al. 1992). These evaluations
are particularly relevant because roads are oftenlocated next to streams with intervening buffer
strips. For example, a survey of Idaho forest practices (Idaho Water Quality Bureau 1988)
found that existing roads near stream channels were, the most important factor contributing to
degradation of water quality.

The key factors controlling sediment filtering are slope and density of obstructions (i.e., woody
debris and ground vegetation). The steeper the side slope, the wider the buffer should be to
filter sediment. Belt et al. (1992) and Johnson and Ryba (1992) reviewed numerous studies
whose recommendations ranged from 25 ft (7.6 m) for 0% slopes to 200 ft (61 m) for steep
(>50%) slopes. Johnson and Ryba (1992) recommended a 100-ft (30-m) buffer to filter
sediment. Regardless of width, buffers are ineffective at stopping sediment that moves through
them in gullies and small stream channels (Duncan et al. 1987).

SHADING

The shade-producing canopy is a key function of riparian vegetation in moderating stream
temperature (Beschta et al. 1987). Other factors that affect shading include stream size,
orientation,. local topography, tree species, stand age, and stand density. The relationship
between canopy density and buffer width is variable (Fig. 6.4), but buffers that are 30 m-wide
provide about the same shade as in old-growth forest.

Shade for the stream can be provided by unmerchantable trees and, along small streams, even
strearnside shrubs. Thus, buffers can be selectively harvested and still function effectively as
shade. This may be adequate for non-fish streams managed only for water quality, but it would
be inadequate for fish-bearing streams. Early riparian buffers used in the 1970s and 1980s were

60 Technical Foundation

100

80
0-01
2@%
60 -

40 -
0

Ca
20

0j
0 10 20 30 40 50
Buffer width (m)

Figure 6.4. The relationship between canopy density and buffer width (uncut buffers)
in western Oregon. Square symbols and solid line represent data from (A) Brazier and
Brown (1973); diamond symbols and dotted line represent data from Steinblums et al.
(1984). (After Beschta et al. 1987.)

mainly designed to prevent adverse increases in temperature (Beschta et al. 1987). Little
consideration was given to other important features of fish habitat; consequently, these buffers
failed to adequately protect fish habitat (Phinney et al. 1989).

PROVIDING LARGE WOODY DEBRIS

After the importance of LWD was recognized in the 1980s, efforts were made to design buffers
to provide for long-term maintenance of LWD in streams. Studies focused on determining
where LWD comes from (i.e., LWD recruitment) and how long it lasts in streams.

The basis for determining width of buffer zones for maintaining LWD was the "source distance"
measured from the streambank to the spot where the tree once stood. The probability of a tree's
falling into a stream decreases rapidly with increasing distance from the stream (Robison and
Beschta 1990). In southeast Alaska, 99% of LWD is recruited from up to 30 m (100 ft) away
from the stream (Fig. 6.5; Murphy and Koski 1989). The NMFS Alaska Region issued a policy
statement in 1988 calling for 30-m, no-harvest buffer zones along streams in Alaska to protect
LWD sources (USDC 1988). In western Oregon and Washington, LWD can be recruited from
up to 55 m (180 ft) away (McDade et al. 1990). Thus, a wider buffer is needed in western
Oregon and Washington than in Alaska to provide the same protection for LWD sources.

Technical Foundation 61

100 - - - - - - -

(D
80-

0
U) Cumulative %
60

0 40-

(D
2 20-

0 EUM
<1 1-5 6-10 11-15 16-20 21-25 26-30 >30
Distance from stream to LWD sources (m)

Figure 6.5. Source distance for large woody debris (LWD) in southeast Alaska. (After
Murphy and Koski 1989.)

Source distance of LWD also depends on the type of stream (Lienkaemper and Swanson 1987;
Murphy et al. 1987). Valley-bottom streams in unconstrained channels receive much of their
LWD from the immediate streambank because the stream can undercut trees lining its banks.
Over the long term, an unconstrained stream can wander across its flood plain and undercut trees
growing far from its present channel. A constrained stream's channel consists of bedrock and
is therefore more stable, and trees along the banks are safer from undercutting. Constrained
streams, however, often have steep adjacent slopes, and LWD can slide into the stream from far
away.

Selective harvest possibly could be used to harvest valuable trees within buffer zones without
decreasing LWD sources if selected 'trees are unlikely to fall into the stream (Robison and
Beschta 1990). Such harvest, however, must be done carefully to avoid damaging remaining
trees. Further, too much harvest can open up buffers to wind damage and exacerbate potential
succession to shrub vegetation (Hibbs et al. 1990). The most stable buffers have a dense stand
of trees rather than individual trees protruding above an understory (Johnson and Ryba 1992).

Selective harvest needs to leave trees that are large enough to Provide stable LWD. The size
Of LWD needed to form stable habitat depends on stream size (Fig. 6.6; Bisson et al. 1987).
Small streams less than 5 in wide need LWD that is at least 30 cm in diameter and 5 in long;
large streams more than 20 m wide need LWD that is at least 60 cm diameter and, 12 m long.
These "key pieces" of LWD serve as "anchors" to trap and -stabilize other smaller pieces. Loss
of such large "key pieces" of LWD reduces stability of LWD accumulations and diminishes the

62 Technical Foundation

-70 -
,E

E 50
co

40 -

C) 30 -

20
0 5 10 15 20 25
Channel width (m)

Figure 6.6. The diameter of stable large woody debris as a function of channel width.
(After Bisson, et al. 1987.)

beneficial functions of LWD in the stream (Heimann 1988). Once the key pieces are gone, the
smaller LWD will not remain in place for long.

Selective harvest within buffers offers an opportunity to restore degraded riparian vegetation in
second-growth areas (Bilby and Bisson 1991). Degraded riparian areas can be improved by
appropriate silviculture if rules allow entry into buffer zones during timber harvest (T. O'Dell,
Simpson Timber Company, Korbel, CA, pers. comm.. 1994). By using silvicultural treatments,
such as patch cutting, thinning, and conifer planting, alder-dominated riparian areas can be
treated to restore vegetative diversity and provide LWD for recovery of productive stream
habitat (Bilby and Bisson 1991). Active management of riparian areas may be necessary to meef
the long-term needs of fish habitat (Sedell et al. 1999). Reestablishing conifers in riparian areas
offers potential long-term benefits for both fisheries and timber managers.

OVERALL EFFECTIVENESS OF BUFFER ZONES

Evaluating the overall effectiveness of riparian buffer zones is difficult because of the long time
periods involved for impacts to occur and for ecosystems to recover. Full impacts on LWD,
for example, may take 100 years to occur, and habitat may take centuries to recover (Murphy
and Koski 1989). Evaluation of the long-term effectiveness of riparian buffers relies heavily on
modeling and extrapolation of data into the future.

Technical Foundation 63

Although one of the major functions of riparian buffer zones is to provide 'LWD for the stream,
blowdown of trees in buffers sometimes results in a more abrupt loading of debris than intended.
Blowdown is more likely in areas of poorly drained soil, where buffers are perpendicular to
prevailing wind, and where the trees are conifers (C. Andrus, ODF, Salem, OR, unpublished
manuscript; DeWalle 1983; Steinblums et al. 1984). Blowdown is not highly correlated with
buffer width; however, wider buffers may still provide greater protection for the stream because
blowdown is often concentrated at the buffer edge. Blanket prescriptions for buffer width and
feathering edges may be ineffective in reducing blowdown (Boughton 1993). The risk of
blowdown, however, can be reduced by adjusting the buffer layout so that boundaries take
advantage of local windbreaks, such as mature forest, ridge lines, and rock outcrops (Gregory
and Ashkenas 1990; Boughton 1993).

Blowdown in,buffer zones is not considered a management failure nor a major problem for the
stream (Murphy et al. 1986; C. Andrus, ODF, Salem, OR, unpublished). Blowdown accelerates
LWD recruitment faster than in natural stands, but it is not an ecological disaster (Gregory et
al. 1990). In southeast Alaska, where wind is a major ecological factor, only 10-15% of trees
in buffers blow down (S. Paustian, FS, Sitka, AK, pers. comm. 1995). Upturning of roots can
contribute sediment, but this not usually a problem (C. Andrus, ODF, Salem, OR, unpublished
manuscript). Blowdown can also eliminate undercut banks, but this loss is offset by added cover
from LWD (Heifetz et al. 1986). In specific cases where blowdown creates a problem, such as
A barrier to fish migration, debris accumulations can be modified, but as little as possible to
-achieve desired results.

Physical exposure of the riparian community to increased light and.wind could cause the buffer
to deteriorate. When timber is harvested to the outer limit of the riparian zone, an edge is
created that affects the interior microclimate of the riparian forest (Fig. 6.3). Relative humidity
within the buffer declines, air temperature varies more, and windthrow and tree breakage
increase. Increased side light accelerates shrub development which reduces herbaceous cover
and tree regeneration (Hibbs et al. 1991). These factors may accelerate senescence of overstory
trees and succession to shrub-dominated communities. Thus, wider riparian buffer zones may
be needed to not only protect the stream but to ensure the long-term viability of riparian
functions (Cederholm 1994).

Natural disturbance regimes that operate over long cycles could be important in the long-term
effectiveness of buffer zones for maintaining habitat quality and diversity. The size of buffer
zones generally does not account for natural disturbances that involve larger landscape scales
(Everett et al. 1993). Attempts to maintain stable buffer zones against the natural tendency for
disturbances in dynamic forest ecosystems may be ineffective or even counterproductive because
stream productivity, unique habitats, or sensitive species often require disturbance events for
long-term sustainability (Everett et al. 1993).

Over the long term, habitat formation in streams may depend on infrequent catastrophic
disturbance events, such as major floods and landslides occurring after wildfire (Reeves.et al.,
in press). The most significant outcome of natural disturbances was the episodic delivery of
large quantities of mixed sediment and LWD into fish-bearing streams from hillslope failures
and debris torrents triggered along headwater stream channels- (Swanson et al. 1987; Hogan and

64 Technical Foundation

Schwab 1991). This material provided complex and productive fish habitat during subsequent
decades as the stream reworked and exported the material downstream. During later stages of
the disturbance cycle, fish habitat becomes less productive after most of this LWD and sediment
has been exported or decayed.

Disturbances from timber harvest differ from natural disturbances in frequency, severity, and
legacy of changes in the watershed (Reeves et al., in press). On the natural landscape, wildfires
and ma or hillslope failures were less frequent, covered less area, and left large amounts of
standing and downed wood in upslope areas. Modem timber harvest leaves much less large
wood, except in riparian areas along fish-bearing streams. Hillslope failures after timber harvest
deliver mostly sediment without the large quantities of LWD that accompanied natural
disturbances. Because of the reduced LWD, stream channels that develop in watersheds
managed for timber will be simpler than the complex channels that developed after natural
disturbances.

Thus, disturbance is not necessarily negative, but is needed to provide productive fish habitat
over the long tenn (Marcot et al. 1994; Everett et al. 1993; Reeves et al., in press). The
challenge is to develop management regimes that put timber harvest in the context of disturbance
regimes so that human patterns of land use do not substantially exacerbate natural disturbance
mechanisms and leave the necessary legacy for the development of required habitat conditions.
For the long-term development of productive fish habitat, the legacy of timber harvest needs to
include more large wood in upslope areas, particularly in buffer zones along headwater streams
and channels with the greatest potential for delivering this material to fish;-bearing streams
(Reeves et al., in press). In critical watershed areas, such as riparian zones and unstable soils,
natural disturbance regimes allowed to predominate can provide the necessary habitat-forming
amounts of sediment and LWD.

BEST MANAGEMENT PRACTICES

Best Management Practices are specific rules designed to prevent nonpoint-source pollution,
particularly from fine sediment (Lynch and Corbett 1990). They are measures used by agencies
to meet pollution control. needs under the Clean Water Act (MacDonald et al. 1991). Best
Management Practices can also refer to forest practices in general (Bisson et al. 1992). In this
synthesis, BMPs are used in the stricter sense related to the Clean Water Act.

Most BMPs address activities within buffer zones, harvest activities on hillslopes, road
construction and maintenance, and silvicultural practices (Boyette 1993). The Pacific Northwest
states generally have several categories of BMPs that pertain directly to streams: 1) directional
felling of trees and bucking and limbing of logs within streams and buffer zones; 2) yarding of
logs across streams and buffers by either cable or tractor methods; 3) treatment of soil and slash
deposited in stream channels; 4) prevention of erosion during tractor logging on hillslopes,
including construction and maintenance of skid trails; 5) mechanical site preparation in buffer
zones for replanting; and 7) road design, construction, maintenance, and obliteration. Usually,
BMPs are applied as a system of practices rather than a single practice.

Technical Foundation 65

BMP Programs

The Clean Water Act gives state water-quality agencies authority to certify their state forest
practices rules as approved BMPs for controlling pollution. Nationwide, 12 states had or were
developing a forest practices act as of 1992, and in 10 of these states, the act required
implementation of water quality BMPs (Boyette 1993). A state agency may certify BMPs of
federal agencies and delegate responsibility for streams under federal jurisdiction. Sixteen states
have a formal arrangement with the FS, and 5 states have an. arrangement with the BLM
(Boyette 1993).

The most frequent barriers to implementation of the forestry nonpoint-source management
program under the Clean Water Act are lack of adequate funding, staffing constraints, and lack
of technical personnel (Boyette 1993). As of 1992, 42 states had revised or developed BMPs
for forestry, and five states used federal cost-share programs to develop their BMPs (Boyette
,1,993). Complexity of the concept itself adds to the problem.

State approaches to controlling nonpoint-source pollution from forestry activities can be
regulatory or voluntary (Brown and Binkley 1994). States with regulatory programs impose
requirements on forest practices and assess penalties for noncompliance. They usually require
approval of harvest and road construction plans, inspection of projects in progress to improve
compliance, and final inspections to determine the need to assess penalties. States with
voluntary programs emphasize education, training, and on-site inspections if requested.
Nationwide in 1992, 23 states had voluntary programs, 13 had regulatory programs, 5 had a
combination of regulatory and voluntary measures, and 9 still lacked any formal program
(Brown and Binkley 1994). All the Pacific Northwest states and Alaska have a regulatory
program and monitor implementation and effectiveness to a limited degree.

BMP Monitoring

Monitoring is an important and required component of BMP programs. Monitoring for
implementation and effectiveness are most common, but a comprehensive BMP monitoring
program should include seven objectives (California Board of Forestry 1993):

1. Determine whether critical problem areas are recognized and appropriate practices are
specified;

2. Determine whether BMPs are adequately applied (implementation monitoring);

3. Determine whether BMPS are effective in meeting their intent (effectiveness monitoring);

4. Determine whether properly implemented BMPs meet water quality standards (compliance
monitoring);

5. Determine whether BMPs for given projects protect the stream's beneficial uses (project
monitoring);

66 Technical, Foundation

6. Provide results to the regulatory agency and public for review; and

7. Provide means for improving monitoring procedures and BMPs.

Nationwide, at least 40 states have some program for monitoring BMP implementation (Brown
and Binkley 1994). Eleven states have a monitoring program for BMP effectiveness, and most
of these states have published reports (Boyette 1993). As of 1992, 22 states had monitoring
programs for BMP compliance, and 15 states have published results of compliance monitoring
(Boyette 1993). Nine states used results of BMP effectiveness monitoring to modify BMPs
(Boyette 1993).

Implementation monitoring should be the first priority because BMP effectiveness can not be
evaluated unless the BMPs are actually implemented as specified. The most important aspect
of implementation monitoring is sample design [e.g., which timber harvest units or roads should
be included in the sample (Ferguson 1995)]. Ideally, all units would be measured, but
constraints on time, personnel, and funds usually necessitate visiting only a representative sample
of ongoing activities. A numerical rating system is also needed in determining whether or to
what extent practices have been implemented (Ferguson 1995).

Effectiveness of BMPs can be monitored individually or collectively (MacDonald et al. 1991).
Monitoring individual BMPs, such as the spacing of water bars on skid trails, is important in
controlling nonpoint-source pollution, but this is different from monitoring to determine whether
the BMPs protect water quality (Dissmeyer, in press). Individual BMPs are often best evaluated
at the site of the practice, such as a skid trail, which may be far away from the stream and
riparian zone. Legally, however, sediment generated from roads may not be a concern until it
enters a stream. Thus, there should be a clear linkage between the upslope measurements and
water quality (MacDonald and Smart 1994). In contrast, overall effectiveness of combined
BMPs for a project is usually evaluated by directly measuring water or other stream
characteristics. Such instream measurements may be difficult to relate to individual BMPs. The
states evaluate BMP effectiveness from a variety of perspectives. Parameters monitored most
often include turbidity, suspended solids, bioassessments, macroinvertebrates, and cobble
embeddedness (Boyette 1993).

Monitoring of BMP implementation and effectiveness is, conducted by numerous entities,
including federal and state agencies, tribes, and private landowners (e.g., TFW 1992; California
Board of Forestry 1993; Hoelscher et al. 1993; Simpson Timber Company 1994). Such a
decentralized approach allows monitoring to be tailored to individual land-use activities, but
disadvantages are that monitoring methods vary and data are not easily aggregated and compared
(Boyette 1993; Brown and Binkley 1994).

One example of a recent BMP monitoring project is Hoelscher et al.'s (1993) audit of Idaho's
BMPs. As with many projects of this type, objectives were to inspect the level of compliance
with forest practices rules and judge whether BMPs were effective in preventing sediment
pollution in streams. Over 1,000 activities were evaluated, and methods included assessment
of upland erosion, sediment pathways, and instream sedimentation. In this audit, BMPs were
implemented 92 % of the time. Where BMPs were not implemented, sediment pollution occurred

Technical Foundation 67

in 75 % of the cases, which emphasized the importance of enforcement. More than one-half of
the projects, however, were judged to have an adverse cumulative effect by causing sediment
pollution. Recommendations were made to the Idaho Land Board to modify the forest practices
rules to improve BMP effectiveness, and new rules were approved in 1995 (J. Colla, IDL,
Coeur d'Alene, ID, pers. comm. 1995).

'In general, recent assessments of BMPs indicate that f6rest practices can protect water quality
ifB.MPs are carefully developed and implemented (Brown and Binkley 1994). Most state BMPs,
however, do not carefully protect small perennial and intermittent streams from disturbance, and
BMPs for minimizing erosion on unstable slopes are still being developed.

The BMPs for protecting small perennial and intermittent streams must be effective because such
BMPs are the only practical means for protecting these headwater streams while other resource
activities continue. These BMPs prescribe whether trees can be felled into stream channels,
whether felled trees can be bucked and limbed there, and whether logging slash must be
removed from the channels after yarding. Certain BMPs also,determine whether logs can be
dragged or yarded across stream channels, and whether tractors and other logging equipment can
enter and cross them. Present state BMPs often allow trees to be felled, bucked, and limbed in
small perennial and intermittent streams, with few restrictions on yarding. These activities. can
destabilize small stream channels, causing the release of large amounts of sediment to
,downstream fish habitats (Toews and Moore 1982; Bilby 1984b).

Many'current problems with water quality also result from poor BMP implementation (Brown
and Binkley 1994). Compliance is generally lower for small private holdings than for public or
industrial lands (Brown and ginkley 1994). Having qualified field personnel available to provide
site-specific BMP recommendations is probably, the most efficient way to improve
implementation (Brown and Binkley 1994).

Ongoing changes in forest practices regulations should consider effects on profitability of timber
companies because decreased profitability could cause large land holdings to be subdivided or
converted to other land uses (R. Bettis, Pacific Lumber Company, Scotia, CA, pers. comm.
1994). This change would make implementing watershed management more difficult and lessen
habitat protection because the larger holdings are generally better at implementing BMPs and
often have experienced technical personnel. Watershed management is easier to implement on
watersheds with single owners than in watersheds with many small parcels with different owners.

CUMULATIVE'EFFECTS MANAGEMENT

The, analysis and management of cumulative effects are means of controlling non-point source
pollution that might be missed by planning at the project level (Coboum 1989). Forest managers
can reduce or prevent undesirable cumulative effects on fish habitat by effective planning at the
watershed level (Klock 1985). The National Environmental Policy Act (NEPA), Clean Water
Act, and laws of several states require that effects of past, present, and future management
activities be considered to prevent undesirable cumulative impacts.

68 Technical Foundation

Evaluations of potential cumulative effects of a given project should consider watershed erosion
potential; slope stability; current disturbances from roads, timber harvest, and other land uses;
rate of recovery after disturbance; and project area compared to the total watershed (Bach 1993).
More comprehensive methods for assessing cumulative watershed effects rely on interpreting
watershed condition and stream dynamics. Patterns of disturbance visible on aerial photographs
can be used to evaluate changes in channel conditions and to link such changes to upstream
causes (Grant 1988).

In evaluating cumulative effects, land managers need to account for land uses other than
forestry. Grazing, mining, agriculture, water development projects, recreation, usage, and
urbanization can all cause incremental cumulative impacts on a watershed and its fish populations
(Clark and Gibbons 1991; Nelson et al. 1991; Platts 1991). Forest management must be
considered in the context of these other activities and may need to be more conservative because
of their cumulative impacts. An example would be watersheds where livestock grazing has
damaged riparian vegetation and streambanks (Platts 1991), warranting greater protection of
riparian areas along streams during timber harvest to compensate for grazing impacts.

Watershed Analysis

Watershed analysis is the newest tool for assessing watershed features and providing the
information needed by planning bodies charged with managing cumulative effects. Watershed
analysis is a systematic procedure for describing current conditions and hazard areas in a
watershed to provide a basis for prescribing appropriate precautionary measures (FEMAT 1993).
It assumes that managers can protect the overall condition of the watershed by managing
sensitive areas appropriately and applying standard practices in less-sensitive areas (Washington
Forest Practices Board 1993). Watershed analysis, in itself, does not result in decisions
regarding management of cumulative effects, although the overall watershed analysis process can
include both resource assessment and decision-making modules (e.g., Washington Forest
Practices Board 1993). Usually, however, watershed analysis is merely a management tool that
provides part of the basis for management decisions. Watershed analysis is currently being
developed, and several approaches are being tried on both public and private lands (Grant et al.
1994).

With information from watershed analysis, managers should be b etter able to avoid cumulative
impacts. Because watershed analysis is a new methodology, its efficacy in promoting better
habitat protection has not been widely tested. One evaluation on the Tongass National Forest,
Alaska, found that several impacts resulting from logging activities could have been avoided if
watershed analysis had been conducted (USDA 1995).

A comprehensive cumulative effects assessment process should be 1) systematic, the same steps
used to assess each watershed; 2) structured, each step supported by a written guide with
decision criteria; 3) reproducible, different observers obtaining the same conclusions;
4) defensible, supported by established scientific principles of watershed management; and 5)
adaptive, periodically revised to reflect new science and technology (IDL 1994).

Technical Foundation 69

Besides watershed analysis, other approaches to analyzing cumulative effects rely on stream
surveys to identify possible problems in a watershed, but these' methods are less able to identify
management-sensitive landforms. Problems with roads and erosion-hazards would be more
readily identified by watershed analysis than by stream surveys.

The concept of Equivalent Clearcut Area (ECA) provides a method for setting threshold levels
of concern for cumulative effects (Belt 1980; King 1989). This method monitors areas in young
seral stages resulting from clearcutting, fire, or other disturbances, and establishes ECA levels
at which cumulative effects begin to show. Exceeding these levels indicates a need for caution
before additional activities. For example, an ECA level of 15 % of forest stands < 30 years old
confers a low risk of impacts, whereas risks increase above the 15% ECA level (McCammon
1993).

The ECA approach has potential drawbacks that can limit its application (Rhodes and
McCullough, in press). It does not represent factors that modify impacts from disturbance, such
as proximity to streams and soil hazard. Timber harvest in riparian areas, for example, has
greater impacts than in uplands. The recovery time (15-30 years) for hydrologic processes does
not reflect recovery time for other habitat functions (200 years to recover LWD; Minshall et al.
1989). The ECA approach also omits other causes of degradation, particularly grazing.

Despite its drawbacks, the ECA approach is useful in establishing thresholds of concern for the
level of timber harvest in watersheds where harvest in riparian areas is closely controlled. In
determining ECA level, coefficients based on amount of tree crown cover can be used to relate
harvest prescriptions other than clearcutting (e.g., shelterwood cuts) to equivalent clearcut area
(L. Bailey, Payette National Forest, pers. comm. 1994; N. Gerhardt, Nez Perce-National Forest,
pers. comm. 1994). Such conversion coefficients may need to be adjusted to fit local watershed
conditions.

Habitat Conservation Plans

Management cooperation across property boundaries is essential for effective watershed
management (Trout Unlimited 1994). Habitat protection and restoration on a watershed basis
will require integrating federal land management with other regulatory programs that affect
aquatic habitats, particularly through the Clean Water Act and Endangered Species Act (Williams
1993). Habitat Conservation Plans developed under the ESA have an important role in
watershed planning on private lands.

Under Section 10 of the ESA, non-federal landowners can voluntarily develop Habitat
Conservation Plans (HCPs) in consultation with NMFS and the U.S. Fish and Wildlife Service
(FWS). The HCPs outline long-term (> 50 years) plans for land management and show how
critical habitat of a listed species will be managed to mitigate or prevent its "taking" through
adverse modifications to habitat falling under the "harass and harm" definitions of the ESA.
The plans can include various measures, such as extended rotation cycles for timber harvest in
critical habitat, watershed analysis to prevent cumulative effects, habitat restoration, and
monitoring. Landowners can also address biotic communities rather than individual species

70 Technical Foundation

through "multi-species" HCPs that cover listed or non-listed species other than fish and foster
conservation of biodiversity.

Once HCPs are approved by NMFS/FWS, the landowners are assured that management of their
lands will not be disrupted by new regulations or restrictions for those species. In issuing an
incidental take permit, NMFS/FWS accept the applicant's proposed package of conservation
measures and agree that the number of fish that would be taken during an otherwise lawful
activity (e.g., land management) would not jeopardize the species' existence.

General provisions considered in an HCP include 1) 'strearnside buffers that provide the full
range of riparian functions (LWD, shade, nutrients, sediment filtering, and bank stability); 2)
planning for road systems to minimize road density; 3) road maintenance to reduce sediment
delivery; 4) avoidance of erosion-prone areas; 5) restoration projects for riparian or fish habitats
that are integrated with comprehensive watershed management; 6) consideration of natural
disturbance regimes; 7) removal of artificial barriers to restore fish passage to natural habitats;
8) measures to provide optimum quantity and quality of water for fish resources; 9) monitoring
programs; and 10) planning for adaptive management (R. Baker, NMFS, Portland, OR, pers.
comm. 1995).

Because of the wide-ranging migrations of anadromous salmonids, planning at the basin,
regional, and even larger scales is also necessary for managing cumulative effects from activities
in addition to timber management. The NMFS Proposed Recovery Plan for Snake River Salmon
(USDC 1995) is a good example of the comprehensive planning needed to address all factors that
can cumulatively affect salmon stocks. In this comprehensive plan, timber management and
other watershed uses are just one of five major planning areas that also include main-stem river
and estuarine habitat, fisheries harvest management, hatchery propagation, and changes in
institutional structure to improve decision making. These other four components are beyond the
scope of this synthesis, but forestry-fisheries issues should properly be considered in this larger
context.

Comprehensive watershed management must involve more than improved scientific
understanding; it also must encompass economic, social, and political concerns (Shepard 1994).
Watershed management needs to involve all stakeholders, including landowners, industries,
environmental groups, and local citizens, in formulating and implementing watershed
management. The challenge is to develop a broad base of support and participation representing
a wide spectrum of interests (Brouha 1991; Daniels et al. 1994).

Working groups consisting of government agencies, industry, and citizen groups can be
instrumental in obtaining consensus on forest practices issues (AWGCFFR 1991) and watershed
management for entire river basins (Doppelt et al. 1993). Working groups provide forums for
complete discussion of alternative viewpoints and their supporting rationale as they strive for
consensus on scientific, technical, and management recommendations. The NMFS Proposed
Recovery Plan for Snake River Salmon (USDC 1995) recognizes that such working groups must
be integrated into decision making processes.

Technical Foundation 71

Economic incentives can be provided for local communities and landowners to support and
participate in habitat protection and restoration. On public lands, contracts awarded by
competitive bidding can provide effective habitat protection and restoration while offsetting some
employment losses in timber-dependent communities (Lippke and Oliver 1994). Using
restoration funds to create local jobs encourages support for habitat restoration (Pacific Rivers
Council 1993b; Weigand 1994). Tax credits and cost-sharing programs can be expanded to
compensate private landowners for measures taken to protect public aquatic resources (Henly
and Ellefson 1987). On private lands, the desire to conserve public fish and wildlife competes
with the landowner's rights of private property. Thus, economic impacts on landowners from
new regulations requiring expanded buffer zones or retention of additional leave trees along
streams should be offset through public funds. Compensation would help to ease resistance to
new regulations.

Chapter 7
Current. Forest Practices

Forest practices are constantly being revised in light of new scientific information, political
compromises, and changing public awareness and demands for forest and water resources. On
federal timberlands, forest practices are generally consistent across the Pacific Northwest, with
management direction from the Northwest Forest Plan (NFP) and PACFISH aquatic conservation
strategies (USDA and USDI 1994a, 1994b). Management direction for federal timberlands in
Alaska comes from the Tongass Land Management Plan as amended to include the 1990 Tongass
Timber Reform Act. Forest practices on private and state lands are governed by their respective
laws, and forest practices vary from state to state.

Regulations governing forest practices are developed through a public process with comments
sought from the public on proposed regulations (Brouha 1991). Federal rules are developed and
implemented with public review according to NEPA. State rules are written by the responsible
resource agency, usually the forestry agency with input from the water quality agency and fish
and wildlife agency, to implement intent of forest practices acts passed by state legislatures and
rulings by state boards of forestry.

'this chapter describes current forest, practices on both federal and private lands in the Pacific
Northwest and Alaska. These areas have recently revised their forest practices toward ecosystem
management aimed at protecting and restoring habitat for anadromous salmonids.

FEDERAL LANDS

Northwest Forest Plan and PACFISH

Considering that aquatic and riparian habitats on federal lands are critical for anadromous
salmonids and other species, the President's Forest Ecosystem Management Assessment Team
(FEMAT) proposed an Aquatic Conservation Strategy to maintain and restore aquatic ecosystems
in the range of the northern spotted owl (FEMAT 1993). The FEMAT (1993) report provided
the scientific basis for adopting the strategy in the Northwest Forest Plan. A similar strategy
has since been adopted to include areas not covered by the NFP over the entire range of Pacific
anadromous salmonids from California to Alaska, and has been termed PACFISH (for Pacific
Anadromous Fish Habitat Management Strategy).

74 Current Forest Practices

Both the NFP and PACFISH strategies are new and not yet wholly implemented, and application
of PACFISH in Alaska is on hold pending further study (USDA 1995). The NFP has been
adopted through the NEPA process and a Record of Decision (USDA and USDI 1994a). The
PACFISH strategy is interim until the FS and BLM complete formal NEPA Environmental
Impact Statements (EISs) for the non-NFP areas. In 1994, the FS and BLM initiated EISs to
develop and adopt coordinated ecosystem management strategies for the interior Columbia River
Basin. This effort is supported by a biological, social, and economic assessment known as the
Eastside Ecosystem Management Project (USDA and USDI'1994c). The PACFISH interim
direction for the region will be replaced by new management direction when the EISs are
completed and the selected alternatives result in revision of national forest and BLM management
plans.

The NFP and PACFISH strategies have four principal components:

1 . Riparian Reserves (in NFP) and Riparian Habitat Conservation Areas (in PACFISH):
Unds along streams and unstable areas where special rules govern land use;

2. Key watersheds: A system of priority watersheds critical to at-risk fish stocks;

3. Watershed Analysis: An evaluation of watershed processes, functions, conditions, and
capabilities to enable planning and informed decision making; and

4. Restoration: Programs to restore watershed conditions, riparian functions, and fish
habitats.

The NFP and PACFISH strategies, however, are not identical. In the region covered by NFP,
riparian reserve boundaries can be modified and management activities can continue in key
watersheds only after watershed analysis, and no new roads can be built in roadless areas of key
watersheds. For PACFISH, watershed analysis is required before authorization to build roads
in or across RHCAs can be granted, but it is not required before other management activities.
The PACFISH strategy also does not address the construction of new roads in roadless areas.
Key watersheds have been identified for NFP areas, and in the Snake River Basin, all
watersheds are designated as key watersheds. For other areas outside the Snake River Basin,
but within the PACFISH range, key watersheds have not yet been identified (R. Baker, NMFS,
Portland, OR, pers. comm. 1995).

Riparian Reserves and Riparian Habitat Conservation Areas (RHCAs) are similar to riparian
buffer zones but are more comprehensive. They generally follow the stream network but also
include unstable hillslopes and other areas necessary to maintain stream ecosystem processes.
They are also designed to maintain wildlife habitat in addition to fish habitat.

The NFP and PACFISH strategies prescribe buffer widths for three categories of streams:

1. Fish-bearing streams: Riparian reserves or RHCAs include either the stream's inner
gorge, the 100-year floodplain, the extent of riparian vegetation, the area within two site-

Current Forest Practices 75

potential tree heights (the average maximum height given site conditions), or within 300
ft (91 m), whichever is greatest.

2. Perennial non-fish streams: Riparian reserves or RHCAs include either the stream's
inner gorge, the 100-year floodplain, the extent of riparian vegetation, the area within
one site-potential tree height, or within 150 ft (46 m), whichever is greatest.

3. Intermitte nt streams, wetlands less than 1 acre, and unstable areas: Riparian reserves
under NFP include the stream's inner gorge, unstable,areas and riparian vegetation, the
area within one site-potential tree heights, or within 100 ft (30 in), whichever is greatest.
Width of RHCAs under PACFISH in key watersheds is similar to NFP's riparian
reserves, but may be narrower (50 ft or 15 in) in non-key watersheds.

Key watersheds are a system of watersheds where habitat for anadromous fish and other at-risk
fish species, particularly bull trout (Salvelinus confluentus), receive special attention and
treatment. Key watersheds can include those with ESA-listed stocks, excellent habitat for mixed
salmonid assemblages, and degraded watersheds with good restoration potential. Watersheds
in good condition serve as "anchors" for the potential recovery of depressed stocks and provide
colonists for adjacent degraded areas. Those in poor condition can provide good habitat after
restoration. Key watersheds should not be in isolated blocks, but should be linked by connective
corridors (Payne and Bryant 1994) to allow recolonization and expansion of populations during
recovery.

Watershed analysis provides information for use in planning and other decision-making
processes. It describes conditions and ecosystem processes in a watershed so that project
planning can focus on site-specific environmental issues (Reid et al. 1994). Watershed analysis,
however, is not a decision-making process, and it does not develop alternatives or limit
management options.

Principal aquatic objectives of watershed analysis are to 1) determine ecosystem processes
affecting the flow of water, sediment, and organic matter through a watershed; 2) identify areas
that are sensitive or critical to beneficial uses; and 3) determine the distribution, abundance,
.habitat requirements, and limiting factors of critical Species. The size of watersheds analyzed
is from 20 to 200 square miles (50-500 kin), which is smaller than regional and basin
assessments, but larger than project analyses (Reid and Ziemer 1994b).

Information from watershed analysis provides a basis for general land use planning (Benda
1993), including transportation planning, cumulative effects assessments, and monitoring.
Watershed analysis also identifies beneficial uses, environmental issues, and societal concerns,
and it may develop recommendations for management options based on physical and biological
conditions. It is an iterative process. As activities are conducted, as new data become available,
and as habitat recovers, analyses can be kept current for the next planning cycle.

76 Current Forest, Practices

Tongass Land Management Plan

The Tongass Land Management Plan (TLMP) was adopted in 1979 following the 1976 National
Forest Management Act to provide guidelines for managing natural resources on the Torigass
National Forest of southeast Alaska (USDA 1989). As with forest plans for other national
forests, TLMP addresses fish and wildlife, recreation, timber, soil and water, and other
multiple-use values. The TLMP is currently undergoing revision and NEPA review, which
began in 1989. In 1990, congress passed the Tongass Timber Reform Act (TTRA), reforming
forest practices on federal lands in the Tongass National Forest. The Act amended the Alaska
National Interest Lands Conservation Act "to protect certain lands in the Tongass National
Forest in perpetuity, to modify certain long-term contracts, to provide for protection of riparian
habitat, and for other purposes. " Management under TLMP was changed to reflect requirements
in TTRA.

The TTRA requires buffer zones at least 100 ft (30 in) wide on each side of all Class I streams
(i.e., those with anadromous fish) and on those Class 11 streams (i.e., those with resident fish
only) that flow directly into a Class I stream. Timber harvest is prohibited within these buffer
zones. No buffer zones are required along Class III streams (i.e., those without fish), but BMPs
must be followed to protect water quality and prevent downstream sedimentation and excessive
increases in temperature. The TTRA does not specifically classify or address intermittent
channels.

The NFP concepts of riparian reserves, watershed analysis, and key watersheds were developed
after TTRA was passed. Hence, they were not considered under TTRA. The TLMP, however,
does have aquatic habitat management units analogous to riparian reserves and many other
similarities with NFP concepts. The PACFISH strategy was intended to include Alaska to bring
habitat protection on the Tongass National Forest up to the same level employed on other federal
lands with anadromous salmonids. Because of the generally healthy status of anadromous
salmonids in Alaska and other reasons, application of PACFISH to Alaska was postponed until
further study of the need for the additional protection (USDA 1995).

STATE AND PRIVATE LANDS

Alaska

The Alaska Legislature amended the Alaska Forest Resources and Practices Act in 1990 to
reform forest practices on state and private lands (ADNR 1993). On state lands, timber harvest
is prohibited within 100 ft (30 m) of anadromous fish streams, and timber harvest must be'
consistent with the maintenance of fish and wildlife habitat between 100 and 300 ft (30-100 m).
On private lands, streams have a lower level of protection, and riparian buffer zones depend on
stream type and size.

Current Forest Practices 77

Streams on private lands. in coastal forests of southern Alaska are classified into three types:

Type A. Streams with anadromous fish; unconstrained channels; banks held in place
by vegetation.

Type B. Streams also with anadromous fish; channels constrained by bedrock not
vegetation.

Type C. Streams without anadromous fish; perennial streams or intermittent channels
incised more than 28 degrees.

Other streams and less-incised intermittent channels are not specifically classified.

Standards for management within
buffer zones on private lands differ
according to stream type. Along ------
Type A streams, no timber can be
harvested within 66 ft (22 in) of the
streambank (Fig. 7.1). Along Type
B streams, all trees may be
harvested, but timber harvest within
100 ft (30 in) of the stream or to the
slope break (whichever is smaller)
must comply with BMPs. Along
Type C streams, timber harvest
within 50 ft (15 in) of the stream or
to the slope break must comply@with
BMPs.

The BMPs used within buffer zones
are designed to protect stream
channels from disturbance and
prevent sediment and small debris
from entering streams during felling
and yarding. Trees must be felled
away from streams in "V-notches,"
and operators must achieve at least
partial suspension of logs while
yarding within buffer zones. Trees
felled into fish-bearing streams must
WA-

be removed immediately, and trees
felled into non-fish streams must be
removed as -soon as feasible to Figure 7. 1. A 66-ft no-harvest buffer on both
prevent destabilizing the channel and sides of a Type A anadromous fish stream on
downstream impacts. Alaska's private land in southeast Alaska. (Photo by
BMPs also cover other forestry R. Harris, Sealaska Corporation.)

78 Current Forest Practices

activities including slash disposal, harvest unit layout, rehabilitation of mass wasting, road
construction and maintenance, and reforestation.

The Alaska rules also grant "variations" from requirements in some cases, such as selective
harvest of specific trees within the buffer zone. A landowner may propose a variation to the
State Forester, and the Department of Fish and Game has due deference in such requests
concerning fish and wildlife habitat. The variation is approved if the State Forester determines
that the activity is not likely to cause significant harm to fish habitat because of site-specific
circumstances. An automatic variation is granted for small streams [<5 ft (1.5 m) wide],
allowing harvest of 25 % of the trees between 25 ft (7.6 in) and 66 ft (20 m) from the stream.

The only opportunity for evaluating possible cumulative effects is during review of a "detailed
plan of operations" which timber operators must file before beginning work. The plan showing
locations of water bodies, stream crossings, road layout, unstable slopes, and other information
can be reviewed by affected agencies, coastal districts, and the public. The review results in
either allowing forest practices to proceed with standard BMPs or designing new management
prescriptions to prevent impacts.

California

The California Department of Forestry and Fire Protection (CDF) enforces the state's forest
practices rules which are promulgated through the Board of Forestry. Rules relating to
protection of water quality and other resources were revised in 1994 (CDF 1994).

California's waters are grouped into four classes:

Class 1. Fish always or seasonally present or a supply of domestic water.

Class II. Fish always or seasonally present within 1,000 ft (304 in) downstream or
habitat for non-fish aquatic species.

Class III. No aquatic life present; channel has definite bed and banks and is capable of
transporting sediment to Class I or 11 streams.

Class IV. Artificial watercourses with established beneficial uses.

The width of riparian buffer zones (called Watercourse and Lake Protection Zones; WLPZs)
depends on stream class, side slope, and yarding method (Table 7. 1). For Class I (fish-bearing)
streams, the WLPZ width ranges from 75 ft .(23 in) where side slopes are less than 30 % to 150
ft (46 in) where side slopes exceed 50%. For Class II (non-fish) streams, the WLPZ width
ranges from 50 ft (15 in) where side slopes are less than 30% to 100 ft (30 in) where side slopes
exceed 50%. The WLPZ along Class I and 11 streams in areas of steep side slopes can be
reduced if cable yarding is used instead of tractor yarding. The need for and width of WLPZs
along Class III (no aquatic life) and IV (artificial) watercourses are determined by on-site
inspection.

Current Forest Practices 79

Table 7.1. California's requirements for width of Watercourse and Lake Protection
Zones (WLPZs) by slope class and stream class.

Class I Class 11 Class III
Slope (f ish-bearing) (non-fish) (no aquatic life)

< 30% 75 50 Site specif ic'

30-50% 100 75 Site specific'

> 50% 1501 1003 Site specific'

'The need for and width of WLPZ is determined by on-site inspection.
'Subtract 50 ft for cable yarding.
'Subtract 25 ft for cable yarding.

Within the WLPZ, harvest prescriptions depend on stream class and side slope. For Class I
streams, at least 50% of the overstory and 50% of the understory canopy must be left
representative of the preharvest stand, and the residual overstory canopy must be composed of
at least 25% of the preharvest conifers. For Class 11 streams, at least 50% of the total canopy
must be left representative of the preharvest stand and composed of at least 25% of the
preharvest overstory conifers. Where less than 50% canopy exists along Class I and II streams
before harvest, only salvage that protects riparian functions is allowed. To provide LWD, at
least two living conifers [ @!! 16 inch ( @!: 41 cm) dbh and 50 ft (15 m) tall] must be retained per
acre within 50 ft of all Class I and II streams. When WLPZs are required along Class III
channels, at least 50% of the preharvest understory vegetation must be left living and well
distributed.

Activities within the WLPZ and adjacent slopes are regulated by BMPs designed to protect
channel stability and prevent sediment and small debris from entering the stream channel. These
BMPs direct the operator to fell trees away from streams and to avoid damaging residual
vegetation. Yarding operations are also regulated by BMPs that prohibit use of tractors on
unstable soils, during winter Wet periods, on slopes greater than 65 %, and on slopes greater than
50% that do not flatten to Class I or 11 streams. A prepared structure (e.g., temporary log
culvert) must be used when crossing any watercourse that may carry water during the life of the
crossing structure. When necessary to protect beneficial uses, a WLPZ or equipment limitation
zone may be required for Class III watercourses. Regulations require that, when needed, the
width and protection of WLPZs on Class III and IV watercourses prevent degradation of
downstream beneficial uses.

Soil and debris that accidentally enter Class I, II, and IV watercourses must be removed
immediately to prevent destabilizing the channel. Soil and debris deposited in Class III
watercourses must be removed before ending operations or before October 15, and constructed

80 Current Forest Practices

temporary stream crossings there must be removed before winter. Continuous areas of disturbed
soil > 800 square ft (> 74 square m) along Class I and II streams must be treated to reduce soil
loss. Within the WLPZ, at least 75 % of the ground cover must remain undisturbed for sediment
filtering. Where necessary, WLPZs are seeded and mulched to maintain or improve ground
cover for sediment filtering. Other BMPs regulate road construction and maintenance, landings,
and silvicultural operations.

The required review process begins with filing of timber harvest plans prepared by registered
professional foresters and submitted to CDF for approval. As part of a field examination, the
forester evaluates riparian areas for erodible streambanks, debris jam potential, overflow
channels, flood-prone areas, and other sensitive areas. The forester proposes WLPZ widths and
protection measures. The CDF conducts on-site inspections as part of the plan review.

Cumulative impacts must be assessed in the timber harvest plan, based on the Board of
Forestry's Cumulative Impacts Assessment Process. The forester preparing the plan must
consult sources that are reasonably available, but the forester's duties are limited to closely
related past, present, and probable future projects. State agencies can supplement this
information. Factors considered in the Cumulative Impacts Assessment include sediment, water
temperature, organic debris, peak flows, and watercourse condition (gravel embeddedness, pool
filling, channel aggradation, bank cutting). The CDF makes the final determination regarding
sufficiency of the assessment and presence of cumulative impacts.

California's rules also include provisions for a form of watershed planning in which the
landowner may submit an optional "Sustained Yield Plan." This plan is intended to provide a
means for addressing long-term issues of sustained timber production and cumulative effects on
fish and wildlife on a landscape basis. This plan analyzes potential cumulative impacts on water
quality and fish habitat, includes maps of unstable soils and planned roads, and discusses feasible
measures to avoid impacts. The rules encourage landowners in watersheds with multiple owners
to cooperate in these watershed assessments.

Certain watersheds that are particularly sensitive to impacts from further timber harvest can be
classified as "sensitive" after public hearings and given additional protection. Justification for
sensitive status can include ongoing impacts on fish habitat from erosion problems related to past
or ongoing land-use activities and potential impacts from accelerated proposed road construction
and timber harvest. For all such watersheds, the Board of Forestry identifies specific mitigation
measures that will protect sensitive resources.

Idaho

The Idaho Legislature enacted a forest practices act in 1974 and amended it seven times since
1980 UDL 1992). Forest practices rules are developed by the Idaho Land Board, enacted by
the legislature, and enforced by the Idaho Department of Lands (IDL). The Idaho Land Board
has recently approved new changes in the forest practices rules which will likely become
effective in 1996 (J. Colla, IDL, Coeur d'Alene, ID, pers. comm. 1995). The rules
incorporating these changes are described here.

Current Forest Practices 81

Idaho recognizes only two stream classes:

Class I. Streams important for fish or domestic water supply.

Class 11. Minor drainages with perceptible streambed and banks, used by few if any
fish; principal value is influence on water quality or quantity downstream..

Buffer zones (called Stream Protection Zones; SPZs) depend on stream class. For Class I
streams, the minimum SPZ is 75 ft (23 m) on both sides of the stream. For Class II streams,
the minimum.SPZ is 30 ft (9 m) on both sides of the stream if it contributes surface flow into
a Class I stream; other Class II streams have a 5-ft (1.5-m) SPZ. The number of leave trees
required differs by stream class and stream width (Table 7.2). More leave trees are required
for large streams [ > 20 ft (6 m) wide] than for smaller streams. In addition, 75 % of the existing
shade canopy over Class I streams must be left intact.

Idaho's BMPs establish enforceable standards for all forestry activities, including felling,
yarding, slash disposal, road construction and maintenance, and silvicultural treatments (Almas
et al. 1993). Trees must be felled, bucked, and limbed away from Class I streams (but not
Class 11 streams) wherever possible. Slash deposited in Class I streams must be removed during
operations; slash in Class II streams must be removed after yarding if accumulations could block
the stream or be transported downstream. Removing felled timber from the SPZ must be done
carefully to avoid damaging shade and sediment filtering functions. Cable yarding across or

Table 7.2. Idaho's requirements for width of Stream Protection Zones (SPZs) and
leave trees within SPZs per 1,000 ft (304 m) along each side of streams by diameter
breast height (dbh).

Class I stream width
< 10 ft 10-20 ft > 20 ft Class 11

SPZ width (ft) 75 75 75 301

Tree diameter
(inches dbh)
3-7.9 200 200 200 140'

8-11 42 42 42 0

12-19.9 0 21 21 0

20 0 0 4 0

ISPZ width is 5 ft and no standing trees are required for Class II streams that do not contribuie-
surface flow into Class I streams.

82 Current Forest Practices

within SPZs must minimize disturbance to vegetation and the stream channel. Ug skidding
across streams is prohibited, and tractors and other ground-based equipment are prohibited
within SPZs or in streams, except at constructed temporary crossings. Log skidding next to SPZs
on slopes > 45 % needs an approved variance. Operators also must avoid conducting operations
along non-classified waters (i.e., without definite bed and bank), such as "wet draws" where the
presence of water is indicated.

Cumulative effects are managed with an approach analogous to federal watershed analysis (IDL
1994). Its purpose is to give trained evaluators an understanding of current watershed condition,
hydrologic processes, and disturbance history. The process assesses streambed sediment,
channel stability, sediment delivery, water temperature, shade, nutrients, and hydrology. It
provides a key to determine whether cumulative effects exist, and guidancefor landowners to
design practices to correct adverse conditions and prevent future cumulative effects.

The cumulative effects assessments are conducted by a committee of forest landowners in the
watershed. This committee selects certified evaluators who prepare an assessment report
identifying problem conditions and guidelines for forest practices. The committee develops
management prescriptions based on the assessment report. The IDL reviews the assessment and
prescriptions for consistency, completeness, and compliance with the Forest Practices Act. The
process results in either allowing forest practices to proceed with standard BMPs, or designing
new management prescriptions to prevent problems.

Monitoring is conducted by the IDL to evaluate implementation and effectiveness. Audits of
forest practices determine compliance with approved prescriptions, and effectiveness monitoring
is conducted through standard assessment techniques by the landowner committee every 5 years
and filed with the IDL. More detailed monitoring is done in some cases by other state agencies.

Oregon

Oregon recently revised its forest practices rules, taking an innovative approach to provide for
long-term productivity of stream and riparian habitats (ODF 1994). Innovative aspects include
incentives for timber operators to actively manage riparian stands to develop desired future
conditions characteristic of mature streamside forests.

Oregon classifies streams according to beneficial use and stream size into three types:

Type E Fish-bearing streams.

Type D. Domestic water supply.

Type N. Non-fish streams, including intermittent streams with well-defined
channels.

Each type is subdivided according to mean annual streamflow (in cubic ft per second [cfs]) into
three size categories: large [ @t 10 cfs (:@: 0.28 m'/s)]; medium [2-10 cfs (0.06-0.028 m/s)]; and
small [<2 cfs (:!90.06 m@ls)]. Timber operators are required to submit written plans for

Current Forest Practices 83

approval by the Department of Forestry (ODF) if operations are within 100 ft (30 m) of a 1@pe
F or D stream.

Riparian buffer zones (called Riparian Management Areas; RMAs) depend on stream class and
size (Table 7.3). Width of RMAs ranges from 50 to 100 ft (15-30 m) for Type F streams; 20
to 70 ft (6-21 m) for I)rpe D streams, and 0 to 70 ft (0-21 m) for Type N streams. Small
perennial Type N streams have RMAs in some regions (eastern Cascades, Blue Mountains) but
not in others (Coast Range and Western Cascades). Intermittent Type N streams do not have
RMAs. Non-classified areas without defined channels, such as ephemeral overland flow and
seeps, also do not have RMAs, but operators must protect soil and vegetation from disturbances
that could affect beneficial uses, and operators are encouraged to leave green trees and snags in
these areas.

The rules are designed to move RMAs toward desired future conditions. For 1"ype F streams,
the desired future condition for RMAs is to grow and retain riparian vegetation so that, over
time, average conditions across the landscape become similar to those of mature strearnside
stands. Requirements for leave trees are based on objectives for tree basal area that would
emulate an average mature streamside stand 120 years old halfway through a 50-year timber
rotation. For Type,N streams, the desired future condition is to have sufficient strearnside
vegetation to support functions that are important to downstream fish use and supplement wildlife
habitat.

Management goals distinguish between areas with different native forests. Where the native
forest would be conifers, the rules aim to retain a sufficient number of conifers to attain a
mature conifer stand along large and medium streams by halfway through the next timber

Table 7.3. Oregon's requirements for width (in ft) of riparian management areas by
stream size.

Stream Type F Type D Type N
size' (fish-bearing) (domestic water) (non-f ish)

Large 100 70 70
@!t 10 cfs

Medium 70 50 50
2-10 cfs

Small 50 20 0_102
5 2 cfs

'Size based on mean annual strearnflow in cubic ft per second (cfs).
210 ft in some regions if stream is perennial; 0 ft in other regions and if stream is
intermittent.

84 Current Forest Practices

rotation. Where the native community is hardwood dominated,, site-specific prescriptions are
used so that regrowth replaces older trees. Where the riparian forest was historically conifers
but currently dominated by hardwoods, the desired action is to manipulate the RMA during
timber harvest to create conditions for reestablishing conifers..

Vegetation retention for Type F streams includes all trees within 20 ft (6 m) of the stream, all
downed wood and snags not safety or fire hazards, and at least 40 live conifers [ > 11 inches
( > 28 cm) dbh] per 1, 000 ft (304 m) along large streams and 30 live conifers ( > 8 in dbh) along
medium streams (Table 7.4). Requirements for Type D and large and medium Type N streams
are similar to Type F streams, except that the number of live conifers to be retained is lower
(Table 7.4). For small perennial 'I'ype N streams, only understory and unmerchantable trees
within 10 ft (3 m) of the stream are required in some regions of the state. No retention is
required along small perennial Type N streams in western Oregon or, any small intermittent Type
N stream, but operations are subject to BMPs.

Enough conifers must also be left to meet standard basal area targets (Table 7.5). These targets
are designed to produce a mature conifer stand halfway through a timber rotation. They. are
calculated based on normal yield of a Douglas-fir forest at 120 years of age after adjusting for
incomplete stocking of riparian areas, tree mortality, and tree growth during the rotation. The
target is also reduced because of the 20-ft no-harvest zone next to the stream, which is not
managed. The basal area target of 230 square ft for large Type F streams equals about 350 11 -
inch-diameter trees. If the preharvest basal area is less than standard targets, no harvest is
allowed and the operator must retain additional conifers or other trees.

An innovative aspect of Oregon's rules is that they provide incentives for private landowners to
take advantage of restoration opportunities during timber harvest. Timber owners can earn a
basal area credit for trees that they place into Type F streams during operations. The basal area

Table 7.4. Oregon's minimum requirements for conifer leave trees [in number per
1,000 ft (304 m) along each side of the stream] in riparian management areas by
stream size.

Stream Type F Type D Type N
size' (f ish-bearing) (domestic water) (non-fish)

Large' 40 30 30

Medium 3 30 10 10

Small 0 0 0

'Size based on mean annual strearnflow in cubic ft per secorid (cfs; see Table 7.3).
'Retained conifers must be at least 11 inches (28 cm) diameter breast height (dbh).
3
Retained conifers must be at least 8 inches (20 cm) dbh.

Current Forest Practices 85

Table 7.5. Oregon's standard targets for conifer basal area in riparian management
areas each side of Type F streams by stream size class in selected regions.
Standard basal area target (square ft per 1,000 ft)

Normal basal Large Medium Small
Region area yield' (RMA = 100 ft) (RMA =@ 70 ft) (RMA = 50 ft)_

Coast Range 457 230 120 40

Western Cascades 473 270 140 40

Siskiyou 411 220 110 40

'Theoretical normal basal area (square ft per 1,000 ft of stream) in a 1 00-ft RMA witF
120-year-old Douglas-fir forest after adjusting for incomplete stocking and tree
mortality (T. Lorensen, ODF, Salem, OR, pers. comm. 1994).

credit is twice the basal area of each conifer log or tree placed in a large or medium Type F
stream, or equal to the basal area of logs and trees placed in small Type F streams. These
credits can be used to reduce the basal area requirement of live conifers in RMAs. The conifer
basal area, however, must not be reduced below "active management" targets, which range from
about 50 to 80% of the standard targets. Specific guidelines and requirements for placing logs
in streams for basal area credit are given by ODF and ODFW (1995).

Activities near streams or within RMAs are regulated by BMPs to protect water quality.
Operators must fell trees away from streams (including small intermittent Type N streams),
except for approved restoration projects. Logging slash must be removed from Type F and D
streams during harvest operations and must not accumulate in Type N streams in amounts that
threaten water quality. Except for small Type N streams, cable yarding across streams must
have prior ODF approval and must achieve full suspension. Cable yarding across small Type
N streams must minimize disturbance to the stream. Tractors and other ground-based yarders
may not enter flowing streams except at constructed temporary crossings. Such crossing
structures are not required for crossing dry streambeds if the disturbance is no greater than
would be caused by construction of the crossing. Mechanical site preparation for tree planting
is not allowed in RMAs on slopes > 35 %, except for certain equipment during dry periods, and
not where soil compaction and erosion are likely. Other BMPs cover all other forestry activities
including road construction and maintenance, landings, skid trails, and silvicultural practices.

Oregon's rules do not provide for analysis of cumulative effects. There is no watershed
analysis, but instead, Oregon takes a "bottom up" approach based on stream surveys. The
ODFW has surveyed most of the streams in Oregon to support forest practices regulation (M.
Solazzi, ODFW, Corvallis, OR, pers. com. m-. 1994). This approach relies on the ability to

86 Current Forest Practices
identify potential problems with cumulative effects by looking at conditions in stream channels
and riparian areas.

Washington

The Washington State Forest Practices Act of 1974 created a Forest Practices Board composed
of state agencies, county governments, industry, and the public. The Board promulgates the
forest practices rules. The first rules did not adequately. address fish habitat (Phinney et al.
1989) because no consideration was given to habitat except for temperature. All riparian trees
could be cut, sparing only the understory on certain temperature-sensitive streams. In 1987, a
group of concerned parties including state agencies, corporations, tribes, citizens, and technical
experts undertook negotiations and reached an agreement on new forest practices rules (the
Timber Fish Wildlife Agreement, or TFW). The rules were again revised to further address
environmental concerns in 1992.

Washington's rules recognize five types of waters:

Type 1. Special "Inventoried Shorelines."

Type 2. High value for fish, wildlife, and human use.

Type 3. Moderate-to-slight fish use.

Type 4. No fish but are important for water quality and have channels > 2 ft (0. 6 m)
wide.

Type 5. Other waters including perennial and intermittent streams with or without
defined channels.

Riparian management zones (RMZs) are required on Type 1, 2, and 3 streams but not on Type
4 and 5 streams unless warranted by site conditions (Table 7.6). In western Washington, the
minimum RMZ width is 25 ft (8 m), and the maximum width depends on stream type and width,
ranging from 100 ft (30 m) on Type I and 2 streams over 75 ft (23 m) wide, down to 25 ft (8
m) on small ( < 5 ft wide) Type 3 streams. For each type, buffer width can vary between the
minimum and maximum values, depending on extent of wetland vegetation or the width needed
for shade.

Prescriptions for the number of leave trees in RMZs in western Washington depends on stream
type and streambed substrate (Table 7.6). Fewer leave trees are required for streams with
boulder-bedrock channels than streams with gravel-cobble channels.. The number of leave trees
is greatest [100 trees per 1,000 ft (304 m) of stream on each side] for Type 2 streams with
gravel-cobble channels. Leave trees for Type 1 and 2 streams must represent the preharvest
stand. For large Type 3 streams, leave trees must be larger than 12 inches ( > 30 cm) dbh and
consist of two conifers per deciduous tree. For small Type 3 streams, leave trees must be larger
than 6 inches ( > 15 cm) dbh and equally conifer and deciduous. Leave. trees can be required
for Type 4 and 5 streams for site-specific reasons.

Current Forest Practices 87

Table 7.6. Washington's requirements for width and leave trees per 1,000 ft (304 m)
in Riparian Management, Zones (RMZs) in western Washington under the 1992 forest
practices rules.
Trees/1,000 ft, each side
Stream RMZ width Type/size Gravel- Boulder-
Class (ft) of leave trees cobble bedrock

1 and 2 25-100' Representative 50 25
(@!t75 ft wide) of stand

1 and 2 25-751 Representative 100 50
(<75 ft wide) of stand

3 25-50' Two 12-inch conifers 75 25
(@t 5 ft wide) per deciduous tree

3 25 One 6-inch conifer 25 25
(< 5, ft wide) per deciduous tree

4 & 5 0-252 Site specific 0-25 0-25

'Width can vary within the range shown, depending on extent of wetland vegetation.
2An RMZ is used only if deemed needed due to site-specific conditions.

Rules for eastern Washington are generally similar to those for western Washington. The RMZ
width for T@pe 1, 2, and 3 streams is 30 to 50 ft (9-15 m) on each side of the stream for areas
of partial harvest, and must average 50 ft (15 m) for clearcutting. The minimum leave tree
requirements [trees @!t 4 inches ( @t 10 cm) dbh] are 75 trees per acre (185 trees/hectare) for
boulder-bedrock streams, and 135 trees per acre (334 trees/hectare) for gravel-cobble streams.

Washington's BMPs comprehensively cover all forestry activities including activities within
RMZs, road construction, skid trails, slash disposal, landings, and silvicultural practices. Within
RMZs, trees must be felled away from Type 1, 2, and 3 waters except where impractical or
unsafe. If felled trees get into streams, they must be removed promptly, and'bucking is allowed
only as needed to remove the tree from the water. Trees may be felled into Type 4 waters and
bucked and limbed within the stream channel provided it is done carefully to minimize'
accumulation of slash. Downed logs imbedded in channels '(except Type 5) must be left
undisturbed.

The BMPs also regulate yarding activities. Cable yarding across Type 1, 2, and 3 waters is
prohibited, except where the logs will not damage the stream channel or the RMZ. Reasonable
care must be taken to avoid damaging residual vegetation. Tractors and wheeled skidders are

88 Current Forest Practices

not permitted in Type 1, 2, and 3 waters without Department of Natural Resources (WDNR)
approval and approval by the Departments of Fisheries or Wildlife. Log skidding is not
permitted within RMZs without WDNR approval, and skid trails are not pennitted within the
50-year flood plain. Tractors are not allowed on slopes where, in WDNR's opinion, they would
cause unnecessary damage to public resources. Log skidding across flowing Type 4 waters must
be minimized or employ constructed temporary stream crossings. No such restrictions apply for
Type 5 waters without RMZs.

Washington's forest practices rules address cumulative effects through watershed analysis. The
process is meant to assess watershed problems and sensitivities and be a basis for developing
appropriate prescriptions (Washington Forest Practices Board 1993). Watershed analysis is
divided into seven modules that separately address mass wasting, surface erosion, hydrology,
riparian function, fish habitat, water quality, and public capital improvements (e.g., roads and
bridges). The process is collaborative, involving scientists and managers representing
landowners, agencies, tribes, and interested public. The desired outcome is a management plan
for the watershed that responds to the resource concerns identified by scientific assessment.

Watershed analysis has four phases: startup, resource assessment, prescription writing, and
wrap-up. The startup phase forms teams, collects data, defines responsibilities, distributes
notifications, and develops a plan for watershed evaluations. During resource assessment, an
interdisciplinary team locates sensitive areas and assesses existing and potential impacts by
implementing the seven inventory modules. During prescription writing, a team of managers
and analysts identifies f6rest practices to prevent or minimize impacts. During wrap-up, the
team develops a monitoring plan to measure effectiveness of the prescriptions.

Other States

Forest practices rules of other states are generally not as comprehensive or restrictive as those
of the Pacific Northwest and Alaska. In most states, compliance with streamside protection
requirements is voluntary, and no monitoring is done (Brown and Binkley 1994).

Chapter 8
Analysis of Current Forest Practices

The rules for federal lands and for the five states described above have many elements in
common. They all have stream classification, regulatory BMPs, and riparian buffer zones.
They differ mainly in the size and management of buffer zones and in the assessment of
cumulative effects.

All agencies classify streams according to beneficial use (Table 8.1). They fall into three
classes:

1. Fish-bearing and domestic water supplies;
2. Non-fish perennial streams; and
3. Non-fish intermittent stream channels.

Non-fish intermittent streams generally include streams that show signs of flowing water and
have definite bed and banks. Areas without defined channels (e.g., ephemeral overland flow,
seeps, wet draws) are not classified, except in NFP/PACFISH and Washington. Agencies
primarily use stream classes to modify their requirements for buffer zones and BMPs. Greatest
protection is given to fish-bearing streams and least protection is given to non-fish intermittent
streams.

BUFFER DESIGNS

Forest practices rules affect two parameters of buffer zone design: buffer width and management
prescriptions within them. These parameters together determine how effective a buffer will be
in protecting fish habitat. A wide unharvested buffer obviously provides more protection than
a narrow, heavily harvested buffer. Most buffer zone designs, however, are between these
extremes.

Streams with anadromous fish have the widest buffer zones (Table 8.2). Minimum width ranges
from 25 ft (7.6 m) in Washington to 300 ft (91 m) on federal lands managed under the NFP or
PACFISH. Streams with lesser fish values (i.e., non-anadromous species) receive narrower
buffers in Alaska and Washington, but the others have similar buffers for all fish streams. Non-
fish perennial streams have narrower buffers than the fish-bearing streams or no buffers in some
cases. Non-fish intermittent streams, except on federal lands managed under NFP and
PACFISH or in Idaho, usually do not have buffers.

90 Analysis of Forest Practices

T6ble 8.1. Definitions of stream classes for federal and private lands in five states.

Regulator Stream Class Characteristics

Federal 1. Fish-bearing
(NFP & 2. Perennial non-fish
PACFISH) 3. Intermittent non-fish

Federal 1. Anadromous fish stream
(TLMP in II. Non-anadromous fish stream
Alaska) Ill. Non-fish stream

Alaska A. Anadromous fish stream, unconstrained channel
(southern B. Anadromous fish stream, constrained channel
coastal) C. Tributary to anadromous fish stream

California 1. Fish-bearing or domestic water supply
11. Non-fish stream; fish present within 1,000 ft downstream
Ill. No aquatic life; capable of sediment transport
IV. Artificial watercourse

Idaho 1. Fish-bearing or domestic water supply
11. Headwater stream with few if any fish

Oregon F. Fish-bearing
D. Domestic water supply
N. Non-fish stream

Washington 1. "Inventoried Shorelines"
2. High fish, wildlife, or human use
3. Low-moderate fish, wildlife, or human use
4. Non-fish perennial or intermittent streams
5. Intermittent stream having periods of spring or storm runoff

All agencies use additional site-specific factors to refine buffer requirements to fit local
conditions. On federal lands, riparian reserves and RHCAs can be adjusted for site-specific
conditions after watershed analysis. On private lands, four of the five states adjust buffer width
depending on hillslope gradient, yarding method, stream size, extent of wetland vegetation, or
streambed materials (Table 8.2). Idaho's buffer widths are set without regard to such site-
specific factors.

Analysis of Forest Practices 91

Table 8.2. Width requirements (in feet) for riparian buffer zones on federal [Northwest
Forest Plan (NFP) and Tongass Land Management Plan (TLMP) as amended by the
Tongass Timber Reform Act] and private lands in five states for the three common
stream classes. Buffer width on private lands is measured as horizontal distance in
Washington and as slope distance in the other states; therefore, buffers of the same
nominal width can be wider in Washington than in the other states, depending on the
slope. Reasons for ranges are in footnotes.

Stream Federal Private Lands
Class NFP' TTRA AK CA ID OR WA

Fish-bearing 300 100 66-1 002 75-1503 75 50_ 1004 25-1005

Non-f ish 150 0 506 50-1 003 30 0-704 0-257
perennial

Non-f ish 1 ob o 0 Site7 30 0 0-25 7
Intermittent8 Specif ic

'Riparian reserve widtR can be increased to include 1 00-year flood plain or other
factors.
2Anadromous fish streams: 66-ft no-cut buffer for non-bedrock streams; 100-ft
harvested zone for bedrock-constrained streams.
'Depends on adjacent hill slope and yarding method: wider buffers on steeper slopes
and for tractor yarding.
4Depends on stream size and region; no buffers for small perennial non-fish streams
in western Oregon.
'Depends on extent of wetlan Id vegetation and stream width.
150 ft or to the slope break, whichever is smaller.
7Buffer zone used only if deemed necessary by site-specific conditions determined
during preharvest planning.
'Non-fish stream channels with definite bed and banks that carry water part of the
year.

California requires a wider buffer in areas with steep slopes and further increases buffer width
up to a total of 150 ft (46 m) for tractor yarding. These adjustments respond to two concerns
about erosion and sediment production: buffers need to be wider to effectively filter sediment

92 Analysis of Forest Practices

in areas of steep slopes (Johnson and Ryba 1992) and tractor yarding causes more erosion than
cable yarding (Chamberlin et al. 1991).

Oregon and Washington adjust buffer width depending on stream size, giving larger streams
wider buffers. The assumption is that small streams do not need as wide a buffer as larger
streams. Compared to small streams, large streams are usually associated with a wider
floodplain (Sullivan et al. 1987) and, therefore, require a wider buffer for floodplain protection.
Large streams also have greater total energy for bank cutting and transporting sediment and
debris. Although this appears justified, the specific widths needed to protect different stream
sizes have not been identified by research; hence, the adjustments in buffer width are based only
on professional judgement.

Alaska and Washington adjust buffer width depending on streambed materials. The rationale
is that stream channels that are constrained by bedrock are less dependent on LWD and
vegetation for providing fish habitat than are unconstrained channels with gravel/cobble
substrate. Alaska increases the width of the buffer from 66 ft (20 m) for gravel/cobble channels
to 100 ft (30 m) for bedrock-constrained channels but allows complete harvest within the wider
zone.

Washington adjusts buffer widths for fish-bearing streams according to wetland vegetation. This
approach, however, may not provide long-term protection even to those wetlands. If adequate
sources of LWD are not provided by a strearnside buffer, the stream can downcut, which would
lower the local water table and cause wet areas to dry up (Beschta 1991). A more appropriate
approach in defining buffer zones is from an analysis of functions provided by strearnside areas
to the stream/riparian ecosystem (Belt et al. 1992).

The different prescriptions for minimum amounts of vegetation to be retained in buffers are
difficult to compare. Each state has a different approach. These include requiring no-harvest
zones (Alaska), retaining a percentage of overstory canopy (California), retaining a specified
number and size of trees (Idaho), setting targets for tree basal area (Oregon), and retaining a
percentage of preharvest trees (Washington). Usually, a combination of approaches is used.
Oregon, for example, has a no-harvest zone and specifies a number of conifers in addition to
its targets for basal area.

BUFFER EFFECTIVENESS

Based on review of 38 separate investigations, Johnson and Ryba (1992) concluded that, for
most riparian functions, buffers greater than 30 m (100 ft) are adequate, buffers 15-30 m (50-
100 ft) are minimal, and buffers less than 10 m (30 ft) are inadequate. With these
recommendations as a general guide, buffers on federal lands under NFP and PACFISH are
adequate for all riparian functions plus a safety margin to offset risks to habitat from unknown
or uncontrollable factors (Table 8.2). Buffer widths on private lands are barely adequate for
fish-bearing streams but minimal or inadequate for non-fish streams. This reduced protection
for non-fish streams is- understandable and appropriate given that management objectives for non-

Analysis of Forest Practices 93

fish streams on private lands do not include most riparian functions, but specifically target
sediment control for protection of downstream beneficial uses.

Shade Protection

States usually require a minimum level of shade-producing vegetation to be left along fish-
bearing streams. Washington, for example, determines minimum requirements for shade by
modeling predicted temperature increases based on expected changes in forest canopy. Idaho
requires retention of 75% of shading vegetation, California requires retention of 50% of the
canopy, and Oregon retains all trees within 20 ft of the stream.

Adequate shade usually can be provided by leaving a strip of trees next to the stream in a width
of about 25 rn (80 ft) (Johnson and Ryba 1992). Trees for shade can consist of unmerchantable
hardwoods and conifers. Buffer widths for fish-bearing streams on privatelands average near
the recommended width (Table 8.2) and should be adequate for shade if not harvested too
hea vily. Idaho's 75% shade requirement, for example, is close to the 80-90% canopy of old-
growth forest (Beschta et al. 1987). California's requirement of only 50% canopy retention,
however, appears too low, especially because California is on the southern margin of the range
of several species, including coho salmon, and increased temperature could make some streams
uninhabitable. These prescriptions for shade retention on fish-bearing streams need to be closely
monitored to ensure adequate control of stream temperature.

Shade requirements for non-fish perennial streams may be inadequate in some states, particularly
Idaho, Oregon, and Washington. In these states, buffers on non-fish streams can be less than
one-half the 80-ft recommended buffer width for shade protection (Table 8.2; Johnson and Ryba
1992) and substantial harvest is allowed within the buffers. In Alaska, both federal and private
lands have minimal buffers on non-fish streams, but a cooler climate probably helps mitigate
potential increases in temperature. Because of their usual small size, non-fish streams may be
adequately shaded by shrubs and other understory vegetation, and narrow, harvested buffers may
suffice for shade protection. Effectiveness monitoring is required to determine whether these
narrow buffers prevent cumulative increases in temperature in downstream fish habitats.

LWD Recruitment

Providing for LWD is more expensive than providing shade because it requires leaving
merchantable conifers. Four of the five states have specific requirements for the number of
leave trees in riparian buffer zones, whereas federal rules and Alaska use , no-harvest
prescriptions (Table 8.3). For high-value fish-bearing streams, the number of leave trees per
1,000 ft (304 in) of stream varies widely by state. For example, Idaho requires about 250 trees
ranging down to 3 inch (8 cm) dbh. Oregon's prescriptions are the most comprehensive. For
large fish streams, Oregon requires 40 trees > 11 inch (> 28 cm) dbh per 1,000 ft of stream,
a 20-ft no-harvest strearnside zone, and a standard target for tree basal area of 230 square ft per
1,000 ft of stream. This basal area is equivalent to nearly 350 11-inch-diameter trees.

The approximate level of protection for LWD recruitment can be estimated based on buffer
width and prescriptions for leave trees within the buffer. Buffer width determines the area from

94 Analysis of Forest Practices

Table 8.3. Required leave trees [per 1,000 ft (304 m), each side of stream] within
riparian buffer zones on federal and private lands in five states. NH = no harvest.
Stream Private Lands
Class Federal AK CA ID OR WA

Fish-bearing NH1 NH 2 23 242-267 4 0-40" 25-1 006

Non-fish O-NH 7 0 23 1408 0-309 0-25'0
perennial

Non-fish O-NH 7 0 0 1408 0 0-25'0
intermittent
'No harvest until after watershed analysis under NFP and PACFISH.
'No harvest for unconstrained channels; complete harvest for constrained channels
and streams with non-anadromous fish. Within no-harvest buffers, selective harvest
permitted through variations.
3Retain 50% overstory canopy, including 25% of preharvest overstory conifers and
two conifers > 16 inch (41 cm) dbh per acre within 50 ft of stream (2.3 trees per
1,000 ft).
4Number of leave trees depends on stream width, includes both hardwoods and
conifers, and ranges down to 3 inch (8 cm) dbh.
5All trees within 20 ft (6 m) of the stream are retained; buffer must meet basal area
targets and include from 0 to 40 conifers (minimum 8-11 inches dbh) per 1,000 ft of
stream, depending on stream size.
6Depends on stream size and streambed materials; hardwood/conifers representatiave
of stand.
7No-harvest buffers under NFP and PACFISH; complete harvest allowed under TLIVIR
'Size of retained trees: 3-8 inches (8-20 cm) dbh; includes both hardwoods and
conifers.
leave trees not required on'small Type N streams.
"Leave trees (conifers and hardwoods) required only for site-specific conditions.

which potential source trees can contribute LWD (Murphy and Koski 1989; McDade et al.
1990), and prescriptions determine how much of this potential material remains after timber
harvest.

Considering buffer width for fish-bearing streams, federal NFP and PACFISH buffers provide
full protection of LWD sources because the buffers are at least two site-potential tree heights in
width. Buffers on private lands, however, are generally not wide enough to fully provide for
long-term LWD recruitment. Most prescribed buffers on private lands are narrower than a
single site-potential tree height [usually 100-120 ft (30-40 m)]. The exceptions are streams in

Analysis of Forest Practices 95

California where side slopes exceed 30% (100-150-ft buffers); large streams (mean discharge
> 10 cfs) in Oregon (100-ft buffers); and certain large streams in Washington where buffer
zones are extended to include the wetland plant community (up to 100-ft buffers). All other
fish-bearing streams have narrower buffers with reduced potential sources of LWD, and all
buffers on private lands have some allowable timber harvest.

Based on buffer width, the buffers for representative fish-bearing streams on private lands in the
five states could provide approximately 90% of LWD sources present in mature conifer stands
if the buffers were unharvested and if they contained mature conifer forest or were restored to
that condition (Table 8.4). Timber harvest within the buffers, however, reduces LWD sources
to the minimum requirements for leave trees and other vegetation. These requirements are
lowest in California, where only 25% of the conifer overstory including two large conifers per
acre must be left, and is highest in Alaska, where a variance must be approved to remove
individual trees. The resulting overall protection of conifer LWD sources on private lands
ranges from only 23 % in California to 82 % in Alaska (Table 8.4)-.---`

Growth of trees during the timber rotation increases the trees for potential LWD. In Oregon,
for example, targets for conifer basal area for leave trees are set so that trees will achieve
desired future conditions halfway through a 50-year rotation. Oregon's rules are based on the
expectation that basal area will grow 59% within 25 years, thereby achieving the level of LWD
sources in a mature Douglas-fir streamside forest (T. Lorensen, ODF, Salem, OR, pers. comm.
1994). Assuming a similar growth rate (59% per 0.5 rotation period) in the other states, the
resulting LWD sources at mid-rotation would exceed 90% of the level in mature forest in Alaska
and Oregon, but would still be far below that level in California and Washington (Fig. 8. 1).

These comparisons of LWD recruitment depend on estimates of average or normal mature
forest. The valuefor-Washington, in particular, depends on how many trees occur in an average
mature streamside stand. Basal area and density of trees varies widely, and a single value for
the percentage leave trees in Table 8.4 fails to portray the large variation that occurs in the field.
Nevertheless, the values give a perspective of the relative level of protection for LWD sources
under- similar hypothetical conditions.

For comparison purposes, this evaluation of buffer effectiveness for LWD recruitment assumed
that strearnside areas contained mature forest. Many riparian areas in the Pacific Northwest,
however, have second-growth vegetation consisting of hardwoods and brush (Gregory et al.
1990). In such cases, leaving a higher percentage of existing trees may not increase conifer
LWD for the stream nor help reestablish conifers in the riparian area (Bilby and Bisson 1991).
In these cases, regulations should encourage activities that modify riparian vegetation leading
to desired future conditions of appropriate mature native forest species.

Oregon's approach provides a prototype model for managing second-growth riparian areas to
achieve desired future conditions for both fish and timber. If the buffer lacks enough conife 'rs
to meet targets, no harvest is allowed. Monitoring data in Oregon indicate that because of the
current condition of riparian forests, minimal tree harvest occurs in buffers on private lands (T.
Urensen, ODF, Salem, OR, pers. comm. 1995). To reestablish conifer stands along streams,

96 Analysis of Forest Practices

Table 8.4. Comparison of minimum level of protection for conifer LWD sources for
representative anadromous fish streams in federal (NFP) and private lands in five
states. For comparisons, preharvest buffers are assumed to have mature conifer
forest.
Federal AK _ffA__ ID OR WA
NFP Type A Class I Class I Type F Type 2
Class 1 40% slopes 15 ft wide > 10 cfs' < 75 ft wide

Buffer width (ft) 300 66 100 75 100 25-75

% LWD source trees
in unharvested buffer 2 100% 96% 92% 85% 92% 40-85%

% Prescribed leave trees 100% 85 %3 25 %4 58 %5 63 %6 38 %7

% LWD sources after
timber harvest' 100% 82% 23% 49% 58% 32%

'Mean annual strearnflow in cubic ft per second.
2Values obtained from graphs in Murphy and Koski (1989) for Alaska and in McDade
et al. (1990; model for mature conifers) for the other states. Buffers are assumed to
have mature conifer forest.
3Value based on 15% harvest rate (R. Harris, Sealaska Corp., Juneau, AK, pers.
comm. 1993).
4Value based on 25% retention of overstory conifers.
'Value obtained by comparing estimated basal area of prescribed leave trees (87 sq.
ft per 1,000 ft) to estimated basal area in mature streamside stands on private lands
in eastern Oregon (150 sq. ft/1,000 ft; I Lorensen, ODF, pers. comm. 1994).
'Example for Coast Range. Value obtained by comparing standard basal area target
to the normal yield of mature Douglas-fir forest adjusted for incomplete stocking and
tree mortality (T. Lorensen, ODF, Salem, OR, pers. comm. 1994).
'Example for western Washington. Value obtained by comparing the 100-leave-tree
requirement to the mean number of trees in mature strearnside forest in the Western
Cascades, corresponding to the maximum 75-ft buffer (263 trees/1,000 ft; T.
Lorensen, ODF, Salem, OR, pers. comm. 1994).
'Value calculated by multiplying the % source trees in an unharvested, mature-conifer
buffer times the % prescribed leave trees.

Oregon allows alternative prescriptions, such as increased harvest followed by conifer planting
in "conversion blocks" alternating with "retention blocks" with lesser harvest (Newton et al.
1995). Oregon further ensures some immediate LWD recruitment by providing basal area
credits when operators add trees to streams.

Analysis of Forest Practices 97 97

Predicted % 100 ...........
LWD sources at ...........
Mid-rotation ...........
...........

........... ...........
........... ...........
...........

........... ........... ...........
........... ........... ...
...........
.... . ...........
........... ...........
........... ...........
..........
...........
80
...........

........... ...........

.......... ...........
........... ...........
........... ...........
60
........... ...........
........... ........... ...........
........... ...........

........... ........... ........... ...........
. ........... ........... ........... ...........

...........
.... ...... ...........
...........
...........
40 -
........... .......
...... ..... ...........
....... ........... ...........
....... ........... ........... ...........
............ ............
........... ........... ...........
... .......
........... ............ ............
........... ........... ........... ...........
........... ............ ........... ...
........... ........... .......
20 - ...........
... ........... ...........
.... .... ........... ...........
........... ........... ............ ........... ...........
........... ........... ........... ........... ...........
...........
........... ........... ...........
........... ........... ........... .......

........... ........... ...........
........... ........... ...........
........... ...........
........... ...........
0 ............ ............. ............. .............
Federal Alaska California Idaho Oregon Washington

Figure 8. 1. Predicted sources of conifer LWD in buffer zones at mid-rotation for
representative fish-bearing streams as a percentage of MID sources present in mature
conifer stands. Values are based on federal and state requirements for buffer width
and leave trees and assumes mature conifer forest in preharvest buffers and a 59%
increase in MID sources during the first one-half of a timber rotation (see text for
explanation).

Prescriptions for buffers in other states, such as Washington's requirement for leave trees that
are representative of the existing stand, do not encourage the desired result of improving riparian
stands. Oregon's and California's rules also directly address the need for conifer LWD by
specifying that leave trees consist of conifers, whereas Idaho and Washington allow both conifers
and hardwoods to qualify as leave trees.

A no-harvest buffer zone is most appropriate along fish-bearing streams where strearnside areas
consist of mature native forest. Where riparian forests are degraded by past logging, a no-
harvest prescription limits options for silvicultural treatments for restoring riparian functions for
fish habitat (Bilby and Bisson 1991). A no-harvest prescription, unless it provides for
"variations," also does not allow landowners to harvest valuable timber from the stand in site-
specific cases as long as habitat is protected.

Alaska's approach illustrates the use of no-harvest buffers in mature forest, with "variations"
allowing selective harvest. A 66-ft, no-harvest buffer zone is used along unconstrained
anadromous fish streams to leave over 90% of LWD source trees present before harvest.
Variations can be granted to landowners to harvest additional specific trees whose removal is

98 Analysis of Forest Practices

unlikely to adversely affect fish habitat. State habitat biologists and landowners debate the
harvest of individual trees, and about 80% of variation requests are approved. The variation
process results in about 15 % of trees > 12 inch ( > 30 cm) dbh within the buffer zone being
harvested (R. Harris, Sealaska Corporation, Juneau, AK, pers. comm. 1993; Resource
Development Council 1994). The state resource commissioners have found that the process
generally works satisfactorily (ADFG 1994 Memorandum), but effectiveness of resulting buffers
has not been evaluated.

Specifying a number of leave trees in buffers is a common way to set a minimum level of
protection for I_WD recruitment. Four of the five states require leave trees for fish-bearing
streams, and three states require leave trees for perennial non-fish streams. Usually leave trees
include many small trees [e.g., down to 3 inch (8 cm) dbh in Idaho] and only a few large trees.
The size of these largest trees [ > 11 inch (28-50 cm) dbh] is generally appropriate to provide
stable LWD in streams, but the smaller trees are probably ineffective for LWD (Bilby and
Wasserman 1989). Current requirements in the four states are to leave only an estimated 23 %
to 58% of potential LWD compared to the sources present in mature conifer forest (Table 8.4).
To provide optimal fish habitat, the number and size of leave trees need to be increased where
additional large conifers are available.

Buffers on small non-fish streams, except for federal lands managed under NFP and PACFISH,
are generally not adequate to provide LWD for the stream. All states except Alaska require
leave trees along some non-fish perennial streams, but not enough to fully maintain Lwa Only
Idaho routinely requires leave trees along intermittent channels (Table 8.3).

Longer term, the lack of LWD sources along small headwater streams can adversely affect
downstream habitat in several ways. Reduced sources of -LWD can reduce sediment storage in
small headwater streams, resulting in more rapid sediment delivery to downstream reaches
(Sullivan et al. 1987). Headwalls of small headwater streams can be important sources of LWD
to downstream reaches via debris torrents (Swanson et al. 1987); lack of a buffer zone in these
areas eliminates this function.

Sediment Control

Controlling sediment delivery is most impo rtant along small non-fish streams and intermittent
channels because of their dense distribution [accounting for more than 50% of the total length
of stream channels in a watershed (Reid and Ziemer 1994a)] and their capacity to transport
sediment to downstream reaches. These stre 'ams, however, except on federal lands managed
under NFP and PACFISH, generally have minimal buffers (Table 8.2). Perennial non-fish
streams do have buffers in Idaho and California, and they sometimes have buffers in Washington
if deemed needed by site-specific conditions. Perennial, non-fish streams do not have buffers on
federal lands managed under TLMP, nor along small 1ype N streams in western Oregon.
Where buffers are left on perennial non-fish streams, they are usually heavily harvested (Table
8.3). Intermittent non-fish streams (with deftte bed and banks) consistently have a buffer zone
only on NFP/PACFISH lands and in Idaho; California and Washington sometimes provide a
buffer for site-specific conditions.

Analysis of Forest Practices 99

The buffers for small non-fish streams appear to be minimal or inadequate for sediment control.
The recommended buffer width for sediment filtering ranges from about 26 to 150 ft (8-46 rn),
depending on hillslope (Johnson and Ryba'1992), whereas the average buffer width on private
lands is 40 ft (12 m) for perennial non-fish streams and usually 0 ft for intermittent non-fish
streams (Table 8.2). California is closest to the recommended width by requiring 50-100-ft
buffers on perennial non-fish streams. A high level of timber harvest within the buffers,
however, probably compromises their effectiveness as sediment filters. Because of the narrow
buffers and high level of harvest allowed along small non-fish streams, preventing sediment
pollution relies heavily on BMPs that restrict felling and yarding practices along streambanks.

BEST MANAGEMENT PRACTICES

The BMPs used in the five states are generally similar in that their principal objective is to
prevent sediment pollution. Each state has a suite of BMPs for felling, yarding, slash disposal,
site preparation, road construction and maintenance, and other activities designed to prevent
disturbances to stream channels, riparian areas, and unstable soils, and minimize sediment runoff
from roads and skid trails. Each state monitors effectiveness of its BMPs, but monitoring
programs are only recently being developed, and current BMPs have not yet been fully
evaluated.

Three BMPs pertaining to buffer zones are particularly important in protecting streams from
disturbance and preventing downstream sediment impacts from timber harvest along small non-
fish streams. These BMPs determine 1) whether trees can be felled into and limbed within
stream channels, 2) whether cable yarding can cross streams with full or partial log suspension,
and 3) whether tractors and other ground-based yarders can operate within streams or their
buffer zones.

The states' BMPs for these: activities carefully protect fish-bearing streams, but small non-fish
streams are not as carefully protected. All states require that trees be felled away from and not
bucked and limbed in fish-bearing streams; however, several states allow felling, bucking, and
limbing in small non-fish streams. Washington and Idaho, for example, allow felling, bucking,
and limbing in perennial non-fish streams as long as care is taken to minimize accumulation of
slash. Cable yarding across fish-bearing streams must have full suspension and prior approval
in Oregon and Washington, except for small non-fish streams. Tractor yarding is generally not
allowed across fish-bearing streams and not allowed in most perennial non-fish streams except
at constructed temporary crossings; however, all the states allow some log skidding across
intermittent non-fish channels. For example, Oregon allows log skidding across dry streambeds
where the disturbance is less than it would be to construct temporary crossings. Washington and
California allow log skidding across intermittent non-fish stream channels unless a buffer zone
is deemed necessary by on-site inspection.

Because small non-fish streams are particularly important for controlling sediment delivery and
because buffer zones along them are usually narrow and heavily harvested, BMPs for felling and
yarding must be closely monitored to ensure that they are effective. Effective BMPs are

100 An alysis of Forest Practices

essential because they may be the only practical means of protecting the numerous non-fish
headwater streams in managed timberlands while other resource activities continue.

Differences in BMPs among the five states and among regions within the states are due to
different emphasis in addressing different logging practices, forest types, and watershed.
conditions. For example, tractor yarding is probably the most widely used method in California,
but it is used much less in the other states. Selective tree harvest is also used extensively in
drier regions, such as in eastern Oregon, whereas, clearcutting predominates in coastal regions.
The BMPs in the different states must also contend with very different potential for soil erosion.
Watersheds in different geologic provinces in the five states produce vastly different amounts
of sediment. Streams in Oregon's Coast Range,, for example, annually export 53-102 metric
tonnes per kml compared to 2,600 tonnes per kM2 in,,-northern California's Coast Range
(Hawkins et al. 1983).

The BMPs in regions with high erosion potential need to be more- restrictive to prevent sediment
pollution', yet tractor yarding, the most disruptive yarding method, is allowed on steeper slopes
in California (up to 65 % slope) than in the other states (e. g., up to 45 % in Idaho and 35 % in
Oregon). Preventing sediment pollution in northern California presents a major challenge to
watershed managers because of the combination of extensive tractor yarding on steep slopes in
one of the most erosive landscapes in the world.

CUMULATIVE EFFECTS MANAGEMENT

Both federal and state programs consider cumulative effects. The FS and BLM are devoting
effort to watershed analysis in the Pacific Northwest, but the FS has tried watershed analysis on
only three watersheds in the Tongass National Forest in Alaska. Washington has a watershed
analysis program, and California and Idaho have analogous systems for analyzing watershed
condition and prescribing precautionary, BMPs to help avoid cumulative effects. The states of
Oregon and Alaska do not, conduct watershed analysis nor have a process for evaluating
cumulative effects.

Applying watershed analysis on private land is difficult for several reasons. The cost of
watershed analysis adds a burden to landowners. Most landowners do not have the personnel
to do the analysis and must look outside their companies . for certified analysts.,, Getting
cooperation and coordination among different landowners in a watershed is often difficult.

Current programs address these difficulties in different ways. In Washington, watershed analysis
is conducted by the State Department of Natural Resources in cooperation with landowners. In
California and Idaho, the cumulative effects analysis is done by private certified foresters,and
evaluators. Idaho coordinates its cumulative effects analysis by forming committees of
landowners. In watersheds with mixed federal and private ownerships, the FS and BLM can
reach out to form cooperative arrangements with private landowners.

Watershed analysis is probably most effective if it provides managers with information necessary
to write management prescriptions that address' site-specific concerns identified in the analysis.

Anatysis of Forest Practices 101

Federal watershed analysis does not provide prescriptions or alternatives, but is only a process
to gather and analyze data for input into decision processes. The Washington or Idaho methods
of doing watershed analyses may be the best prototype models to use for prescriptive watershed
analysis on private lands with mixed ownerships. Watershed analysis, however, is in its infancy,
and more experience is needed to develop its potential for increasing the effectiveness of habitat
protection on private timberlands.

Chapter 9
Habitat Restoration

Habitat restoration is one element in a comprehensive program of watershed management that
emphasizes habitat protection. The FS and BLM, for example, recognize watershed restoration
as one of four components in their Aquatic Conservation Strategy which also includes key
watersheds, riparian reserves, and watershed analysis (FEMAT 1993). Habitat restoration is an
interim measure until watersheds recover under good management, not a mitigation or an
exemption from stream protection.

Stream restoration science is founded on hydrologic principles, stream ecosystem theory, fish-
habitat relationships, the concept of limiting factors, and a growing awareness of human impacts
on stream ecosystems (Koski 1992). Although the term "restoration" infers returning to an
original state, restoration of heavily impacted streams to original condition is generally not
practical (Herricks, and Osborne 1985). Habitat restoration is really a pragmatic mix of
protection and rehabilitation to some improved level consistent with multiple use of the
watershed.

The approach to habitat restoration described here applies principally to forest lands affected by
past timber harvest practices. The approach and techniques may need to be modified to apply
to lands affected by mining, grazing, agriculture, urban development, and other uses where
considerations in restoring fish habitat may be different than for streams in altered forests (e.g.,
Ferguson 1991). Numerous reports, workshops, and training sessions have covered the topic of
stream restoration (e.g., Gore 1985; Hunter 1991; Reeves et al. 1991; Koski 1992). This
chapter briefly reviews the procedures for restoration, presents examples of ongoing restoration
programs, and discusses the role of restoration in an overall watershed management program.

RESTORATION PROCEDURES

To be successful, stream habitat restoration requires a holistic approach directed at the entire
watershed to ensure that it addresses all major environmental factors affecting the stream
ecosystem (Koski 1992). Resource analysis at the scale of the river basin should precede
restoration of individual watersheds composing the basin to provide a broad context. Watershed
analysis should be used to detennine how the watershed functions, which parameters are outside
the range of natural variability, and what the restoration potential is (Kershner 1993). Any
restoration project should be nested within a larger program of landscape management that
protects, maintains, and restores ecosystem structure and function (Gregory 1993b).

104 Restoration

Sound restoration requires a solid foundation on ecological principles and a clear recognition of
the dynamic nature of streams and adjacent forests (Gregory 1993b). The goal is to reestablish
the ability of the watershed to maintain its functions and organization without continued human
intervention. Most importantly, practices that caused degradation need to be changed before
attempting restoration.

The most important technical elements of a holistic restoration program are 1) Mpland restoration
to control erosion, 2) riparian restoration to restore functions of strearnside vegetation, and 3)
instrearn restoration to improve habitat structure by physically modifying stream channels or
their flood plains.

Upland Restoration

A first step in restoration is to initiate "upland restoration" to begin recovery of watershed
hydrologic and erosional processes. This is a broad-based program to control erosion from
roads and bare soils, restore natural strearnflow regimes, and manage all uses of the stream and
watershed.

A high priority in upland restoration is to address problems with roads (Pacific Rivers Council
1993a). Existing needed roads should be stabilized, and abandoned and unneeded roads should
be closed and shaped to stable contours and to drain properly without maintenance (Furniss et
al. 1991). Dirt roads should be surfaced with gravel or asphalt to reduce sediment production
from road usage (J. Anderson, FS, Baker City, OR, pers. comm. 1994). Road obliteration can
prevent most future erosion if road surfaces are backfilled, stream crossings are removed, stream
channels are reconstructed to stable configurations, and all bare surfaces are revegetated (Fig.
9.1).

Another first step in restoration is to provide access for fish to suitable habitat where blocked
by road crossings. Culverts should be improved or replaced with bridges when necessary to
allow upstream fish passage. Restoration work should also address migration blockages caused
by other watershed impacts. Fishways can be constructed to provide access where sediment
from landslides or excessive erosion deposited in the stream blocks upstream migration (Flosi
and Reynolds 1991). Where the sediment forms thick alluvial fans at tributary mouths, boulder
fishways can take advantage of scour from the main stem during high strearnflow to maintain
the fishway (S. Downey, CDFG, Redway, CA, pers. comm. 1994).

Riparian Restoration

Restoration of riparian areas promotes long-term recovery of numerous important riparian
functions that strongly influence fish habitat in streams (Beschta 1991; Chan 1993; Everett et
al. 1994). Past riparian harvest combined with other watershed impacts have left many riparian
areas in degraded condition with poor prospects for recovery (Chan 1993). Impacts from
homesteading, grazing, and logging along streams in western Oregon and Washington resulted
in development of homogeneous alder and salmonberry (Rubus spectabilis) communities in nearly
all riparian areas (Gregory et al. 1990; FEMAT 1993). In this region, riparian areas have few

Restoration 105

Befor - -----

After

Figure9.1. Unused and unneeded roads should be "put to bed" by removing culverts
and outsloping road surfaces to drain properly without maintenance. (After Furniss
et al. 1991.)

trees larger than 10 inches (>25 cm) diameter growing within 100-200 ft (30-60 m) of the
stream, and recruitment of large wood may be deficient for decades (FEMAT 1993).

Loss of large conifers from riparian areas worsens the effects of floods on fish habitat. Alder-
dominated flood plains do not have the structural integrity to resist damage from excessive scour
during high water and debris torrents, whereas large conifers naturally could withstand high
flows, hold debris jams, and'reduce scouring of the flood plain (Chan 1993). Once large
conifers are removed, continual floodplain scouring may prevent their natural reestablishment
without restoration to stabilize the stream channel (J. Barnes, FS, Arcata CA, pers. comm.
1994).

In other situation, absence of LWD derived from large conifers can result in channel erosion and
downcutting during high strearnflow because LWD from alder and other hardwoods is inadequate
for structuring the stream channel (Andrus et al. 1988; Heimann 1988). This downcutting
lowers local water tables and breaks the linkages between the stream and its flood plain because
the lower channel prevents the stream from overflowing its banks (Beschta 1991). Off-channel
moisture recharge and storage are reduced and base strearnflow declines, which is particularly
troublesome in regions with summer droughts. A prime objective of adding LWD, boulders,
and other "roughness elements" in bedrock channels is to rebuild the stream's aquifer (L. Hood
and L. Burton, FS, Mapleton, OR, pers. comm. 1994). Downcutting also reduces the natural
disturbance of the flood plain during peak strearriflows. This natural disturbance is needed for

106 Restoration

the establishment of conifers, and lack of floodplain disturbance encourages the succession of
riparian areas to shrub vegetation (S. Chan, FS, Corvallis, OR, pers. Comm. 1994).

Silvicultural treatments have potential long-term benefits in restoring habitat functions of riparian
areas. A treatment being used and tested extensively in western Oregon is underplanting and
thinning to reestablish native conifers (Emmingharn et al. 1989; Chan 1993). Because the
streams have been downcut, boulders and LWD structures are also added to the stream in
conjunction with this riparian restoration to reestablish stream-floodplain linkages (L. Burton,
FS, Mapleton, OR, pers. comm. 1994). Thinning of overdense stands also helps reduce fire
risks. Caution is warranted, however, because few studies have been completed from which to
judge effectiveness.

Under suitable conditions, conifers can naturally become dominant over alder (J. Henderson, FS,
Seattle, WA, pers. comm. 1995). If Douglas-fir begins to grow at the same time that alder
becomes established, it may compete successfully and develop a dominant canopy well before
alder becomes senescent. Western hemlock and western red cedar are shade tolerant and can
grow slowly under alder and become dominant as the alders begin maturing and dying. On
cool, wet sites, Sitka spruce grows steadily in either openings or beneath alder canopy, and may
dominate over alder after 60-90 years (Henderson et al. 1989).

In many cases, however, degraded riparian vegetation may not recover without active
management intervention (Sedell et al. 1989). Alder-dominated riparian areas are especially
susceptible to poor conifer regeneration due to low light, lack of a conifer seed source, and lack
of downed wood or mineral goil needed for a suitable seedbed. The relatively short life of alder,
scarcity of conifer seedlings, dense shrub understories, and lack of natural floodplain disturbance
because of stream downcutting indicate that many areas would eventually succeed to shrubs
without management intervention (Hibbs et al. 1991).

Although some floodplain disturbance is desirable for the establishment of conifers, excessive
flooding and severe scouring of the floodplain may prevent conifer establishment (Henderson
1978). Excessive scouring could result from upland disturbances in the watershed and a lack
of large conifers in the riparian area. Large conifers help protect the floodplain from scour
because they can withstand flood damage better than alder and can trap flood-entrained debris.
Until large conifers can become established, floodplain biological communities are more
susceptible to flood damage and may be kept in an early successional stage dominated by alder
and salmonberry.

Successful restoration of riparian areas must involve active management for a long time (Chan
1993). Because of strong competition, simply planting trees in alder and salmonberry without
at least partially removing the overstory and understory is unlikely to succeed. Managers will
have to monitor tree growth and survival, and periodically remove understory shrubs to ensure
L
the reestablishment of conifers. Enlightened management policies and practices are needed that
will provide the maximum beneficial effects of riparian vegetation on stream hydrology and
channel morphology (Beschta 1991). Managers should move toward policies that protect, re-
establish, and encourage the functional attributes of riparian vegetation.

Restoration 107

Instrearn Restoration

After upland and riparian problems have been addressed, the third restoration step is to improve
instream habitats to increase carrying capacity and fish survival. This is an interim "fix" until
the natural long-term recovery of the watershed has begun (Koski 1992). In this step, the
primary focus is on using instream structures to alleviate limiting factors, such as to retain
spawning gravel or create additional rearing pools. Success of instream structures depends on
condition of upslope areas and the continuing hydrologic response to past and ongoing watershed
disturbance. Thus, instream structures are recommended only as part of a comprehensive
watershed program (Koski 1992).

Though an interim measure, instream restoration may be crucial as part of a program to recover
anadromous salmonids while long-term restoration measures have time to become effective
(Koski 1992). Attaining desired levels of channel complexity and other habitat conditions may
best be achieved in the short term with instrearn structures until the recovery of watershed
hydrologic processes and riparian forests provide for long-term maintenance of the stream
channel.

Instrearn restoration has been criticized as ineffective (Frissell and Nawa 1992). In the past,
many instrearn restoration projects proceeded without adequate planning and evaluation, and
commonly prescribed structures were often inappropriate or counterproductive (J. Anderson, FS,
Baker City, OR, pers. comm. 1994). Many projects failed or did not demonstrate their
effectiveness. Conversely, instrearn restoration projects that were accompanied by careful
planning, monitoring and evaluation, and constructed by experienced biologists have been
successful in improving fish habitat (House et al. 1991; Crispin et al. 1993).

Instrearn projects have mainly targeted one species (coho salmon) although associated species
may also have benefitted. Generally, however, restoration efforts should take a "community
approach" which emphasizes recovery of native biological communities with their full diversity
of aquatic and riparian-dependent species (Sedell and Beschta 1991). Care must be taken to
avoid negative effects on other species when altering instream habitat for salmonids. Species
like the foothill yellow-legged frog (Rana boy1ii), which has declined alarmingly in California
(Welsh et al. 1991), can be adversely affected by instrearn structures (Fuller and Lind 1994).
Similarly, altering habitat to favor one salmonid species may negatively affect another. Adding
LWD for coho rearing, for example, may decrease spawning habitat for pink salmon. The goal
of the community approach is to restore ecosystem structures and functions that support diverse
biological communities including healthy salmonid populations. This approach provides for
continued viability of coexisting salmonid stocks, as well as other fish and wildlife species.

Not taking a watershed perspective has been one of the main reasons for failure of instrearn
structures. Frissell and Nawa (1992) evaluated instrearn structures in 15 streams in western
Oregon and Washington after a flood. Damage to structures was widespread in streams with
recent watershed disturbance, high sediment loads, and unstable channels. Many structures
failed because of altered flow regimes, increased sedimentation, or debris torrents from upslope
activities.

108 Restoration

Instrearn structures are most appropriate where upslope portions of the watershed have stabilized
and the major habitat problem is lack of physical structure in the stream channel. In such cases,
instrearn structures can provide great benefits. For example, after 8 years, 86 % of 812
structures on 10 streams were fully successful in restructuring stream reaches by increasing
gravel substrate, instrearn cover, pool habitat, and total usable habitat (House et al. 1989). In
another evaluation, 98% of 200 instrearn structures were still functioning after 1-4 years, and
they had increased pool area and off-channel habitat for coho salmon nearly five fold (Crispin
et al. 1993). Addition of conifer logs to debris-poor streams and construction of off-channel
alcove habitats can increase salmonid smolt production several fold (Fig. 9.2; Solazzi and
Johnson 1994).

An important benefit of instream structures is that they can help reestablish linkages between the,
stream and its flood plain (House et al. 1991). By trapping sediments and increasing channel
complexity, instrearn structures can raise the water table and expand the stream's flood plain,
thus reversing the downcutting that occurred historically from the loss of LWD (J. Cederholm,
WDNR, Olympia, WA, pers. comm. 1994). The expanded floodplain also helps reduce adverse
effects of peak strearnflows because flood waters can spread out over a broader area. This also
increases water storage capacity, which helps augment streamflow during droughts (Beschta
1991).

Most instrearn restoration projects have been directed at improving migration, spawning, or
rearing habitat. Instrearn projects usually attempt to mimic factors that shape and stabilize the
stream channel, store sediment, create pools, dissipate stream energy, and provide diverse
habitat for either spawning or rearing (Fig. 9.3; Koski 1992). Barriers have been removed and
fishways constructed to provide access to habitat. Stream gravel has been cleaned or trapped
to improve spawnifig, and spawning channels have been constructed to increase spawning area.
Diverse structures have been added to stream channels to provide summer and winter rearing
habitat, and streambanks have been protected to deepen streams and reduce floodplain scour and
channel erosion.

Many types of instrearn structures have been used in stream restoration projects (Koski 1992).
In the Pacific Northwest, most structures have been installed in streams of fourth to fifth order
with normal peak flows about 6-60 m' s-1 and with channel gradients of I to 3 %. In large
streams, boulders, whole trees, or cabled logs are used for main-channel structures, instrearn
cover, and bank protection.

Restoration practitioners use a variety of techniques and structures, including:

Dams formed by cross-stream structures simulate natural debris jams or boulder dams in natural
streams. Dams are used to create plunge pools downstream and dammed pools upstream, to
collect gravel, or sort sediments.

Deflectors, also called wing dams or jetties, have been one of the most commonly used
structures to improve fish habitat. Deflectors simulate obstructions that divert strearnflow and
are used to create pools and cover by narrowing and deepening the stream.

Restoration 109

10,000 3,000
8,000 (A) Treated -0- 2,500 - Trea!@@

2,000
6,000 -
W
Contr 1 .01,500
E
Z q'VUV -
0
0 1,000 -
,2,000 - trol
500 -

0 0
1988 1990 1992 1994 1988 1990 1992 1994-
Pre- Post- Pre- Post-
treatment treatment treatment treatment

Figure 9.2. Increases in number of outmigrant coho salmon smolts after experimental
addition of large woody debris to debris-poor streams in the Alsea River basin (A) and
Nestucca R. basin (13) in western Oregon. Treatment occurred in summer 1990. (Data
are from Solazzi and Johnson 1994.)

Cover consists of overhead or instrearn features, such as overhanging vegetation, woody debris,
brush bundles, root wads, boulders, substrate interstices, turbulence, and depth. Adding cover
combined with other instream structures makes structures more usable by fish. The objective
in adding cover is usually to reduce predation and increase fish survival during peak flows.

Stream banks are sometimes protected from scour with revetments or riprap of boulders, woody
debris, or brush bundles. Planting shrubs and trees also helps establish root systems that
stabilize the-bank and provide overhanging cover for fish. Riprap combined with other instrearn
structures stabilizes the created habitat. Riprapping can be used to armor streambanks to protect
the toe of unstable hillslopes (Flosi and Reynolds 1991). Excessive riprapping, however, can
be detrimental by eliminating side channels, pools, and other complex features (Andrus 1991).

Off-channel habitat, such as pools and alcoves, are constructed or blasted into flood plains along
low-gradient streams to provide cover for rearing juvenile salmonids. Protected from peak
strearnflows, these areas can provide important overwinter habitat (Cederholm et al. 1988;
Nickelson et al. 1992b).

Beavers can be introduced or encouraged to colonize an area by planting aspen and other food
trees (Andrus 1991). Beaver ponds provide important winter habitat for juvenile coho salmon
1,eateu
@Contr @1

(Nickelson et al. 1992a), as well as provide other beneficial hydrologic functions for stream
ecosystems (Beschta 1991).

110 Restoration

Flow

LOG DEFLECTORS
- scours pool
- direct meanders
rearing, adult holding
TREE
- provides cover
- bank protection
rearing
adult holding
BOULDER GROUPS

- scours pool

- cover
rearing loop ROOTWADS

cover
rearing
DOUBLE LOG DIAGONAL WIER

scours pool
- collects gravel CHANNEL MARGIN
- dissipates velocity LOG-BOULDER COMPLEX
adult holding/ spawning - complex of pools,
rearing velocities and cover
- collects, stabilizes gravel
rearing
adult holding

BOULDER DEFLECTORS
wler water away 0-61
collects bedload
scours pool
bank prollectlon
rearing LOG UPSTREAM V WIER
- scours pool
- collects gravel
rearing
adult holding/ spawning

Figure 9.3. Diagram showing variety of instrearn projects that shape and stabilize the
stream channel, store sediment, create pools, dissipate stream energy, and provide
diverse habitat for either spawning or rearing. (From Koski 1992; reprinted with
permission from Maryland Sea Grant College.)

Restoration

Restoration Planning and Evaluation

Planning for habitat restoration has two levels: program planning and project planning (Everest
et al. 1991). At the program level, managers should consider coordination of financial and
personnel resources and priorities for watersheds and species. The target species and the
proposed methodology should be carefully considered so that one species is not emphasized at
the expense of others. An ecosystem approach to restoration targets habitat complexity to
benefit a diversity of species and life stages (Sedell and Beschta 1991). Interdiscipynary
consultations are needed to ensure success, and projects should be designed by professionals
from as many relevant disciplines as possible (Gregory 1993b).

Project-level planning considers specific details of proposed projects, including the size of the
area and the time allotted for inventory of fish and habitat. The project watershed should be
larger than 50 km@ to account for seasonal changes in fish distribution (Everest et al. 1991).
Ideally, data on fish distribution and habitat should be available for at least'l year for analysis
of limiting factors (Koski 1992). Project planning is closely coordinated with program planning
and follows a stepwise sequence: 1) pre-improvement inventory, 2) limiting factor analysis, 3)
site selection, 4) techniques and materials selection, 5) implementation plan, and 6) project
evaluation.

The pre-improvement inventory is a crucial step in which watershed attributes are- inventoried
to identify habitat problems. Surveys of hillslope erosion, riparian vegetation, stream channels,
and fish populations provide a basis for analyzing limiting factors and baselines for later
evaluations (M. Solazzi, ODFW, Corvallis, OR, pers. comm. 1994).

An analysis of limiting factors must be completed before any habitat enhancement is begun
(Reeves et al. 1989). Although,the limiting factor concept is useful in identifying limitations to
salmonid production, it can oversimplify complex ecological processes (Hall and Baker 1982).
As many as 73 factors could potentially limit fish production in a hypothetical stream with three
or more salmonid species, each with different age classes and habitat requirements (Everest and
Sedell 1984). Historical logging affected multiple habitat factors in streams, and one or more
of the factors could be limiting to fish production. A thorough knowledge of habitat
requirements and life histories of the endemic stocks in a stream can help in identifying limiting
factors and indicating approaches for restoration.

The emphasis in choosing materials and techniques should be to simulate natural habitat. If
woody debris is the dominant feature, log structures are used; if bedrock or boulders dominate,
boulder structures are used. Particularly in areas of heavy use by sport angling, the project
should appear natural. Conifer logs are more effective and persist longer than alder logs.
Adding alder logs to a stream is inexpensive, but has only limited value, as effects diminish after
several years (C. J. Cederholm, WDNR, pers. comm. 1995). Experienced practitioners use
observations of stable materials in the stream to guide them in choosing appropriate size and type
of materials to use (S. Downie, CDFG, Redway, CA, pers. comm. 1994).

Evaluating effectiveness of habitat restoration has been neglected but is important in improving
restoration technology and demonstrating the benefits of restoration (Hall 1984; Koski 1992).

112 Restoration

The most meaningful evaluations simultaneously examine habitat, fish production, and cost
effectiveness (Everest et al. 1991). Habitat indicates whether the project attained the desired
changes; fish production indicates whether the habitat changes produced the desired effect on the
target species. Cost effectiveness shows whether the increased benefits were worth the expense.

Effectiveness monitoring can be conducted for both specific restoration practices and for the
cumulative effects of the set of practices applied in the watershed (MacDonald et al. 1991).
Monitoring of a specific practice, such as planting vegetation to prevent erosion, indicates
whether that practice was successful in a specific situation. Monitoring the entire set of practices
determines whether the cumulative effect of all the individual practices was successful in
attaining objectives. Effectiveness evaluations are not needed for every individual project, and
they are actually outlawed by some appropriation legislation (e.g., California's restoration
program), but evaluations should be done for a representative sample of projects across a range
of stream types, and for the overall program. Many worthwhile evaluations can be based on
professional judgement and unpublished observations of experienced professional practitioners.
However, formal scientific studies that fully evaluate projects are needed for at least
representative restoration projects.

Cost effectiveness should include the costs of planning, implementation, and maintenance, and
the benefits derived from the increased fish production and other attributes in the basin over the
longevity of the project. The benefit-cost ratio and present net worth of habitat improvement
projects can be assessed (Everest and Talh6lm 1982; Everest and Sedell 1984; Everest et al.
1987b). However, this is rarely done because of the time required to thoroughly evaluate
changes in habitat and fish production. Almost no literature exists that treats appropriate
economic monitoring for ecosystem restoration (Weigand 1994). Evaluating benefits is also
difficult because the widespread depressed level of fish populations causes underutilization of
improved habitats (J. Barnes, FS, Arcata, CA, pers. comm. 1994).

Evaluations of over 1,200 stream improvement projects in Oregon (Andrus 1991) resulted in
nine recommendations for increasing effectiveness: 1) allocate more funds for examining limiting
factors and for monitoring results; 2) consolidate smaller projects and treat long channel
segments in a few watersheds rather than dispersing projects among many watersheds; 3) exploit
the potential of beaver in'enhancing channel structure; 4) explore less costly methods for
improving channel structure; 5) prioritize funding for areas with important fisheries resources;
6) increase funding for demonstration projects on urban streams to promote public awareness;
7) reevaluate the practice of riprapping streambanks (consider the loss of fish habitat that occurs
when riprapping results in a channelized stream); 8) identify streams where high temperature
limits production and restore riparian vegetation before using instrearn structures; and 9)
continue to archive information on completed projects so that results can be evaluated.

RESTORATION PLANNING ON PRIVATE LANDS

Restoration opportunities on private lands depend on providing incentives and obtaining access.
Some of the best potential fish habitat is located in the low-elevation watersheds that have been
intensively managed for timber. To effect restoration on private lands, landowners must be
given incentives to conduct restoration or cooperate with federal and state efforts.

Restoration 113

Although restoration projects should generally undergo extensive planning and evaluation, these
types of reviews are most appropriate for programs on public lands or large projects. For
smaller projects on private lands, such rigorous reviews are not easily accomplished nor always
necessary. Addition of LWD during timber harvest to restore stream habitat complexity is an
example of a small project not requiring extensive planning and evaluation (C. Andrus, ODF,
Salem, OR, pers. comm. 1994). A representative sample of such projects, however, must be
evaluated for effectiveness, and the projects must address known limiting factors.

A streamlined permitting and design process is needed to take better advantage of restoration
opportunities on private lands (T. O'Dell and L. Diller, Simpson Timber Company, Korbel, CA,
pers. comm. 1994). Opportunities for adding LWD or boulders to debris-poor streams on
private lands often arise during timber harvest when equipment, materials, and labor are on site.
To help willing landowners to contribute and to ensure proper design of instrearn structures,
trained agency personnel should be available to help companies obtain permits and advise them
on placing instream structures so that the work can be done cost-effectively without delay.

CURRENT RESTORATION PROGRAMS

Many federal, state, and local agencies, tribes, and private interests are actively involved in
habitat restoration. On federal lands, the FS and BLM have large programs aimed at all aspects
of habitat restoration, including watershed, riparian, and instream restoration.

Restoration on National Forests

The restoration program on the Six Rivers National Forest in northern California provides an
example of ongoing restoration on federal lands (J. Barnes, FS, Arcata, CA, pers. comm. 1994).
Streams in this region were heavily impacted by landslides during a 100-year flood in 1964.
The stream channels have aggraded and become wide and shallow, lacking pools and habitat
complexity. Changed logging practices and road improvements have stabilized watershed
processes, allowing instream restoration to be effective. Funds for restoration come mainly from
the State of California restoration program through its Wildlife Conservation Board.

The primary restoration objective for streams on the Six Rivers National Forest is to provide
rearing habitat for juvenile steelhead by the addition of instrearn structures to narrow and deepen
stream channels, create pools and winter cover, and stabilize channels to allow reestablishment
of riparian vegetation (Fig. 9.4). Although steelhead are the target species, fall chinook and
cutthroat trout are also of concern (Fuller 1990; McCain 1992), as well as other aquatic non-fish
species (Fuller and Lind 1994). Although some early mistakes were made in boulder placement,
updated techniques effectively increase salmonid carrying capacity in treated reaches (J. Barnes,
FS, Arcata, CA, pers. comm. 1994).

Other funds for restoration on national forests can come from the Knutson-Vandenberg Act
(1930, amended 1976) which provides money out of timber sale receipts for improving
productivity of renewable resources on Forest lands. Use of these "K-W funds for restoration,
however, is hampered because they must be spent within the boundaries of the specific timber

114 Restoration

sale, which frequently does not
coincide with restoration needs.

California's Habitat
Restoration Program

California has an active program of
habitat restoration for private lands
administered by the CDFG. The
total budget for the restoration
program is about $5 million per year
Q. Steele, CDFG, Sacramento, CA,
pers. comin. 1994). Funding can
come from sales of commercial
fishing stamps, angling licenses,
permit fees, statewide initiatives, and
other sources. Private-sector
cooperation, involving matching
funds, in-kind contributions, and
volunteer efforts, is growing
strongly.

A cottage industry of restoration
in
companies has developed
California to conduct restoration 9
activities in cooperation with CDFG.
Proposals are submitted by the Figure 9.4. Boulders and logs used to form
companies and evaluated by CDFG, pools and provide complex cover in Redcap
and a restoration manual (Flosi and Creek, Six Rivers National Forest, California.
Reynolds 1991) gives specific The stream was heavily impacted by
direction for preliminary watershed landslides and resulting sedimentation and
assessments, habitat inventories,
p r o j e c t planning, a n d scour. Before instream structures were
implementation. Research on added,. the channel was stabilized by
effectiveness evaluations, however, riprapping to allow establishment of the dense
are prohibited by the legislation riparian vegetation shown in the photo.
establishing the program. Since (Photo courtesy of J. Barnes, FS.)
1981, nearly 3,000 restoration
projects administered by CDFG have
been completed to control erosion,
improve fish passage, stabilize stream banks, and improve instrearn habitat (Fig. 9.5).
California's restoration program also supports supplemental hatchery and educational programs.

Restoration 115

ifi

Figure 9.5. Distribution of nearly 3,000 restoration projects administered by California
Department of Fish and Game done to control erosion, improve fish passage, stabilize
stream banks, and improve instrearn habitat. (Figure courtesy of J. Hopelain,
California Department of Fish and Game.)

Oregon's Basal Area Credits

Oregon incorporated incentives for habitat restoration into its new forest practices rules that
provide credits for instrearn or other restoration projects conducted by landowners during timber
harvest (ODF 1994; ODF and ODFW 1995). Other aspects of Oregon's rules encourage active
silvicultural management of riparian areas to reestablish mature conifers, an important objective
of riparian restoration. The principal objective of the credit for instrearn restoration is to

116 Restoration

improve fish habitat in streams that lack LWD. The credits provide incentives for operators to
place logs in streams or take other actions to immediately improve fish habitat.

Subject to prior approval-of the State Forester, operators may place conifer logs or downed trees
in fish-bearing -streams and receive "basal area credit" toward meeting the requirement for
retaining live trees in a stream's riparian management area. Basal area is determined by
measuring cross-sectional area of the large end of the log or at the spot on a downed tree that
would be equivalent to breast height. For large and medium Type F streams, the credit is twice
the basal area of the placed log (i.e., for every log placed in'a stream, 'approximately two trees
can be harvested from the RMA). For small Type F streams, the credit is equal to the basal
area of the placed log. The basal area credit, however, can not reduce the standing tree
retention below "active management" targets specified in the rules. These active-man@gement
targets are usually about 75 % of the standard targets.

In placing logs, operators must follow prescriptions from the State Forester (ODF and ODFW
1995). Operators may also propose other enhancement projects for basal area credit, such as
creation of off-channel alcoves and fencing to exclude cattle. Such enhancement projects are
reviewed-by-ODFW, and the basal area credit is negotiated,.among the operator, ObF, and
ODFW.

The advantages of this program are that it costs much less than other , instrearn restoration
practices, and it encourages landowners to participate in restoration on private lands. Possible
disadvantages include the lack of careful pre-project evaluations and the current lack of research
on the program's effectiveness and possible unintentional adverse effects. "Credit trees" usually
do not include the rootwad, which makes them potentially less -stable and less effective than
natural LWD. "Credit trees" are placed without anchoring or cabling to stabilize the logs in
place, which could cause problems with channel instability. In addition, Oregon does not use
watershed analysis in evaluating timber harvest plans. Thus, the projects may not address the
actual factors limiting fish production. They target mainly coho salmon, and other species may
be adversely affected. The "credit-tree" projects are being monitored (C. Andrus, ODF, Salem,
OR, pers. comm. 1994), but results are not yet available. Another potential problem is that
future LWD recruitment could be reduced because an operator can take two trees from the RMA
for each one placed in the stream. Maintaining "active-management" targets for standing trees
in the RMA establishes a minimum level for basal area, which should prevent excessive harvest.

The benefits may outweigh the potential disadvantages. The "credit-tree" projects directly
address the lack of LWD in streams, which is widespread, well documented (e.g., Beschta
1991), and recognized as the most common limiting factor for coho salmon (Nickelson et al.
1992a). Although the logs are not anchored or cabled, movement alone should not be
considered a failure. In many respects, effective redistribution of wood- by the stream may be
ecologically more desirable than if it remained in an original fixed position (Gregory and
Wildman 1994).

Restoration 117

THE ROLE OF RESTORATION IN WATERSHED MANAGEMENT

Watershed restoration-is an integral part of a comprehensive program to recover anadromous fish
habitat that emphasizes habitat protection and uses restoration to stabilize deteriorating conditions
and accelerate recovery in key watersheds. Before initiating restoration, land uses that have
caused the degradation need to be modified to end adverse effects.

Considering the cost of restoration, habitat protection is obviously preferable to allowing habitat
to degrade to the point of needing restoration. Costs of restoration vary depending on limiting
factors, stream size, restoration methods, project objectives, and other factors. Projects that
used mostly instream structures cost an average $24,000 per stream km and ranged up to $1.2
million per stream km (House et al. 1989; Hunter 1991). Costs of stabilizing watershed slopes
and existing roads, obliter-ating abandoned roads, and replanting riparian areas are also high
(Koski 1992). Restoration practitioners are attempting to find less expensive ways to improve
habitat (Cederholm et al. 1988; Cederholm and Scarlett 1991; Flosi and Reynolds 1991), but
.costs remain high.

The need for restoration is great. For example, more than two-thirds of the riparian areas in
the Pacific Northwest are substantially degraded and need reforestation (Pacific Rivers Council
1993a), and 50-85% of all streams in the region need restoration (McMahon 1989). Cost of
upland, riparian, and instrearn restoration in key watersheds comprising about one-third of
federal lands in Oregon, Washington, and northern California would exceed $700 million
(Pacific Rivers Council 1993a). Political will and commitment will be indispensable to
formulating and implementing restoration that would be significant in light of this great need
(Wiegand 1994). '

Restoration costs money, but its return on investment can be considerable. The return of salmon
and recovery of watershed functions have many social and economic benefits, and the direct
restoration work would,generate many jobs (Pacific Rivers Council 1993a). Due to limited
funds and the large amount of degraded habitat, most habitat will have to rely on slow recovery
under effective watershed management. Much can be done ', however, to speed recovery in
priority areas of heavy hurftan use or severely depressed stocks. Major goals should be to
maintain existing wild stocks and promote. recovery of stocks listed under the Endangered
Species Act.

Priorities for restoration should be key watersheds that remain healthy, rather than the most
degraded areas (FEMAT 1993). The most urgent restoration task is protection of key
watersheds and riparian areas and immediate prevention of imminent road-related sedimentation
(Pacific Rivers Council 1993a). The goal should be to secure, expand, and link the healthier
areas in a system of refugia watersheds connected by intact migration corridors (Frissell et al.
1993). This approach would yield a quicker, more widespread, and more cost-effective
response.

Along with habitat restoration, small supplemental rearing programs may help to speed recovery
of wild stocks (e.g., Pacific Lumber Company 1993). Supplemental rearing programs operate
small hatcheries and utilize stocks native to the watershed being restored. Their purpose is to

118 Restoration

speed recovery of the native stock to take full advantage of the restored habitat. Such hatchery
programs can be discontinued once recovery is achieved, to minimize potential genetic effects
on the wild stock.

Restoration programs should also include an educational component to inform the public about
the value of watershed resources and the importance of habitat protection (Koski 1992).
Education offers the best possibility of increasing public awareness of environmental issues and
instilling a conservation ethic needed to ensure long-term sustainability of salmonid habitats.

Restoring habitat alone can not guarantee recovery of anadromous salmonids. Fisheries
management and hydropower also need to contribute (Palmisano et al. 1993; Botkin et al. 1994;
USDC 1995). Concurrent with habitat restoration, fisheries managem. ent needs to ensure
adequate spawner escapements to fully seed the restored habitat. Many depressed wild stocks
can not recover without coordinated fisheries management to curtail harvest (C. J. Cederholm,
WDNR, Olympia, WA, pers. comm. 1994). Operation of hydropower facilities on river main
stems needs to provide upstream passage for adult salmonids and adequate survival of
downstream migrant juveniles.

Habitat restoration is not a panacea for habitat recovery (Koski 1992). Habitat restoration and
protection, however, are critical because even with fisheries closures, depressed stocks can not
recover without habitat.

Chapter 10
Conclusions and Recommend ati-ons

A comprehensive watershed-level approach is essential for maintaining and restoring salmonid
habitat because the watershed is a fundamental unit for both ecological processes and land
management. The failure of past piecework forest management to prevent habitat degradation
or to accomplish restoration of stream reaches shows the need for an ecosystem-based,
watershed-level management strategy.

The main technical elements of the watershed approach are buffer zones, BMPs, watershed
analysis, and restoration. Because any conservation strategy will probably fail without
community support, watershed management also includes outreach programs to recruit support
from local citizens and enlist cooperation irom. private landowners.

BUFFER ZONES

Buffer zones are probably the most important tool for protecting critical riparian and aquatic
processes. Buffer zones along streams, however, can not maintain fish habitat unless sensitive
watershed areas and hydrologic processes are also protected by effective watershed management.

Buffer zones do not need to be "lock-out" zones if management activities within them maintain
or restore critical riparian processes. The appropriate design for buffer zones depends on
management objectives. The widest buffers with greatest restrictions on activities are used along
fish-bearing streams to meet the full range of objectives for fish habitat, as well as for other
wildlife (e.g., owls and amphibians). Narrower buffers with fewer restrictions can be used
along non-fish streams to protect water quality and downstream fish habitat.

To fully protect fish-bearing streams, buffers need to provide all processes that create and
maintain fish habitat, particularly shade, streambank integrity, and recruitment of large woody
debris. Buffer zones need to be wide enough to fully protect the stream and floodplain and to
ensure the long-term viability, of the buffer itself. Buffers wider than one site-potential tree
height (average maximum height given site conditions) may be needed to protect the floodplain
and riparian vegetation where exposure to light and wind could cause succession to shrub
communities. Blowdown in buffer zones, however, is usually not a problem for fish habitat,
and where it does cause a problem, such as a stream blockage, it can be minimally altered to
restore fish passage while leaving most fallen trees in place.

120 Conclusions and Recommendations

Current requirements for buffer width and leave trees on private lands do not fully protect LWD
sources for fish-bearing streams. Four of the five states require leaving only an estimated 23 %
to 58% of potential LWD sources compared to the sources present in mature conifer forest.
More and larger leave trees are needed to provide optimal fish habitat over the long term.

Many areas in the Pacific Northwest, however, have degraded riparian vegetation dominated by
hardwood and shrubs, and lack additional large conifers for leave trees. In these degraded
areas ', buffer zones can be actively managed to improve degraded riparian functions.
Reestablishing conifers offers potential long-term benefits for both fisheries and timber
managers. In riparian areas restored to mature conifers, buffers could be selectively harvested
if monitoring shows it would not harm fish habitat.

Buffer zones are also needed along non-fish streams to protect water quality and provide LWD
for downstream fish habitat. Except for federal lands under NFP and PACFISH, buffers on
small non-fish streams (both perennial and intermittent) are often inadequate or lacking.
Reliance on BMPs alone may be inadequate to protect these headwater areas, and monitoring
studies have not yet shown that BMPs are effective in preventing downstream impacts. The
width and harvest activities within these buffers can be designed specifically to protect headwater
sources of temperature control, sediment, and woody debris.

Management regimes are needed that will put timber harvest in the context of natural disturbance
regimes. Disturbance to streams and flood plains is not necessarily negative, and may be needed
for productive fish habitat over the long term. Unnatural disturbances, however, should be
minimized, and patterns of land use should mimic the natural disturbance process and leave the
necessary legacy for the long-term development of required habitat. Specifically, more large
wood is needed in buffers along headwater channels with the greatest potential for delivery to
fish-bearing streams.

BEST MANAGEMENT PRACTICES

Generally, BMPs can be effective at controlling nonpoint source pollution but need to be closely
monitored for implementation and effectiveness to identify needed improvements. All the Pacific
Northwest states and Alaska have a regulatory BMP program and monitor for implementation
and effectiveness. Monitoring programs are mostly new, however, and BMPs have not been
fully evaluated.

Many current forestry-related problems with water quality result from inadequate BMP
implementation, which is generally worse on small private parcels than on public or large
industrial holdings. On-site inspections are needed to identify sensitive areas and to design
harvest and transportation plans to.suit local conditions. Having well-qualified field personnel
available to provide site-specific BMP recommendations, particularly for small private
landowners, is probably the best way to improve BMP implementation.

Because small non-fish streams are particularly important for preventing sediment pollution and
because buffer zones along them are usually narrow and heavily harvested, BMPs for activities

Conclusions and Recommendations 121

near them need to be closely monitored to ensure that they are effective. State BMPs do not
fully protect small non-fish streams, intermittent channels, and unstable slopes from logging
disturbance. The BMPs pertaining to felling and yarding that apply to fish-bearing streams
generally do not apply to small non-fish streams (particularly intermittent channels) and unstable
slopes. Monitoring with feedback for adaptive management is needed to develop, evaluate, and
improve BMPs for these areas.

WATERSHED ANALYSIS

A watershed program must have some process for analysis and planning at the watershed level.
Watershed analysis is the most thorough method for understanding potential effects of land uses
at the watershed scale. Watershed analysis can be used to describe current conditions, identify
sensitive areas and risks, determine factors limiting salmonid production, and develop
prescriptions to prevent cumulative effects.

Watershed analysis should be instituted wherever possible to provide information for watershed
planning. State agencies can organize and lead working groups of concerned landowners in
cooperative watershed analysis in watersheds with mixed ownerships. The watershed analysis
efforts in Washington and Idaho provide good prototype models for developing prescriptive
watershed analysis for private lands.

RESTORATION

Restoration is an integral part of comprehensive watershed management and is used to stabilize
deteriorating conditions and speed recovery in key watersheds. Effective restoration has a
watershed-level approach and includes upland, riparian, and instrearn components. The upland
component is used to control erosion, stabilize roads, upgrade culverts for fish passage, and
manage watershed uses. The riparian component restores functions of riparian vegetation by
reestablishing mature conifers or other appropriate vegetation. The instrearn component, using
woody debris and other structures to retain spawning gravel and create pools or other features,
should be conducted only after watershed problems have been addressed and limiting factors
identified.

Effectiveness evaluations are a critical part of restoration because they help improve technology
and demonstrate the benefits of restoration. A representative sample of projects needs to be
evaluated over a range of watershed and stream classes for each type of restoration technique.

Although restoration projects should undergo rigorous planning and evaluation, a streamlined
process is needed to take advantage of opportunities arising during timber harvest on private
lands when equipment, materials, and labor are on site. Trained agency personnel are needed
to advise willing companies on obtaining permits and designing projects so that the work can be
done without delay. Monitoring can be used to develop and evaluate standard techniques for
such cases, and incentives can be incorporated in forest practices rules to encourage such

122 Conclusions and Recommendations

projects. Prototype models for this are the Oregon incentives for riparian and instream
restoration.

Priorities for restoration are key watersheds with the best remaining habitat, rather than the most
degraded areas. The goal is to secure, expand, and link key watersheds in a system of refugia
connected by iritact migration corridors. Restoration activities for the best watersheds should
focus on reducing risks to these habitats by obliterating unneeded roads and revegetating upland
and riparian areas. The expectation is that all watersheds, not just key watersheds, will improve
over time, but key watersheds will recover fastest because of their high level of habitat
protection and priority for restoration. Other watersheds are expected to recover as a result of
improved land management.

The best forin of restoration is habitat protection. There is no guarantee that restoration efforts
will succeed, and the cost of restoration is much greater than the cost of habitat protection. The
most prudent approach is to minimize the risk to habitat by ensuring adequate habitat protection.

COMMUNITY OUTREACH

Comprehensive watershed management involves more than improved scientific understanding;
it also encompasses economic, social, and political concerns. In the ideal situation, all
stakeholders, including landowners, industries, and citizen groups, are partners in planning and
implementing watershed management. Working groups of government agencies, industry, and
citizen groups can provide the necessary consensus on forest practices and watershed
management issues.

Habitat protection and restoration on a watershed basis will require integrating federal land
management with other regulatory programs that affect aquatic habitats, particularly the Clean
Water Act and Endangered Species Act. Habitat Conservation Plans developed under the ESA
have an important role in watershed planning on private lands. Ultimately, basin-wide planning
efforts are needed that include all public and private land managers.

Economic incentives can be provided for local communities and landowners to support habitat
protection and restoration. On public lands, contracts awarded by competitive bidding can
provide effective habitat protection and restoration while providing local employment. Tax
credits and cost-sharing programs can be expanded to compensate private landowners for
measures taken to protect public aquatic resources, such as.expanded buffer zones or retention
of additional leave trees along streams.

Although scientific information will always be incomplete and possibly wrong, current
knowledge is adequate to design comprehensive watershed management to reduce risks to
salmonid habitat and to restore degraded habitat. Scientific information can provide the basis
for evaluating trade-offs between timber harvest and habitat protection, but whether society
should take actions needed to recover anadromous- salmonids is a political decision.

Glossary

Alevin: Larval salmonid that has hatched but has not fully absorbed its yolk sac, and generally
has not yet emerged from the spawning gravel.

Anadromous salmonids: Members,of the family Salmonidae (especially salmon, trout, and char)
that move from the sea to fresh water for reproduction.

Basal area: The cross-sectional area of a log or tree measured at breast height.

Bedload sediment: That part of a stream's total sediment load moved along the bottom by
running water, in contrast to suspended sediment which is carried in the water column.

Beneficial use: The designated resource value of a stream, such as domestic water supply or
anadromous fish habitat.

Best Management Practices (BMPs): Methods, measures, or practices designed to prevent or
reduce water pollution.

Blowdown (also windthrow): The uprooting and felling of trees by strong gusts of wind.

Buffer zone: An administratively defined area established along a stream, lake, wetland, or
erosion hazard to provide protection for aquatic resources during land-use activities.

Carrying capacity: Maximum average number of organisms that can be sustained in a habitat.

Clearcutting: Removal of the entire standing crop of trees from an area; in practice, much
unsalable material may be left standing.

Coarse sediment: Sediment with particle sizes generally greater than 2 nun, including gravel,
cobbles, and boulders.

Compliance monitoring: Sampling of stream water,to determine whether properly implemented
Best Management Practices meet applicable water quality standards.

Culvert: Buried pipe structure that allows streamflow or road drainage to pass under a road.

124 Glossary

Cut and fill: Construction of a road on hilly terrain that is partly excavated and partly filled.

Cumulative effects: Effects that result incrementally and collectively from the combined effects
of separate management activities through time and space.

Debris torrent: Deluge of water charged with soil, rock, and woody debris down a steep stream
channel.

-Density: Number of organisms per unit area or volume.

Dewatering: Lowering of the water table in stream channel deposits caused by a channel shift,
flow reduction, or channel downcutting.

Diversity index: Numerical value derived from the number of individuals per taxon and the
number of taxa present.

Ecosystem management: Management of watershed land and aquatic resources based on
perspective of forest and stream ecosystem structure, function, and dynamics aimed at long-term
sustainability of watershed productivity. Ecosystem management integrates scientific knowledge
of ecological relationships within a complex sociopolitical and values framework toward the
general goal of protecting native ecosystem integrity over the long term. Although managing
an entire ecosystem can positively affect a listed species, incorporating an ecosystem approach
into recovery efforts means protecting the processes and functions of ecosystems important for
the conservation of listed, proposed, or candidate species (Grumbine 1994).

Embeddedness: Degree to which coarse sediment (boulders, rubble, gravel) are surrounded or
covered by fine sediment, usually measured in classes according to percent coverage.

Effectiveness monitoring: Sampling of soil erosion, streams, and other features to determine
whether properly implemented Best Management Practices are effeciive in meeting their intent.

Emergence: Departure of fry from the incubation gravel into the water column.

Escapement: That portion of an anadromous fish population that escapes fisheries and reaches
the freshwater spawning grounds.

Evapotranspiration: Loss of water by evaporation from the soil and transpiration from plants.

Fine sediment: Sediment with particle size of 2 mm or less, including sand, silt, and clay.

Flood plain: Level lowland bordering a stream onto which the stream spreads at flood stage.

Forest practices: The full range of forest management activities employed in silviculture and
harvest of timber.

Glossary 125

Freshet: Rapid temporary rise in stream discharge caused by heavy rains or rapid melting of
snow or ice.

Fry: Life stage of a salmonid between full absorption of the yolk sac and the fingerling or parr
stage, which generally is reached by the end of the first summer.

Gradient (topographic slope): Average change in vertical elevation per unit of horizontal
distance. %

Groundwater: That part of the subsurface water that is in the zone of saturation, including
underground streams.

Gullying: Formation or extension of gullies by surface runoff water.

High-lead yarding: Method of powered cable logging in which the mainline blocks are fastened
high on a spar so logs can be skidded with one end off the ground.

Landing: Place where felled trees are accumulated for further transport.

Large woody debris (LWD): Any piece of woody material that intrudes into a stream channel,
whose smallest diameter is greater than 10 cm, and whose length is greater than I m.

Limiting factor: Environmental factor that limits the growth or activities of an organism or that
restricts the size of a population or its geographical range.

Implementation monitoring: Sampling of management activities to determine whether practices
are adequately applied as specified.

Instrearn restoration: Activities conducted to improve physical structure of stream channels, such
as to provide spawning habitat or create pools.

Intermittent channel: A stream channel that carries water only part of the year during snowmelt
or after rain storms.

Main stem: Principal stream or channel of a drainage system.

Mass movement: Downslope transport of soil and rocks due to gravitational stress.

Monitoring: The process of collecting information to evaluate whether anticipated or assumed
results of a management plan are being realized or whether implementation is proceeding as
planned.

Nephelometric turbidity unit (NTU): Measure of the concentration or size of suspended particles
(cloudiness) based on the scattering of light transmitted or reflected by the medium.

126 Glossary

Nonpoint-source pollution: Pollution from sources that cannot be defined as discrete points, such
as areas of timber harvesting, surface mining, and construction.

Old growth: Forest stand dominated by large old trees reaching natural senescence; the last stage
in forest succession. Characters of old-growth forest include 1) storied canopy including
different tree species in the lower levels; 2) openings that allow light into the forest floor where
dense vegetation thrives; 3) presence of snags and downed logs; and absence of major stand-
altering disturbance by humans (Bolsinger and Waddell 1993).

Parr: Young salmonid in the stage between alevin and smolt, which has developed distinctive
dark marks on its sides and is actively feeding in fresh water.

Perennial stream: A stream with flowing water all year long.

Permeability: A measure of the rate at which a substrate can pass water, the rate depending on
substrate composition and compaction.

Pool: Portion of a stream with reduced current velocity, often with deeper water than
surrounding areas and with a smooth surface.

Presmolt: Juvenile salmonid during the parr-smolt transformation, with intermediate coloration
and body form.

Primary production: Production of organic substances by photosynthesis.

Redd: Nest made in gravel, consisting of a depression dug by a fish for egg deposition and then
filled.

Riffle: Shallow section of a stream or river with rapid current and a surface broken by gravel,
rubble, or boulders.

Riparian restoration: Management activities aimed at changing the size, density, species
composition, or other characteristics of riparian vegetation to improve ecosystem functions.

Riparian area: Area between a stream or other body of water and the adjacent uplands.

Riprap: Layer of large, durable materials (usually boulders), used to protect a stream bank or
lake shore from erosion.

Runoff: The part of precipitation and snowmelt that reaches streams by flowing over the ground.

Second growth: Forest stand that has come up after some drastic interference such as logging,
fire, or insect attack.

Sediment: Fragments of rock, soil, and organic material transported and deposited in beds by
wind, water, or other natural phenomena.

Glossa[y 127

Seral stage: One in a series of ecological communities that succeed one another in the biotic
development of an area (also see succession). Forests pass through four recognized stages: 1)
early seral stage, the period from disturbance to crown closure; 2) mid-seral stage, from crown
closure to first merchantability (usually age 15-40 years); 3) late-seral stage, from first
merchantability to culmination of mean annual increment (100 years); and 4) mature seral stage,
from culmination of mean annual increment to old-growth stage (200 years).

Shelterwood cutting: Selective cutting of regenerating plants so as to establish a new tree crop
under the protective remnants of a former stand.

Skid trail: A constructed trail or established path used by tractors or other vehicles for skidding
logs in going to and from landings.

Site-potential tree: A tree that has attained the average maximum height possible given site
conditions where it occurs.

Site class: A measure of an area's relative capacity for producing timber or other vegetation.

Site index: A measure of forest productivity expressed as the height of the tallest trees'in a stand
at an index age.

Skidding: Yarding logs by sliding or dragging with tractors or other ground-based equipment.

Skid trails: Trails on which logs are moved to landings by sliding or dragging.

Skyline yarding: Method of powered cable logging in which a heavy cable (the skyline) is
stretched between two spars and used as an overhead track for a load-carrying trolley.

Slash: Woody residue left after trees are felled, limbed, and yarded.

Smolt: The seaward-migrant stage of an anadromous salmonid that has undergone physiological
changes to cope with the marine environment.

Splash dam: Dam built to create a head of water for driving logs.

Stock: Group of fish that is genetically self-sustaining and isolated geographically or temporally
during reproduction.

Stream order: A number ranked from headwaters to river mouth that designates the relative
position of a stream in a drainage basin. First-order streams have no discrete tributaries; the
junction of two first-order streams forms a second-order stream; the junction of two second-
order streams forms a third-order stream; etc.

Succession: A series of dynamic changes by which one group of organisms succeeds another
through stages leading to potential natural community or climax. An 'example is the
development of series of plant communities (called seral stages) following a major disturbance.

128 Glossary

Tributary: Stream flowing into a lake or larger stream.

Upwelling: The movement of groundwater through stream substrate into the stream water
column.

Waterbar: Shallow channel (cross drain) or raised barrier of packed earth laid diagonally across
the surface of a road to guide water off the road. Also called "waterbreaks."

Watershed: Total land area draining to any point in a stream.

Watershed analysis: A systematic process, to describe current watershed conditions and develop
prescriptions to prevent cumulative impacts.

Watershed restoration: A broad-based program to control erosion from roads and bare soils,
upgrade or remove culverts to restore fish access, restore natural streamflow regimes, and
manage all uses of the stream and watershed.

Yarding: hauling of timber from the point of felling to a yard or landing.

Acronyms

ADFG Alaska Department of Fish and Game

ADNR Alaska Department of Natural Resources

AWGCFFR Alaska Working Group on Cooperative Forestry Fisheries Research

BLM Bureau of Land Management

BMPs Best Management Practices

CDF California Department of Forestry and Fire Protection

CDFG California Department of Fish and Game

cfs Cubic feet per second (a measure of stream discharge)

dbh Diameter at breast height of trees

DO Dissolved oxygen

ECA Equivalent Clearcut Area

EIS Environmental Impact Statement

ESA Endangered Species Act

ESU Evolutionarily Significant Unit

FEMAT Forest Ecosystem Management Assessment Team

FS U.S. Forest Service

HCP Habitat Conservation Plan

130 Acronyms

IDL Idaho Department of Lands

LWD Large woody debris

NEPA National Environmental Policy Act

NFP Northwest Forest Plan

NMFS National Marine Fisheries Service

ODF Oregon Department of Forestry

ODFW Oregon Department of Fish and Wildlife

PACFISH Pacific Anadromous Fish Habitat Management Strategy

RHCA Riparian Habitat Conservation Area

RMAs Riparian Management Areas: buffer zones in Oregon

RMZs Riparian Management Zones: buffer zones in Washington

SpZS Stream Protection Zones; buffer zones in Idaho

TFW Washington State's Timber Fish Wildlife Agreement

TLMP Tongass Land Management Plan

TTRA Tongass Timber Reform Act of 1990

USDI U.S. Department of Interior

USDA U.S. Department of Agriculture

USDC U.S. Department of Commerce

WDNR Washington Department of Natural Resources

WLPZs Watercourse and Lake Protection Zones: buffer zones in California

References

ADNR. 1993. State of Alaska forest resources and practices regulations. Alaska Department
of Natural Resources, Division of Forestry, Juneau, AK.

Almas, D., J. Colla, K. David, and C. Schnepf. 1993. Forestry BMP's for Idaho. Idaho
Department of Lands, Coeur d'Alene, ID.

Andrus, C. W. 1991. Improving streams and watersheds in Oregon: Inventory and evaluation
of efforts to improve the condition of Oregon's stream and watersheds from 1985 to 1990.
Oregon Water Resources Department, Salem, OR.

Andrus, C. W., B. A. Long, and H. A. Froehlich. 1988. Woody debris and its contribution
to pool formation in a coastal stream 50 years after logging. Canadian Journal of Fisheries and
Aquatic Sciences 45:2080-2086.

Amo, S. F, and R. D.Ottmar. 1994. Reducing hazard for catastrophic fire. Pages 18-19 in
Everett, R. L. (compiler) Restoration of stressed sites and processes. General Technical Report
PNW-GTR-330. USDA Forest Service, Pacific Northwest Research Station, Portland, OR.

AWGCFFR. 1991. Alaska Working Group on Cooperative Forestry Fisheries Research:
mission and organization. Informational document 91-01. USDA Forest Service, Juneau, AK.

Bach, L. B. 1993. Risk assessment procedure. In M. D. Purser (editor) Proceedings,
Watershed and stream restoration workshop: shared responsibilities for shared watershed
resources, conducted August 26-27, 1993, Portland OR, not paginated. American Fisheries
Society, Oregon Chapter, Portland.

Bachman, R. A. 1984. Foraging behavior of free-ranging wild and hatchery brown trout in a
stream. Transactions of the American Fisheries Society 113:1-32.

Belt, G. H. 1980. Predicting strearnflow changes caused by forest practices using the
equivalent clearcut area model. College of Forestry, Wildlife, and Range Sciences. Bulletin
32. University of Idaho; Forest, Wildlife, and Range Experiment Station; Moscow, ID.

132 References

Belt, G. H., J. O'Laughlin, and T. Merrill. 1992. Design of forest riparian buffer strips for
the protection of water quality: analysis of scientific literature. Report No. 8. University of
Idaho, Idaho Forest, Wildlife and Range Policy Analysis Group, Moscow, ID.

Benda, L. 1993. Watershed analysis in the context of fish habitat management. In M. D.
Purser (editor) Proceedings, Watershed and stream restoration workshop: shared responsibilities
for shared watershed resources, conducted August 26-27, 1993, Portland OR, not paginated.
American Fisheries Society, Oregon Chapter, Portland.

Berg, L., and T. G. Northcote. 1985. Changes in territorial, gill-flaring, and feeding behavior
in juvenile coho salmon (Oncorhynchus kisutch) following short-term pulses of suspended
sediment. Canadian Journal of Fisheries and Aquatic Sciences 42:1410-1417.

Berkman, H. E., and C. F. Rabini. 1987. Effect of siltation on stream fish communities.
Environmental Biology of Fishes 18:285-294.

Beschta, R. L. 1991. Stream habitat management for fish in the northwestern United States:
The role of riparian vegetation. American Fisheries Society Symposium 10:53-58.

Beschta, R. L., and W. L. Jackson. 1979. The intrusion of fine sediments into a stable gravel
bed. Journal of the Fisheries Research Board of Canada 36:204-210.

Beschta, R. L., R. E. Bilby, G. W. Brown, L. B. Holtby, and T. D. Hofstra. 1987. Stream
temperature and aquatic habitat: fisheries and forestry interactions. Pages 191-232 in E. 0. Salo
and T. W. Cundy (editors) Streamside management: forestry and fishery interactions.
Contribution No. 57, Institute of Forest Resources, University of, Washington, Seattle.

Bilby, R. E. 1984a. Characteristics and frequency of cool water areas in a western Washington
stream. Journal of Freshwater Ecology 2:593-602.

1984b. Removal of woody debris may affect stream channel stability- Journal of
Forestry 82:609-613.

Bilby, R. E., and P. A. Bisson. 1991. Enhancing fisheries resources through active
management of riparian areas. Pages 201-209 in B. White and 1. Guthrie (editors) Proceedings
of the 15th Northeast Pacific pink and chum salmon workshop. Pacific Salmon Commission,
Vancouver, B.C.

Bilby, R. E., and J. W. Ward. 1989. Changes in characteristics and function of woody debris
with increasing size of streams in western Washington. Transactions of the American Fisheries
Society 118:368-379.

Bilby, R. E., and L. J. Wasserman. 1989. Forest practices and riparian management in
Washington State: data based regulation development. Pages 87-94 in R. E. Gresswell (editor)
Riparian Resource Management. U.S. Bureau of Land Management, Billings, MT.

References 133

Bisson, P. A., and R. E. Bilby. 1982. Avoidance of suspended sediment by juvenile coho
salmon. North American Journal of Fisheries Management 2:371-374.

Bisson, P. A., and J. R. Sedell. 1984. Salmonid populations in streams in clearcut vs. old-
growth forests of western Washington. Pages 121-129 in W. R. Meehan, I R. Merrell, Jr.,
and T. A. Hanley (editors) Proceedings, fish and wildlife relationships in old-growth forests
symposium. American Institute of Fishery Research Biologists, Asheville, NC.

Bisson, P., A., R. E. Bilby, m. a Bryant, C. A. Dolloff, G. B. Grette, R. A. House, M. L.
Murphy, K V. Koski, and J. R. Sedell. 1987. . Large woody debris in forested streams in the
Pacific Northwest: past, present, and future. Pages 143-191 in E. 0. Salo and T. W. Cundy
(editors) Streamside management: forestry and fishery interactions. Contribution No. 57,
University of Washington, Institute of Forest Resources, Seattle.

Bisson, R A., T. R Quinn, G. H. Reeves, and S. V. Gregory., 1992. Best management
practices, cumulative effects, and long-term trends in fish abundance in Pacific Northwest river
systems. Pages 189-232 in R. J. Naiman (editor) Watershed management: balancing
sustainability and environmental change. Springer-Verlag, New York.

Bjornn, T. C., and D. W. Reiser. 1991. Habitat requirements of salmonids in streams. Pages
83-138 in W. R. Meehan (editor) Influences of forest and rangeland management on salmonid
fishes and their habitats. Special Publication 19. American Fisheries Society, Bethesda, MD.

Bjornn, T. C., M. A. Brusven, N. J. Hetrick, R. M. Keith, and W. R. Meehan. 1992. Effects
of canopy alterations in second-growth forest riparian zones on bioenergetic processes and
responses of juvenile salmonids to cover in small, southeast Alaska streams. Technical Report
92-7. Idaho Cooperative Fish and Wildlife Research Unit, Moscow, ID.

Bolsinger, C. L., and K. L. Waddell. 1993. Area of old-growth forests in California, Oregon,
and Washington. General Technical Report PNW-RB-197. USDA Forest Service, Pacific
Northwest Research Station, Portland, OR.

Botkin, D., K. Cummins, T. Dunne, H. Regier, M. Sobel, and L. Talbot. 1994. Status and
future of salmon of western Oregon and northern California: findings and options. Report #8.
The Center for the Study of the Environment, Santa Barbara, CA.

Boughton, J. 1993. How to accomplish soil and watershed stewardship objectives through
integrated prescriptions. Pages 8-12 In: Brock, T. (editor) Proceedings of Watershed '91: A
conf@rence on the stewardship of soil, air, and water resources. Juneau, Alaska, April 16-17,
1991.' USDA Forest Service, Juneau, AK.

Boyette, W. G. 1993. Composite of reports on implementation of silvicultural nonpoint source
programs. North Carolina Division of Forest Resources, Raleigh, NC.

134 References

Bradley, W. P. 1988. Riparian management practices on Indian lands. Pages 201-206 in K.
J. Raedeke (editor) Streamside management: Riparian wildlife and forestry interactions.
Contribution No. 59. University of Washington, Institute of Forest Resources, Seattle, WA.

Brazier, J. R., and G. W. Brown. 1973. Buffer strips for stream temperature control.
Research Paper 15. USDA Forest Service, Pacific Northwest Research Station, Portland, OR.

Brown, T. C., and D. Binkley. 1994. Effect of management on water quality in North
American forests. General Technical Report RM-248. USDA Forest Service, Rocky Mountain
Forest and Range Experiment Station, Fort Collins, CO.

Brownlee, K. 1990. Food habits of juvenile salmonids in nursery areas on the lower Taku
River, Southeast Alaska. MS Thesis. University of Alaska, Fairbanks.

Brouha, P. 1991. Fish habitat planning. Pages 587-598 in W. R. Meehan (editor) Influences
of forest and rangeland management on salmonid fishes and their habitats. Special Publication
19. American Fisheries Society, Bethesda, MD.

Bryant, M. D. 1980. Evolution of large, organic debris after timber harvest: Maybeso Creek,
1949 to 1978. General Technical Report PNW-101. USDA Forest Service, Pacific Northwest
Forest and Range Experiment Station, Portland, OR.

1984. The role of beaver dams as coho salmon habitat in southeast Alaska streams.
Pages 183-192 in J. M. Walton and D. B. Houston (editors) Proceedings, Olympic wild fish
conference. Peninsula College, Fisheries Technology Program, Port Angeles, WA.

1985. Changes 30 years after logging in large woody debris, and its use by salmonids.
Pages 329-334 in R. R. Johnson, C. D. Ziebell, D. R. Patton, P. E Folliott, and R. H. Hamre
(editors) Riparian ecosystems and their management: Reconciling conflicting uses. First North
American Riparian Conference, Tucson, Arizona. General Technical Report RM-120. USDA
Forest Service, Rocky Mountain Forest and Range Experiment Station, Fort Collins, CO.

Burger, C. V., R. L. Wilmot, and D. B. Wangaard. 1985. Comparison of spawning areas and
times for two runs of chinook salmon (Oncorhynchus tshauytscha) in the Kenai River, Alaska.
Canadian Journal of Fisheries and Aquatic Sciences 42:693-700.

Bums, a c. 1991. Cumulative effects of small modifications to habitat. Fisheries 16:12-17.

Burroughs, E. R., Jr., and B. R. Thomas. 1977. Declining root strength in Douglas-fir after
felling as a factor in slope stability. Research Paper INT-190. USDA-Forest Service,
Intermountain Forest and Range Experiment Station, Ogden, UT.

California Board of Forestry. 1993. Assessing the effectiveness of California's forest practice
rules in protecting water quality. California State Board of Forestry, Sacramento, CA.

References 135

CDR 1994. , California Forest Practice Rules, Title 14, California Code of Regulations.
California Department of Forestry and Fire Protection, Sacramento, CA.

CDFG. 1994. Petition to the Board of Forestry to list coho salmon (Oncorhynchus kisutch) as
a sensitive species. California Department of Fish and Game, Sacramento, CA.

Cederholm, C. J. 1994. A suggested landscape approach for salmon and wildlife habitat
protection in western Washington riparian ecosystems. Pages 78-90 in A. B. Carey and C.
Elliott (compilers) Washington Forest Landscape Management Project-Progress Report Number
1. Washington State Department of Natural Resources, Olympia, WA.

Cederholm, C. J., and N. P. Peterson. 1985. The retention of coho salmon (Oncorhynchus
kisutch) carcasses by organic debris in small streams. Canadian Journal of Fisheries and Aquatic
Sciences 42:1222-1225.

Cederholm, C. J., and L. M. Reid, 1987. Impacts of forest management on coho salmon
(Oncorhynchus kisutch) populations of the Clearwater River, Washington: a project summary.
Pages 373-398 in E. 0. Salo and T. W. Cundy (editors) Strearnside management: forestry and
fishery interactions. Contribution No. 57, University of Washington Institute of Forest
Resources, Seattle.

Cederholm, C. J., and W. J. Scarlett. 1991. The beaded channel: a low-cost technique for
enhancing winter habitat of coho salmon. American Fisheries Society Symposium 10: 104-108.

Cederholm, C. J., W. J. Scarlett, and N. P. Peterson. 1988. Low-cost enhancement technique
for winter habitat of juvenile coho salmon. North American Journal of Fisheries Management
8:438-441.

Chamberlin, T. W., R. D. Harr, and F. H. Everest. 1991. Timber harvesting, silviculture,
and watershed processes. Pages 181-205 in W. R. Meehan (editor) Influences of forest and
rangeland management on salmonid fishes and their habitats. Special Publication 19. American
Fisheries Society, Bethesda, MD.

Chan, S. 1993. Regeneration of coastal riparian areas: second year tree performance. Coastal
Oregon Productivity Enhancement Program 6:4-6.

Chapman, D. W. 1988. Critical review of variables used to define effects of fines in redds of
large salmonids. Transactions of the American Fisheries Society 117:1-21.

Chen, G. K. 1992. Use of basin survey data in habitat modeling and cumulative watershed
effects analysis. FHR Currents, Fish Habitat Relationships Technical Bulletin No. 8. USDA
Forest Service, Pacific Southwest Region, Eureka, CA.

Clark, R. N., and D. R. Gibbons. 1991. Recreation. Pages 459-482 in W. R. Meehan (editor)
Influences of forest and rangeland management on salmonid fishes and their habitats. Special
Publication 19. American Fisheries Society, Bethesda, MD.

136 Rgferences

Cobourn, J. 1989. An application of cumulative watershed effects (CWE) analysis on the
Eldorado National Forest in California. Pages 449-460 in W. W. Woessner and D. F. Potts
(editors) Proceedings of the symposium on headwaters hydrology. American Water Resources
Association, Bethesda, MD.

Crispin, V., R. House, and D. Roberts. -1, 1993. Changes in instream habitat, large woody
debris, and salmon habitat after the restructuring of a coastal Oregon stream. North American
Journal of Fisheries Management 13:96-102.

Crone, R. A. I and C. E. Bond. 1976. Life history of coho salmon, Oncorhynchus kisutch, in
Sashin Creek, southeastern Alaska. Fishery Bulletin 74:897-923.

Daniels, S. E., G. B. Walker, J. R. Boeder, and J. E. Means. 1994. Managing ecosystems
and social conflict. Pages 327-339 in M. E. Jensen and P. S. Bourgeron (editors) Eastside
forest ecosystem health assessment. Volume 11: Ecosystem management: principles and
applications. General Technical Report PNW-GTR-318. USDA Forest Service, Pacific
Northwest Research Station, Portland, OR.

Daterman, G. E. 1994. Protecting unique habitats and riparian areas from insect attack. Pages
43-46 in Everett, R. L. (compiler) Restoration of stressed sites and processes. General
Technical Report PNW-GTR-330. USDA Forest Service, Pacific Northwest Research Station,
Portland, OR.

DeWalle, D. R. 1983. Wind damage around clearcuts in the ridge and valley province of
Pennsylvania. Journal of Forestry 81:158-172.

Dill, L. M., R. C. Ydenberg, and A. H. G. Fraser. 1981. ' Food abundance and territory size
in juvenile coho salmon (Oncorhynchus kisutch). Canadian Journal of Zoology 59:1801-1809.

Dissmeyer, G. E. In press. Evaluating the effectiveness of forestry best management practices
in meeting water quality goals or standards. USDA Forest Service, Atlanta, GA.

Dolloff, C. A. 1986. Effects of stream cleaning on juvenile coho salmon and Dolly Varden
in Southeast Alaska. Transactions of the American Fisheries Society 115:743-755.

Dolloff, C. A. 9 D. G. Hankin, and G. H. Reeves. 1993. Basinwide estimation of habitat and
fish populations in streams. General Technical Report SE-83. USDA Forest Service,
Southeastern Forest Experiment Station, Asheville, NC.

Doppelt, B., M. Scurlock, C. Frissell, J. Karr, and Pacific Rivers Council. 1993. Entering the
watershed: a new approach to save America's river ecosystems. Island Press, Washington, D.C.

Duncan, S. H., R. E. Bilby, J. W. Ward, and J. T. Heffner. 1987. Transport of road surface
sediment through ephemeral stream channels. Water Resources Bulletin 23:113-119.

References
137

Duncan, W. F. A., and M. A. Brusven. 1�85. Energy dynamics of three low-order southeast
Alaska streams: Allochthonous processes. Journal of Freshwater Ecology 3:233-248.

Eastside Forests Scientific Society Panel. 1993. Inter protection for late-successio I f s
im na ore ts
fisheries, and watersheds, national forests east of the Cascade Crest, Oregon and Washington.
A Report to the United States Congress and the President. Available from the American
Fisheries Society, Bethesda, MD.

Eiler, J. H., B. D. Nelson, a,nd R. E Bradshaw. 1992. Riverine spawning by sockeye salmon
in the Taku River, Alaska and British Columbia. Transactions of the American Fisheries Society
121:701-708.

Elliott, J. M. 1973. The food of brown and rainbow trout (Salmo trutta and S. gairdneti) in
relation to the abundance of drifting invertebrates in a mountain stream. Oecologia 12:329-347.

Emmingham, W. H., M. Bondi, and D. E. Hibbs. 1989. Underplanting western hemlock in
an alder thinning: early survival, growth, and damage. New Forests 3:31-43.

Everest, F. H., and W. R. Meehan. 1981. Forest management and anadrornous fish habitat
productivity. Pages 521-530 in Transactions of the 46th North American Wildlife and Natural
Resources Conference. Wildlife Management Institute, Washington DC.

Everest, R H., R. L. Beschta, J. C. Scrivener, K V. Koski, J. R. Sedell, and C. J. Cederholm.
1987a. Fine sediment and salmonid production: a paradox. Pages 98-142 in E. 0. Salo and T.
W. Cundy (editors) Strearnside management: forestry and fishery interactions. Contribution
No. 57, University of Washington, Institute of Forest Resources, Seattle.

Everest, F. H., G. H. Reeves, J. R. Sedell, D. B. Hohler, and T. Cain. 1987b. The effects
of habitat enhancement on steethead trout and coho salmon smolt production, habitat utilization,
and habitat availability in Fish Creek, Oregon, 1983-86. 1986 Annual Report, Project 84-11.
Bonneville Power Administration, Division of Fish and Wildlife, Portland, OR.

Everest, F. H., and J. R. Sedell. 1984. Evaluating effectiveness of stream enhancement
projects. Pages 246-256 in T. J. Hassler (editor) Proceedings of the Pacific Northwest stream
habitat management workshop. California Cooperative Fishery Research Unit, Humboldt State
University. Western Division, American Fisheries Society, Arcata, CA.

Everest, F. H., J. R. Sedell, G. H. Reeves, and M. D. Bryant. 1991. Planning and evaluating
habitat projects for anadromous salmonids. American Fisheries Society Symposium 10:68-77.

Everest, F. H., and D. R. Talhelm. 1982. Evaluating projects for improving fish and wildlife
habitat onnational forests. General Technical Report PNW-146. USDA Forest Service,
Portland, OR.

138 References

Everett, R. L. 1994. Eastside forest ecosystem health assessment. Volume IV: Restoration of
stressed sites and processes. General Technical Report PNW-GrR-330. USDA Forest Service,
Pacific Northwest Research Station, Portland, OR.

Everett, R. L., P. F. Hessburg, and T. R. Lillybridge. 1993. Emphasis areas as an alternative
to buffer zones and reserved areas in the conservation of biodiversity and ecosystem processes.
Pages 1-16 in American forests scientific workshop, Assessing forest health in the inland West.
USDA Forest Service, Portland, OR.

Fast, D., J. Hubble, M. Kohn, and B. Watson. 1991. Yakima River spring chinook
enhancement study. Final Report Project 82-16. U.S. Department of Energy, Bonneville Power
Administration, Division of Wildlife, Portland, OR.

Fausch, K. a 1984. Profitable stream positions for salmonids: relating specific growth rate
to net energy gain. Canadian Journal of Zoology 62:441-451.

FEMAT. 1993. Forest ecosystem management: an ecological, economic, and social
assessment. Report of the Forest Ecosystem Management Assessment Team. USDA Forest
Service, USDI Fish and Wildlife Service, USDC National Marine Fisheries Service, USDI
National Park Service, USDI Bureau of Land Management, U.S. Environmental Protection
Agency. U.S. Government Printing Office: 1993-793-071.

Ferguson, B. K. 1991. Urban stream reclamation. Journal of Soil and Water Conservation
Septem. ber-October: 324-328.

Ferguson, J. M. 1995. Handbook of criteria for forest practices implementation monitoring.
Alaska Working Group on Cooperative Forestry/Fisheries Research. Technical Report Number
95-02. USDA Forest Service, Wildlife and Fisheries, Juneau, AK.

Flosi, G., and E L. Reynolds. 1991. California salmonid stream habitat restoration manual.
California Department of Fish and Game, Sacramento, CA.

Franklin, J. E 1992. Scientific basis for new perspectives in forests and streams. Pages 25-72
in R. J. Naiman (editor) Watershed management: balancing sustainability and environmental
change. Springer-Verlag, New York.

Frissell, C. A., W. J. Liss, and D. Bayles. 1993. An integrated, biophysical strategy for
ecological restoration of large watersheds. Pages 449-456 in D. E. Potts (editor) Changing roles
in water resources management and policy, June 1993. Technical Publication Series TPS-93-2.
American Water Resources Association, Bethesda, MD.

Frissell, C. A., and R. K. Nawa. 1992. Incidence and causes of physical failure of artificial
habitat structures in streams of western Oregon and Washington. North American Journal of
Fisheries Management 12:182-197.

References 139

Froehlich, H. A., and D. H. McNabb. 1984. Minimizing soil compaction in Pacific Northwest
forests. Pages 159-192 in E. L. Stone (editor) Forest soils and treatment impacts. Proceedings,
Sixth North American Forest Soils Conference, University of Tennessee, Knoxville.

Fuller, D. D. 1990. Seasonal utilization of instream boulder structures by anadromous
salmonids in Hurdygurdy Creek, California. FHR Currents, Fish Habitat Relationships
Technical Bulletin No. 3. USDA Forest Service, Pacific Southwest Region, Eureka, CA.

Fuller, D. D., and A. J. Lind. 1994. Reducing negative impacts to foothill yellow-legged frogs
when implementing fish habitat improvement projects. Page 11 in Fish Habitat Relationships
Program, Fiscal Year 1993. USDA Forest Service, Pacific Southwest Region, Eureka, CA.

Furniss, M. J., T. D. Roelofs, and C. S. Yee. 1991. Road construction and maintenance.
Pages 297-323 in W. R. Meehan (editor) Influences of forest and rangeland management on
salmonid fishes and their habitats. Special Publication 19. American Fisheries Society, Bethesda,
MD.

Gibbons, D. R., and E. 0. Salo. 1973. An annotated bibliography of the effects of logging on
fish of the western United States and Canada. General Technical Report PNW-10. USDA Forest
Service, Portland, OR.

Gibbons, D. R., W. R. Meehan, K V. Koski, and T. R. Merrell, Jr. 1987. History of fisheries
and forestry interactions in southeastern Alaska. Pages 297-329 in E. 0. Salo and T. W. Cundy
(editors) Streamside management: forestry and fishery interactions. Institute of Forest Resources,
University of Washington, Seattle, WA.

Golding, D. L. 1987. Changes in strearnflow peaks following timber harvest of a coastal
British Columbia watershed. International Association of Hydrological Sciences 167:509-517.

Gore, J. A. (editor). 1985. The restoration of rivers and streams: theories and experience.
Butterworth. Stoneham, Massachusetts.

Grant, G. 1988. The, RAPID technique: a new method for evaluating downstream effects of
forest practices on riparian zones. General Technical Report PNW-220. USDA Forest Service
Pacific Northwest Research Station, Portland, OR.

Grant, G., C. McCain, and J. Cissel. 1994. Summary of the watershed-landscape analysis
workshop: H. J. Andrews Experimental Forest. General Technical Report PNW-GTR-338.
USDA Forest Service, Pacific Northwest Research Station, Portland, OR.

Gray, P. L., J. F. Koerner, and R. A. Marriott. 1981. The age structure and length-weight
relationship of southeastern Alaska coho salmon (Oncorhynchus kisutch), 1969-1970.
Informational Leaflet 195. Alaska Department of Fish and Game, Juneau, AK.

140 References

Gregory, R. S. 1993a. Effect of turbidity on the predator avoidance behavior of juvenile
chinook salmon (Oncorhynchus tsha"tscha). Canadian Journal of Fisheries and Aquatic
Sciences 50:241-246.

Gregory, S. V. 1993b. The basis for integrated watershed and stream restoration. In M. D.
Purser (editor) Proceedings, Watershed and stream restoration workshop: shared responsibilities
for shared watershed resources, August 26-27, 1993, Portland OR, not paginated. American
Fisheries Society, Oregon Chapter, Portland.

Gregory, S. V., and L. Ashkenas. 1990. Riparian management guide, Willamette National
Forest. USDA Forest Service. Portland, OR.

Gregory, S. V., and R. Wildman. 1994. Aquatic ecosystem restoration project, Quartz Creek,
Willamette National Forest. Department of Fisheries and Wildlife, Oregon State University,
Corvallis, OR.

Gregory, S. V., G. A. Lamberti, a C. Erman, K V. Koski, M. L. Murphy, and J. R. Sedell.
1987. Influence of forest practices on aquatic production. Pages 233-255 in E. 0. Salo and T.
W. Cundy (editors) Strearnside management: forestry and fishery interactions. Contribution No.
57. Institute of Forest Resources, University of Washington, Seattle.

Gregory, S. V., R. L. B@eschta, R J. Swanson, J. R. Sedell, G. H. Reeves, and F. H. Everest.
1990. Abundance of conifers in Oregon's riparian forests. Coastal Oregon Productivity
Enhancement Program 3:4-6.

Grette, G. B. 1985. The role of large organic debris in juvenile salmonid rearing habitat in
small streams. MS Thesis. University of Washington, Seattle.

Groot, C., and L. Margolis (editors). 1991. Pacific salmon life histories. Canada Department
of Fisheries and Oceans, Vancouver, B.C.

Grumbine, R. E. 1994. What is ecosystem management? Conservation Biology 8:27-38.

Hall, J. D. 1984. Evaluating fish response to artificial stream structures: problems and
progress. Pages 214-221 in T. J. Hassler (editor) Proceedings, Pacific Northwest stream habitat
management workshop. California Cooperative Fishery Research Unit, Humboldt State
University, Arcata, CA.

Hall, J. D., and C. 0. Baker. 1982. Influence of forest and rangeland management on
anadromous fish habitat in western North America. General Technical Report PNW-138. USDA
Forest Service Pacific Northwest Research Station, Portland, OR.

Harding, R. D. 1993. Abundance, size, habitat utilization, and instrearn movement of juvenile
coho salmon in a small southeast Alaska stream. MS Thesis. Juneau, AK.

References 141

Harr, R. D. 1983. Potential for augmenting water yield through forest practices in western
Washington and western Oregon. Water Resources Bulletin 19:383-393.

--. 1986. Effects of clearcutting on rain-on-snow runoff in western Oregon: a new look
at old studies. Water Resources Research 22:1095-1100.

Harr, R. D., R. L. Fredriksen, and J. S. Rothacher. 1979. Changes in stream flow following
timber harvest in southwestern Oregon. Research Paper PNW-249. USDA Forest Service,
Pacific Northwest Forest and Range Experiment Station, Portland, OR.

Hartman, G. H. 1988. Some preliminary comments on results of studies of trout biology and
logging impacts in Carnation Creek. Pages 175-180 in T. W. Chamberlin (editor) Proceedings
of the workshop: applying 15 years of Carnation Creek results. Pacific Biological Station,
Carnation Creek Steering Committee, Nanaimo, British Columbia.

Hartman, G., J. C. Scrivener, and L. Powell. 1987. Some effects of different strearnside
treatments on physical conditions and fish population processes in Carnation Creek, a coastal
rainforest stream in British Columbia. Pages 330-372 in E. 0. Salo and T. W. Cundy (editors)
Strearnside management: forestry @nd fishery interactions. Contribution No. 57. University of
Washington, Institute of Forest Resources, Seattle.

Hawkins, C. P., M. L. Murphy, and N. H. Anderson. 1982. Effects of canopy, substrate
composition, and gradient on the structure of macroinvertebrate communities in Cascade Range
streams of Oregon. Ecology 63:1840-1856.

Hawkins, C. P., M. L. Murphy, N. H. Anderson, and M. A. Wilzbach. 1983. Density of fish
and salamanders in relation to riparian canopy and physical habitat in streams of the
northwestern United States. Canadian Journal of Fisheries and Aquatic, Sciences 40:1173-1185.

Healey, M. C. 1983. Coastwide distribution and ocean migration patterns of stream- and
ocean-type chinook salmon, Oncorhynchus tshawytscha. Canadian Field-Naturalist 97:427-433.

Heggberget, T. G. 1988. Timing of spawning in Norwegian Atlantic salmon (Salmo salar).
Canadian Journal of Fisheries and Aquatic Sciences 45:845-849.

Heifetz, J., M. L. Murphy, and K V. Koski. 1986. Effects of logging on winter habitat of
juvenile salmonids in Alaska streams. North American Journal of Fisheries Management 6:52-
58.

Heifetz, J., S. W. Johnson, K V. Koski, and M. L. Murphy. 1989. Migration timing, size,
and salinity tolerance of sea-type sockeye salmon (Oncorhynchus nerka) in an Alaska estuary.
Canadian Journal of Fisheries and Aquatic Sciences 46:633-637.

Heimann, D. C. 1988. Recruitment trends and physical characteristics of coarse woody debris
in Oregon Coast Range streams. MS Thesis. Oregon State University, Corvallis.

142 Referen ces

Henderson, J. A. 1978. Plant succession on the Alnus rubralRubus spectablis habitat type in
western Oregon. Northwest Science 52:156-167.

Henderson, J. A., a H. Peter, R.. D.Lesher, and a c. Shaw. 1989. Forested plant
associations of the Olympic National Forest. Region Six Ecological Technical Paper 001-88.
USDA Forest Service, Portland, OR.

Henly, R. K., and P. V. Ellefson. 1987. State-administered forestry programs: Current status
and prospects for expansion. Renewable Resources Journal Autumn 1987:19-23.

Herricks, E. E., and L. L. Osborne. 1985. Water quality restoration and protection in streams
and rivers. Pages 1-20 in J. A. Gore (editor) The restoration of rivers and streams.
Butterworth, Boston.

Hetherington, E. D. 1988. Hydrology and logging in the Carnation Creek watershed-What
have we learned? Pages 11-15 in T. W.. Chamberlin (editor) Proceedings of the workshop:
applying 15 years of Carnation Creek results. Pacific Biological Station, Carnation Creek
Steering Committee, Nanaimo, British Columbia.

Hibbs, D. E., P. Giordano, and S. Chan. 1990. Managing riparian vegetation. Coastal Oregon
Productivity Enhancement Program 3:7-8.

--. 1991. Vegetation dynamics in managed coastal riparian areas. Coastal Oregon
Productivity Enhancement Program 4:3-5.

Hicks, B. J. 1990. The influence of geology and timber harvest on channel geomorphology and
salmonid populations in Oregon Coast Range streams. Ph.D. Dissertation. Oregon State
University, Corvallis.

Hicks, B. J., J. D. Hall, R A. Bisson, and J. R. Sedell. 1991a. Responses of salmonids to
habitat changes. Pages 483-518 in W. R. Meehan (editor) Influences of forest and rangeland
management on salmonid fishes and their habitats. Special Publication 19. American Fisheries
Society, Bethesda, MD.

Hicks, B. J., R. L. Beschta, and R. D. Harr. 1991b. Long-term changes in strearnflow
following logging in western Oregon and associated fisheries implications. Water Resources
Bulletin 27:217-226.

Hoelscher, B., J. Colla, M. Hanson, J. Heimer, D. McGreer, S. Poirier, and J. Rice. 1993.
Forest practices water quality audit 1992. Idaho Department of Health and Welfare, Division
of Environmental Quality, Boise, ID.

Hogan, D. L., and J. W. Schwab. 1991. Stream channel response to landslides in the Queen
Charlotte Islands, B.C.: Changes affecting pink and chum salmon habitat. Pages 222-236 in B.
White and L Guthrie (editors) Proceedings of the 15th Northeast Pacific pink and chum salmon

References 143

workshop, Parksville, B.C., February 27-March 1, 1991. Pacific Salmon Commission,
Vancouver, B. C.

Holtby, L. B. 1988. Effects of logging on stream temperatures in Carnation Creek, British
Columbia, and associated impacts on the coho salmon (Oncorhynchus kisutch). Canadian Journal
of Fisheries and Aquatic Sciences 45:502-515.

Holtby, L. B., and I C. Scrivener. 1989. Observed and simulated effects of climatic
variability, clear-cut logging and fishing on the numbers of chum salmon (Oncorhynchus keta)
and coho salmon (0. kisutch) returning to Carnation Creek, British Columbia. Canadian Special
Publication of Fisheries and Aquatic Sciences- 105:62-81.

Holtby, L. B., T. E. McMahon, and J. C. Scrivener. 1989. Stream temperatures and inter-
annual variability in the emigration timing of coho salmon (Oncorhynchus kisutch) smolts and
fry and chum salmon (0. keta) fry from Carnation Creek, British Columbia. Canadian Journal
of Fisheries and Aquatic Sciences 46:1396-1405.

House, R. A., V. Crispin, and R. Monthey. 1989. Evaluation of stream rehabilitation projects:
Salem District (1981-1988). BLM-OR-PT-90-10-6600.9. USDI Bureau of Land Management,
Salem,'OR.

House, R. A., V. A. Crispin, and J. M. Suther. 1991. Habitat and channel changes after
rehabilitation of two coastal streams in Oregon. American Fisheries Society Symposium 10: 150-
159.

Hunt, R. L. 1969. Effects of habitat alteration on production, standing crops and yield of
brook trout in Lawrence Creek, Wisconsin. Pages 281-312 in T. G. Northcote (editor)
Symposium on salmon and trout in streams. H.R. MacMillan Lectures in Fisheries. University
of British Columbia, Institute of Fisheries, Vancouver, B.C.

Hunter, C. J. 1991. Better trout habitat-a guide to stream restoration and management.
Montana Land Reliance, Island Press. Washington, D.C.

Huppert, D. D., and R. D. Fight. 1991. Economic considerations in managing salmonid
habitats. Pages 559-585 in W. R. Meehan (editor) Influences of forest and rangeland
management on salmonid fishes and their habitats. Special Publication 19. American Fisheries
Society, Bethesda, MD.

IDL. 1992. Rules and regulations pertaining to the Idaho Forest Practices Act, Title 38,
Chapter 13, Idaho Code. Idaho State Board of Land Commissioners, Department of Lands,
Boise, ID.

1994. A draft cumulativi watershed effects process for Idaho. Idaho Department of
Lands, Boise, ID.

144 References

Idaho Water Quality Bureau. 1988. Final report: Forest Practices Act water quality audit.
Idaho Department of Health and Welfare, Division of Environmental Quality, Boise, ID.

Jensen, M. E., and P. S. Bourgeron (editors). 1994. Ecosystem Management: Principles and
Applications. Volume 11. Eastside Forest Ecosystem Health Assessment. USDA Forest
Service, Pacific Northwest Research Station, Portland, OR.

Johnson, A. W., and D. M. Ryba. 1992. A literature review of recommended buffer widths
to maintain various functions of stream riparian areas. King County Surface Water Management
Division, Seattle, WA.

Johnson, S. W., J. Heifetz, and K V. Koski. 1986. Effects of logging on the abundance and
seasonal distribution of juvenile steelhead in some southeastern Alaska streams. North American
Journal of Fisheries Management 6:532-537.

Johnson, S. W., J. F. Thedinga, and K V. Koski. 1992. Life history of juvenile ocean-type
chinook salmon (Oncorhynchus tshanywha) in the Situk River, Alaska. Canadian Journal of
Fisheries and Aquatic Sciences 49:2621-2629.

Kershner, J. L. 1993. Concepts of integrated restoration planning In M. D. Purser (editor)
Proceedings, Watershed and stream restoration workshop: shared responsibilities for shared
watershed resources, conducted August 26-27, 1993, Portland OR, not paginated. American
Fisheries Society, Oregon Chapter, Portland.

King, J. G. 1989. Strearnflow responses to road building and hanesting: A comparison with
the equivalent clearcut area procedure. Research Paper INT-401. USDA Forest Service,
Intermountain Forest and Range Experiment Station, Ogden, Utah.

Klock, G. 0. 1975. Impact of five postfire salvage logging systems on soils and vegetation.
Journal of Soil and Water Conservation 30:78-81.

Klock, G. 0. 1985. Modeling the cumulative effects of forest practices on downstream aquatic
ecosystems. Journal of Soil and Water Conservation 40:237-241.

Konopacky, R. C. 1984. Sedimentation and productivity in a salmonid stream. Ph.D.
Dissertation. University of Idaho, Moscow.

Koski, K V. 1966. The survival of coho salmon (Oncorhynchus kisutch) from egg deposition
to emergence in three Oregon coastal streams. MS Thesis. Oregon State University, Corvallis.

Koski, K V. 1975.. The survival and fitness of two stocks of chum salmon (Oncorhynchus keta)
from egg deposition to emergence in a controlled-stream environment at Big Beef Creek. Ph.D.
Dissertation. University of Washington, Seattle.

References 145

, 1992. Restoring stream habitats affected by logging activities. Pages 343-404 in G.
@V. Thayer (editor) Restoring the nation's marine environment. Publication UM-SG-TS-92-06.
Maryland Sea Grant College, College Park, MD.

Larsson, P. 0. 1985. Predation on migrating smolt as a regulating factor in Baltic salmon,
Salmo salar L., populations. Journal of Fish Biology 26:391-397.

Leider, S. A., M. W. Chilcote, and J. J. Loch. 1986. Movement and survival of presmolt
steelhead in a tributary and main stem of a Washington river. North American Joumal of
Fisheries Management 6:526-531.

Leidy, R. A. 1984. Distribution and ecology of stream fishes in the San Francisco Bay
drainage. Hilgardia 52.

Leopold, L. B., M. G. Wolman, and J. P. Miller. 1964. Fluvial processes in geomorphology.
W. H.Treeman and Company, San Francisco.

Li, H. W.,, C. B. Schreck, C. E. Bond, and E. Rexstad. 1987. Factors influencing changes
in fish assemblages of Pacific Northwest streams. Pages 193-2.2 in W. J. Matthews and D. C.
Heins (editors) Community and evolutionary ecology of North American stream fishes.
University of Oklahoma Press, Norman, OK.

Lienkaemper, G. W., and E J. Swanson. 1987. Dynamics of large woody debris in streams
in old-growth Douglas-fir forests. Canadian Journal of Forest Research 17:150-156.

Lippke, B., and C. Oliver. 1994. An economic tradeoff system for ecosystem management.
Pages 318-326 in M. E. Jensen and P. S. Bourgeron (editors) Eastside forest ecosystem health
assessment. Volume 11: Ecosystem management: Principles and applications. PNW-GFR-314.
USDA Forest Service, Pacific Northwest Research Station, Portland, OR.

Lisle, T. E. 1986. Effects of woody debris on anadromous salmonid habitat, Prince of Wales
Island, southeast Alaska. North American Journal of Fisheries Management 6:538-550.

Lisle, T. E., and S. Hilton. 1992. The volume of fine sediment in pools: an index of sediment
supply in gravel-bed streams. Water Resources Bulletin 28:371-383.

Lorenz, J. M., and J. H. Eiler. 1989. Spawning habitat and redd characteristics of sockeye
salmon in the glacial Taku River, British Columbia and Alaska. Transactions of the American
Fisheries Society 118:495-502.

Lynch, J. A., and E. S. Corbett. 1990. Evaluation of best management practices for
controlling nonpoint pollution from silvicultural operations. Water Resources Bulletin 26:41-52.

Macdonald, J. S., G. Miller, and R. A. Stewart. 1988. The effects of logging, other forest
industries and forest management practices on fish: an initial bibliography. Canadian Technical
Report of Fisheries and Aquatic Sciences No. 1622.

146 References

MacDonald, L. H., and A. Smart. 1993. Beyond the Guidelines: practical lessons for
monitoring. Environmental Monitoring and Assessment 26:203-218.

MacDonald, L. H., A. W. Smart, and R. C. Wissmar. 1991. Monitoring guidelines to evaluate
effects of forestry activities on streams in the Pacific Northwest and Alaska. EPA/910/9-91-001.
USEPA Region 10 in Cooperation with the University of Washington Center for Strearnside
Studies, Seattle, WA.

Marcot, B. G., M. J. Wisdom, H. W. Li, and G. C. Castillo. 1,994. Managing for featured,
threatened, endangered, and sensitive species and unique habitats for ecosystem sustainability.
General Technical Report PNW-GTR-329. USDA Forest Service, Pacific Northwest Research
Station, Portland, OR.

Mason, J. C. 1976. Response of underyearling coho salmon. to supplemental feeding in a
natural stream. Journal of Wildlife Management 40:775-788.

Mason, J. C., and a w. Chapman. 1965. Significance of early emergence, environmental
rearing capacity, and behavioral ecology of juvenile coho salmon in stream channels. Journal
of the Fisheries Research Board of Canada 22:173-190.

Mason, R. R., and B. E. Wickman. 1994. Procedures to reduce landscape hazard from insect
outbreaks. Pages 20-21 in Everett, R. L. (compiler) Restoration of stressed sites and processes.
General Technical Report PNW-GTR-330. USDA Forest Service, Pacific Northwest Research
Station, Portland, OR.

Mathews, S. B., and F. W. Olson. 1980. Factors affecting Puget Sound coho salmon
(Oncorhynchus kisutch) runs. Canadian Journal of Fisheries and Aquatic Sciences 37:1373-1378.

McCain, M. E. 1992. Comparison of habitat use and availability for juvenile fall chinook
salmon in a tributary of the Smith River, CA. FHR Currents, Fish Habitat Relationships
Technical Bulletin No. 7. USDA Forest Service, Pacific Southwest Region, Eureka, CA.

McCain, M., D. D. Fuller, L. Decker, and K. Overton. 1990. Stream habitat classification
and inventory procedures for northern California. FHR Currents, Fish Habitat Relationships
Technical Bulletin No. 1. USDA Forest Service, Pacific Southwest Region, Eureka, CA.

McCammon, B. 1993. Determining the risk of cumulative watershed effects resulting from
multiple activities-Endangered Species Act, Section 7. USDA Forest Service, Portland, OR.

McDade, M. H., F. J. Swanson, W. A. McKee, J. E Franklin, and J. Van Sickle. 1990.
Source distances for coarse woody debris entering small streams in western Oregon and
Washington. Canadian Journal of Forest Resources 20:326-330.

McIntosh, B. A., J. R. Sedell, J. E. Smith, R. C. Wissmar, S. E. Clarke, G. H. Reeves, and
L. A. Brown. 1994. Management history of eastside ecosystems: changes in fish habitat over

References 147

50 years, 1935 to 1992. General Technical Report PNW-GTR-321. USDA Forest Service,
Pacific Northwest Research Station, Portland, OR.

McMahon, T. 1989. Large woody debris and fish. In Silvicultural management of riparian
areas for multiple resources, not paginated. A COPE Workshop. USDA Forest Service, Pacific
Northwest Research Station, and Oi-egon State University Forestry College, Corvallis, OR.

McNeil, W. J. 1964. Effect of the spawning bed environment on reproduction of pink and
chum salmon. Fishery Bulletin 65:495-523.

Meehan, W. R. (editor) 1991. Influences of forest and rangeland management on salmonid
fishes and their habitats. Special Publication 19. American Fisheries Society, Bethesda, ma

Meehan, W. R., and T. C. Bjornn. 1991. Salmonid distributions and life histories. Pages 47-
82 in W. R. Meehan (editor) Influences of forest and rangeland management on salmonid fishes
and their habitats. Special Publication 19. American Fisheries Society, Bethesda, MD.

Meehan, W. R., F. J. Swanson, and J. R. Sedell. 1977. Influences of riparian vegetation on
aquatic ecosystems with particular reference to salmonid fishes and their food supply. Pages
137-145 in Importance, preservation, and management of riparian habitat: a symposium. USDA
Forest Service, Pacific Northwest Forest and Range Experiment Station, Corvallis, OR.

Megahan, W. F. 1983. Hydrologic effects of clearcutting and wildfire on steep granitic slopes
in Idaho. Water Resources Research 19:811-819.

Minshall, G. W., J. T. Brock, and J. D. Varley. 1989. Wildfires and Yellowstone's stream
ecosystems. BioScience 39:707-715.

Minshall, G. W., D. A. Andrews, J. T. Brock, C. T. Robinson, and D. E. Lawrence. 1990.
Changes in wild trout habitat following forest fire. Pages 111- 119 in Proceedings of Wild Trout
IV Symposium. Yellowstone National Park, Wyoming, Sept. 18-19, 1989.

Minshall, G. W., and J. T. Brock. 1991. Observed and anticipated effects of forest fire on
Yellowstone stream ecosystems. Pages 123-135 in R. B. Keiter and M. S. Boyce (editors) The
Greater Yellowstone Ecosystem: Redefining America's wilderness heritage. Yale University
Press.

Minshall, G. W., J. T. Brock, and J. D. Varley. 1989. Wildfires and Yellowstone's stream
ecosystems. BioScience 39:707-715.

Murphy, M. L. 1985. Die-offs of pre-spawn adult pink salmon and chum salmon in
southeastern Alaska. North American Journal of Fisheries Management 5:302-308.

Murphy, M. L., and J. D. Hall. 1981. Varied effects of clearcut logging on predators and their
habitat in small streams on the Cascade Mountains, Oregon. Canadian Journal of Fisheries and
Aquatic Sciences 38:137-145.

148 References

Murphy, M. L., and K V. Koski. 1989. Input and depletion of woody debris in Alaska streams
and implications for strearnside management. North American Journal of Fisheries Management
9:427-436.

Murphy, M. L., and W. R. Meehan. 1991. Stream ecosystems. Pages 17-46 in W. R.
Meehan (editor) Influences of forest and rangeland management on salmonid fis'hes and their
habitats. Special Publication 19. American Fisheries Society, Bethesda, MD.

Murphy, M. L., C. P. Hawkins, and N. H. Anderson. 1981. Effects of canopy modification
and accumulated sediment on stream communities. Transactions of the American Fisheries
Society 110:469-478.

Murphy, M. L., K V. Koski, J. Heifetz, S. W. Johnson, D. Kirchoffer, and J. F. Thedinga.
1984a. Role of large organic debris as winter habitat for juvenile salmonids in Alaska streams.
Proceedings of the Annual Conference of the Western Association of Fish and Vtrildlife Agencies
64:251-262.

Murphy, M. L., J. F Thedinga, K V. Koski, and G. B. Grette. 1984b. A stream ecosystem
in an old-growth forest in southeast Alaska. Part 5:. seasonal changes in habitat utilization by
juvenile salmonids. Pages 89-98 in W. R. Meehan, T. R. Merrell, Jr., and T. A. Hanley
(editors) Proceedings, fish and wildlife relationships in old-growth forests symposium.
American Institute of Fishery Research Biologists, Asheville, NC.

Murphy, M. L., J. Heifetz, S. W. Johnson, K V. Koski, and J. R Thedinga. 1986. Effects
of clear-cut logging with and without buffer strips on juvenile salmonids in Alaskan streams.
Canadian Journal of Fisheries and Aquatic Sciences 43:1521-1533. .

Murphy, M. L., J. M. Lorenz, J. Heifetz, J. F. Thedinga, K V. Koski, and S. W. Johnson.
1987. The relationship between stream classification, fish, and habitat in Southeast Alaska.
Wildlife and Fisheries Habitat Management Notes, R10-MB-10. USDA Forest Service, Tongass
National Forest, Juneau, AK.

Murphy, M. L., J. E Thedinga, and K V Koski. 1988. Size and diet of juvenile Pacific
salmon during seaward migration through a small estuary in southeastern Alaska. Fishery
Bulletin 86:213-222.

Murphy, M. L., J. Heifetz, S. W. Johnson, K V. Koski, and J. F. Thedinga. 1989. Habitat
utilization by juvenile Pacific salmon (Oncorhynchus) in the glacial Taku River, southeast
Alaska. Canadian Journal of Fisheries and Aquatic Sciences 46:1677-1685.

Naiman, R. J. (editor). 1992. Watershed management: balancing sustainability and
environmental change. Springer-Verlag, New York.

Naiman, R. j., aG. Lonzarich, T. I Beechie, and S. C. Ralph. 1991. General principles of
classification and the assessment of conservation potential in rivers. Pages 93-123 in P. J.

References 149

Boon, P. Calow and G. E. Petts (editors) River conservation and management. John Wiley and
-Sons, New York.

Narver, D. W. 1971. Effects of logging debris on fish production. Pages 100-111 in J. T.
Krygier and J. D. Hall (editors) Proceedings of a symposium on forest land uses and stream
environment. Oregon State University, Corvallis.

Nehlsen, W., J. E. Williams, and J. A. Lichatowich. 1991. Pacific salmon at the crossroads:
Stocks at risk from California, Oregon, Idaho, and Washington. Fisheries 16:4-21.

Nelson, R. L., M. L. McHenry, and W. S. Platts. 1991. Mining. Pages 425-458 in W. R.
Meehan (editor) Influences of forest and rangeland management on salmonid fishes and their
habitats. Special Publication 19. American Fisheries Society, Bethesda, MD.

Netboy, A. 1974. The salmon: their fight for survival. Houghton Mifflin Company, Boston.

Newton, M., R. Willis, J. Walsh, E. Cole, and S. Chan. 1995. Enhancing riparian habitat for
fish, wildlife and timber in managed forests. Biological conservation symposium. Annual
meeting of the Weed Science Society of America, February 1995, Seattle, WA.

Nickelson, T. E., J. D. Rodgers, S. L. Johnson, and M. F. Solazzi. 1992a. Seasonal changes
in habitat use by juvenile coho, salmon (Oncorhynchus kisutch) in Oregon coastal streams.
Canadian Journal of Fisheries and Aquatic Sciences 49:783-789.

Nickelson, I E., M. F. Solatzi, S. L. Johnson, and J. D. Rodgers. 1992b. Effectiveness of
selected stream improvement techniques to create suitable summer and winter rearing habitat
for juvenile coho salmon (Oncorhynchus kisutch) in Oregon coastal streams. Canadian Journal
of Fisheries and Aquatic Sciences' 49:790-794.

Norris, L. A., H. W. Lorz, and S. V. Gregory. 1991. Forest Chemicals. Pages 207-296 in
W. R. Meehan (editor) Influences of forest and rangeland management on salmonid fishes and
their habitats. Special Publication 19. American Fisheries Society, Bethesda, MD.

ODF. 1994. Water classification and protection rules. Oregon Department of Forestry, Salem,
OR.

ODF and ODFW. 1995. A guide to placing large wood in streams. Oregon Department of
Forestry, Forest Practices Section, Salem, OR.

Pacific Lumber Company. 1993. Cooperative fisheries program. Annual report. Pacific
Lumber Company, Scotia, CA.

Pacific Rivers Council. 1993a. The protection and restoration of watersheds and salmon habitat
on federal lands throughout the Pacific Northwest: technical, legal, and economic requirements.
Pacific Rivers Council, Eugene, OR.

150 References

Pacific Rivers Council. 1993b. The new watershed imperative: A new approach to restore
America's river ecosystems and biodiversity. Pacific Rivers Council, Eugene, OR.

Palmisano, J. E, R. H. Ellis, and V. W. Kaczynski. 1993. The impact of environmental and
management factors on Washington's wild anadromous salmon and trout: summary and
recommendations. Washington Department of Natural Resources, Olympia, WA.

Payne, N. F., and F. Bryant. 1994. Techniques for wildlife management of uplands.
Biological Resource Management Series, McGraw Hill, New York.

Peterson, N. P. 1982. Immigration of juvenile coho salmon (Oncorhynchus kisutch) into
riverine ponds. Canadian Journal of Fisheries and Aquatic Sciences 39:1308-1310.

Phinney, D. E., M. S. Deusen, S. M. Keller, and P. A. Knudsen. 1989. A new approach to
riparian management in Washington State. Pages 11-15 in Practical approaches to riparian
resource management, proceedings of an educational workshop, May 8-11, 1989, Billings,
Montana. American Fisheries Society, Bethesda, MD.

Platts, W. S. 1991. Livestock grazing. Pages 389-424 in W. R. Meehan (editor) Influences of
forest and rangeland management on salmonid fishes and their habitats. Special Publication 19.
American Fisheries Society, Bethesda, MD.

Platts, W. S., and W. F. Megahan. 1975. Time trends in riverbed sediment composition in
salmon and steelhead spawning areas: South Fork Salmon River, Idaho. Transactions of the
North American Wildlife and Natural Resources Conference 40:229-239.

Powell, D. S., J. L. Faulkner, D. R. Darr, Z. Shu, and D. W. MacCleery. 1993. Forest
resources of the United States, 1992. General Technical Report RM-234. USDA Forest
Service, Rocky Mountain Forest and Range Experiment Station, Fort Collins, CO.

Ralph, S. C., G. C. Poole, L. L. Conquest, and R. J. Naiman. 1994. Stream channel
morphology and woody debris in logged and unlogged basins of western Washington. Canadian
Journal of Fisheries and Aquatic Sciences 51:37-51.

Resource Development Council. 1994. Harvesting trees in the buffer zones. Resource Review:
October 1994. Resource Development Council, Inc., Anchorage, AK.

Reeves, G. H., L. E. Benda, P. A. Bisson, and J. R. Sedell. In press. A disturbance-based
ecosystem approach to maintaining and restoring freshwater habitats of evolutionarily significant
units of anadromous salmonids in the Pacific Northwest. In Proceedings of a symposium on
significance of evolutionary concepts in salmonid habitat management, July 1994, Monterey,
CA. American Fisheries Society, Bethesda, MD.

Reeves, G. H., F. H. Everest, and T. E. Nickelson. 1989. Identification of physical habitats
limiting the production of coho salmon in western Oregon and Washington. General Technical

References 151

Report PNW-GrR-245. USDA Forest Service, Pacific Northwest Research Station, Portland,
OR.

Reeves, G. H., F. H. Everest, and J. D. Hall. 1987. Interactions between the redside shiner
(Richardsonius balteatus) and the steelhead trout (Salmo gairdneri) in western Oregon: the
influence of water I temperature. Canadian Journal of Fisheries and Aquatic Sciences 44:1603-
1613.

Reeves, G. H., E H. Everest, and J. R. Sedell. 1993. Diversity of juvenile anadromous
salmonid assemblages in coastal Oregon basins with different levels of timber harvest.
Transactions of the American Fisheries Society 122:309-317.

Reeves, G. H., L D. Hall, T. D. Roelofs, T. L. Hickman, and C. 0. Baker. 1991.
Rehabilitating and modifying stream habitats. Pages 519-557 in W. R. Meehan (editor)
Influences of forest and rangeland. management on salmonid fishesand their habitats. Special
Publication 19. American Fisheries Society, Bethesda, MD.

Reeves, G. H., and J. R. Sedell. 1992. An ecosystem approach to the conservation and
management of freshwater habitat for anadromous salmonids in the Pacific Northwest. North
American Wildlife Natural Resource Conference 57:408-415.

Reid, L. M., and T. Dunne. 1984. Sediment production from forest road surfaces. Water
Resources Research 20:1753-1761.

Reid, L. M., and R. R. Ziemer. 1994a. Evaluating the biological significance of intermittent
streams. Pages 17-28 in Notes from seminars held at Humboldt Interagency Watershed Analysis
Center. USDA Forest Service, McKinleyville, CA.

Reid, L. M., and R. R. Ziemer. 1994b. Basin assessment and watershed analysis. Pages 63-
78 in Notes from seminars held at Humboldt Interagency Watershed Analysis Center. USDA
Forest Service, McKinleyville, CA.

Reid, L. M., R. R. Ziemer, and M. J. Furniss. 1994. Watershed analysis in the federal arena.
Pages 1-16 in M. J. Furniss (editor) Notes from seminars held at the Humboldt Interagency
Watershed Analysis Center. USDA Forest Service, McKinleyville, CA.

Rhodes, J., and D. A. McCullough. In press. Coarse screening process. Columbia River
Inter-Tribal Fish Commission, Portland, OR.

Ricker, W. E. 1972. Hereditary and environmental factors affecting certain salmonid
populations. Pages 19-160 in R. C. Simon and P. A. Larkin (editors) The stock concept in
Pacific salmon. University of British Columbia, Vancouver, B.C.

Rieman, B. E., R. C. Beamesderfer, S. Vigg, and T. P. Poe. 1991. Estimated loss of juvenile
salmonids to predation by northern squawfish, walleyes, and smallmouth bass in John Day
Reservoir, Columbia River. Transactions of the American Fisheries Society 120:448-459.

152 References

Ritter, D. E 1978. Process geomorphology. Wm. C. Brown Company Publishers, Dubuque,
IA.

Robison, E. G., and R. L. Beschta. 1990. Identifying trees in riparian areas that can provide
coarse woody debris to streams. Forest Science 36:790-801.

Rodgers, J. D., R. D. Ewing, and J. D. Hall. 1987. Physiological changes during seaward
migration of wild juvenile coho salmon (Oncorhynchus kisutch). Canadian Journal of Fisheries
and Aquatic Sciences 44:452-457.

Salo, E. 0., and T. W. Cundy.' (editors) 1987. Strearnside management: Forestry and fishery
interactions: Proceedings of a symposium; 12-14 February 1986, Seattle, WA. Contribution No.
57, Institute of Forest Resources, University of Washington. Seattle.

Schlosser, 1. J. 1982. Trophic structure, reproductive success, and growth rate of fish in a
natural and modified headwater stream. Canadian Journal of Fisheries and Aquatic Sciences
39:968-978.

Scott, J. B., C. R. Steward, and Q. J. Stober. 1986. Effects of urban development on fish
population dynamics in Kelsey Creek, Washington. Transactions of the American Fisheries
Society 115:555-567.

Scrivener, J. C. 1991. An update and application of the production model for Carnation Creek
chum salmon. Pages 210-219 in B. White and I. Guthrie (editors) Proceedings of the 15th
Northeast Pacific pink and chum salmon workshop. Pacific Salmon Commission, Vancouver,
B.C.

Scrivener, J. C., and M. J. Brownlee. 1989. Effects of forest harvesting on gravel quality and
incubation survival of chum salmon (Oncorhynchus keta) and coho salmon (0. kisutch) in
Carnation Creek, British Columbia. Canadian Journal of Fisheries and Aquatic Sciences 46:681-
696.

Sedell, J. R., and R. L. Beschta. 1991. Bringing back the "bio" in bioengineering. American
Fisheries Symposium 10: 160-175.

Sedell, J. R., and J. L. Froggatt. 1984. Importance of streamside forests to large rivers: the
isolation of the Willamette River, Oregon, U.S.A., from its floodplain by snagging and
streamside forest removal. Internationale Vereinigung ffir Theoretische und Angewandte
Limnologie Verhandlungen 22:1828-1834.

Sedell, J. R., and K. J. Luchessa. 1982. Using the historical record as an aid to salmonid
habitat enhancement. Pages 210-223 in N. B. Armantrout (editor) Acquisition and utilization
of aquatic habitat inventory information symposium. American Fisheries Society, Western
Division, Bethesda, MD.

References 153

Sedell, J. R., and F. J. Swanson. 1984. Ecological characteristics of streams in old-growth
forests of the Pacific Northwest. Pages 9-16 in W. R. Meehan, T. R. Merrell, Jr., and T. A.
Hanley (editors) Proceedings, fish and wildlife relationships in old-growth forests symposium.
American Institute of Fishery Research Biologists, Asheville, NC.

Sedell, J. R., P. A. Bison, R J. Swanson, and S. V. Gregory, 1988. What we know about
large trees that fall into streams and rivers. Pages 47-81 in C. Maser (editor) From the forest
to the sea: a story of fallen trees. General Technical Report PNW-GTR-229. USDA Forest
Service, Pacific Northwest Research Station, Portland, OR.

Sedell, J. R., F H. Everest, and a R. Gibbons. 1989. Strearnside vegetation management for
aquatic habitat. Pages 115-125 in Proceedings, national silvicultural workshop-silviculture for
all resources. USDA Forest Service, Timber Management, Washington, D.C.

Sedell, J. R., F. N. Leone, and W. S. Duval. 1991. Water transportation and storage of logs.
Pages 325-368 in W. R. Meehan (editor) Influences of forest and rangeland management on
salmonid. fishes and their habitats. Special publication 19. American Fisheries Society,
Bethesda, MD.

Shepard, W. B. 1994. Modem forest management: It's about openmg up, not locking up.
Pages 218-227 in W. W. Covington and L. F. DeBano (editors) Sustainable ecological systems:
Implementing an ecological approach to land management. Report RM-247. USDA Forest
Service, Rocky Mountain Forest and Range Experiment Station, Fort Collins, CO.

Sheridan, W. L., and A. M. Bloom. 1975. Effects of canopy removal on temperatures of some
small streams in southeast Alaska. USDA Forest Service, Alaska Region, June 1975.

Sidle, R. C., A. J. Pearce, and C. O'Loughlin. 1985. Hillslope stability and land use.
American Geophysical Union, Washington, D.C.

Simpson Timber Company. 1994. Monitoring impacts of timber harvest on streams within
Simpson Timber Company lands in northern California. Simpson Timber Company, Korbel,
CA.

Smoker, W. A. 1955. Effects of strearnflow on silver salmon production in western
Washington. Ph.D. Dissertation. University of Washington, Seattle.

Solazzi, M. F., and S. L. Johnson. 1994. Development and evaluation of techniques to
rehabilitate Oregon's wild coho salmon. Fish Research Project F-125-R. Annual Progress
Report. Oregon Department of Fish and Wildlife, Portland, OR.

Sparks, R. E., R B. Bayley, S. L. Kohler, and L. L. Osborne. 1990. Disturbance and
recovery of large floodplain rivers. Environmental Management 14:711-724.

Steinblums, I. J., H. A. Froehlich, and J. K. Lyons. 1984. Designing stable buffer strips for
stream protection. Journal of Forestry 82:49-52.

154 References

Strahler, A. N. 1957. Quantitative analysis of watershed geomorphology. Transactions of the
American Geophysical Union 38:913-920.

Sullivan, K., T. E. Lisle, C. A. Dolloff, G. E. Grant, and L. M. Reid. 1987. Stream
channels: the link between forests and fishes. Pages 39-97 in E. 0. Salo and T. W. Cundy
(editors) Strearnside management: forestry and fisheries interactions. Contribution No. 57,
University of Washington, Institute of Forest Resources, Seattle.

E.
Swanson, F. J., L. E. Benda, S. H. Duncan, G. Grant, W. F. Megahan, L. M. Reid, and
R. R. Ziemer. 1987. Mass failures and other processes of sediment production in Pacific
Northwest forest landscapes. Pages 9-38 in E. 0. Salo and T. W. Cundy (editors) Strearnside
management: forestry and fishery interactions. Contribution No. 57, Institute of Forest
Resources, University of Washington, Seattle.

Swanston, D. N. 1991. Natural processes. Pages 139-179 in W. R. Meehan (editor)
Influences of forest and rangeland management on salmonid fishes and their habitats. Special
Publication 19. American Fisheries Society, Bethesda, MD.

Tagart, J. V. 1976. The survival from egg deposition to emergence of coho salmon in the
Clearwater River, Jefferson County, Washington. MS Thesis. University of Washington,
Seattle.

Talley, K. 1990. Stats Pack-everything you wanted to know about landings, prices, and
predictions- for the major Pacific Coast fisheries. Pacific Fishing 11: 126.

Thedinga, J. F., M. L. Murphy, J. Heifetz, K V. Koski, and S. W. Johnson. 1989. Effects
of logging on size and age composition of juvenile coho salmon and density of presmolts in
southeast Alaska streams. Canadian Journal of Fisheries and Aquatic Sciences 46:1383-1391.

Thedinga, J. F., S. W. Johnson, K V. Koski, J. M. Lorenz, and M. L. Murphy. 1993.
Potential effects of flooding from Russell Fiord on salmonids and habitat in the Situk River,
Alaska. U.S. Department of Commerce, NOAA Technical Memorandum NMFS F/NWC-221.

Thedinga, J. F., M. L. Murphy, S. W. Johnson, J. M. Lorenz, and K V. Koski. 1994.
Determination of salmonid smolt yield with rotary-screw traps in the Situk River, Alaska, to
predict effects of glacial flooding. North American Journal of Fisheries Management 14:837-
851.

Timber Fish, Wildlife. 1992. Effectiveness of Washington's forest practice riparian management
zones; regulations for protection of stream temperature. Report No. TFW-WQ6-92-001.
Washington State Department of Ecology, Olympia, WA.

Toews, D. A. A., and M. K. Moore. 1982. The effects of strearnside logging on large organic
debris in Carnation Creek. Report 11. British Columbia Ministry of Forest Land Management,
Victoria, B. C.

References 155

Tripp, D. B.,, and V. A. Poulin. 1986. The effects of logging and mass wasting on salmonid
spawning habitat in streams on the Queen Charlotte Islands. Land Management Report No. 50,
British Columbia Ministry of Forests and Lands, Victoria, B.C.

Trout Unlimited. 1994. Conserving salmonid biodiversity on federal lands: Trout Unlimited's
policy on mining, grazing, and timber harvest. Trout Unlimited, Arlington, VA.

Tschaplinski, R J., and G. F. Hartman. 1983. Winter distribution of juvenile coho salmon
(Oncorhynchus kisutch) before and after logging in Carnation Creek, British Columbia, and some
implications for overwinter survival. Canadian Journal' ,of Fisheries and Aquatic Sciences
40:452-461.

USDA. 1989. Understanding the past, designing the future. Publication R10-MB-85. USDA
Forest Service, Alaska Region, Juneau, AK.

--. 1995. Report to Congress: Anadromous Fish Habitat Ass-ssment. Report R10-MB-
279. USDA Forest Service, Pacific Northwest Research Station and the Alaska Region, Juneau,
AK.

USDA and USDI. 1994a. Record of decision for amendments to Forest Service and Bureau
of Land Management planning documents within the range of the northern spotted owl: standards
and guidelines for management of habitat for late-successional and old-growth forest related
species within the, range of the northern spotted owl. USDA Forest Service and USDI Bureau
of Land Management, Washington, D.C.

1994b. Environmental assessment for the implementation of interim strategies
(PACFISH) for managing anadromous fish-producing watersheds in eastern Oregon and
Washington, Idaho, and portions of California. USDA Forest Service and USDI Bureau of Land
Management, Washington, D.C.

1994c. Interior Columbia River Basin ecosystem management framework and scientific
assessment and Eastside Oregon and Washington Environmental Impact Statement. Eastside
Ecosystem Management Project Charter. Eastside Ecosystem Management Project, Walla
Walla, WA.

USDC. 1988. National Marine Fisheries Service, Alaska Region, policy for riparian habitat
protection in Alaska. NOAA, NMFS, Alaska Region, Juneau, AK.

1990. Fisheries of the United States, 1989. U.S. Department of Commerce, NOAA,
NMFS, Fisheries Statistics Division. Silver Spring, MD.

1995. Proposed recovery plan for Snake River salmon. U.S. Department of
Commerce, NOAA, NMFS, Northwest Region, Seattle, WA.

Vannote, R. L., G. W. Minshall, K. W. Cummins, J. R. Sedell, and C. E. Cushing. 1980.
The river continuum concept. Canadian Journal of Fisheries and Aquatic Sciences 37:130-137.

156 References

Warren, C. E., J. H. Wales, G. E. Davis, and P. Doudoroff. 1964. Trout production in an
experimental stream enriched with sucrose. Journal of Wildlife Management 28:617-660.

Washington Forest Practices Board. 1993. Board manual: Standard methodology for conducting
watershed analysis under Chapter 222-22 WAC. Washington Forest Practices Board, Olympia,
WA.

Weigand, J. E 1994. Economic issues in ecosystem restoration. Pages 6-12 in R. L. Everett
(editor) Eastside forest ecosystem health assessment. Volume IV: Restoration of stressed sites,
and processes. General Technical Report PNW-GTR-330. USDA Forest Service, Pacific
Northwest Research Station, Portland, OR.

Wernple, B. 1994. Hydrologic integration of forest roads with stream networks in two basins,
western Cascades, Oregon. MS Thesis. Oregon State University, Corvallis, OR.

Wilford, D. J. 1984. The sediment-storage function of large organic debris at the base of
unstable slopes. Pages 115-119 in W. R. Meehan, T. R. Merrell, Jr., and T. A. Hanley
(editors) Proceedings: fish and wildlife relationships in old-growth forests. American Institute
of Fisheries Research Biologists, Asheville, NC.

Williams, J. E. 1993. Developing support and participation for watershed restoration. In M.
D. Purser (editor) Proceedings, Watershed and stream restoration workshop: shared
responsibilities for shared watershed resources, August 26-27, 1993, Portland OR, not paginated.
American Fisheries Society, Oregon Chapter, Portland.

Wilzbach, M. A. 1985. Relative roles of food abundance and cover in determining the habitat
distribution of stream-dwelling cutthroat trout (Salmo clarki). Canadian Journal of Fisheries and
Aquatic Sciences 42:1668-1672.

Wissmar, R. C., J. E. Smith, B. A. McIntosh, H. W. Li, G. H. Reeves, and J. R. Sedell.
1994. Ecological health of river basins in forested regions of eastern Washington and Oregon.
General Technical Report PNW-GTR-326. USDA Forest Service, Pacific Northwest Research
Station, Portland, OR.

Ziemer, R. R., and D. N. Swanston. 1977. Root strength changes after logging in southeast
Alaska. Research Note PNW-306. USDA Forest Service, Pacific Northwest Research Station,
Portland, OR.

OU.S. GOVERNKENT PRINTING OFFICE: 1995-404-445/25610

OTHER TITLES IN THE
DECISION ANALYSIS SERIES

No. 1. Able, Kenneth W. and Susan C. Kaiser. 1994. Synthesis of Summer Flounder Habitat
Parameters.

No. 2. Matthews, Geoffrey A. and Thomas J. Minello. 1994. Technology and Success in
Restoration, Creation, and Enhancement of Spartina alterniflora Marshes in the United States. 2
vols.

No. 3. Collins, Elaine V., Maureen Woods, Isobel C. Sheifer and Janice Beattie. 1994.
Bibliography of Synthesis Documents on Selected Coastal Ocean Topics.

No. 4. Hinga, Kenneth R., Heeseon Jeon and Nodlle F. Lewis. 1995. Marine Eutrophication
Review.

No. 5. Lipton, Douglas W., Katherine Wellman, Isobel C. Sheifer and Rodney F. Weiher. 1995.
Economic Valuation of Natural Resources: A Handbook for Coastal Resource Policy Makers.

No. 6. Vestal, Barbara, Alison Rieser et al. 1995. Methodologies and Mechanisms for
Management of Cumulative Coastal Environmental Impacts. Part I -- Synthesis, with Annotated
Bibliography; Part 11 Development and Application of a Cumulative Impacts Assessment
Protocol.

ATM

2
YEARS or
SCIIENCE &
5SERVICI
Aw.. --OFCOWAERCE_
#40 Osp *,9?/C,

3 6668 00003 8598