2006 Practitioners Workshop - Pacific Northwest Aquatic Monitoring

Transcription

The Pacific Northwest Aquatic Monitoring Partnership (PNAMP)
&
The Collaborative Systemwide Monitoring & Evaluation Project (CSMEP)
Pacific Northwest Aquatic
Monitoring Partnership
SECOND ANNUAL RESEARCH, MONITORING AND EVALUATION
(RME) WORKSHOP FOR DECISION-MAKERS, PROGRAM
MANAGERS, SCIENTISTS AND FIELD PRACTITIONERS
March 16-17, 2006
Workshop Preparatory Information:
Brief Summaries from Current Monitoring Programs in the Pacific Northwest
Ambridge Event Center
300 NE Multnomah St
Portland, OR 97232
Invitees:
All Pacific Northwest RME Practitioners, Program Managers &
Decision-Makers RME Coordination Groups such as:
Collaborative Systemwide Monitoring and Evaluation Project
Columbia River Fish and Wildlife Authority
Federal Caucus
NOAA Fisheries Recovery Division
Northwest Power and Conservation Council
Oregon Plan Monitoring Team
Oregon Watershed Enhancement Board
Pacific Northwest Aquatic Monitoring Partnership
Technical Recovery and Domain Teams
Washington Governor’s Forum on Monitoring
Washington Salmon Recovery Funding Board
i
TABLE OF CONTENTS
Agenda…………………….……………………………………………………………..………1
Workshop Objectives…….………………………………………………….……………….….4
Workshop Rationale & Products……………………………………………..………………….5
Pacific Northwest Aquatic Monitoring Partnership (PNAMP)………………..…………..…....6
Collaborative Systemwide Monitoring and Evaluation Project (CSMEP)……....……….……11
CSMEP - Status and Trends Subgroup………………………………….………..….…….…..32
.
CSMEP – Harvest Action Effectiveness Subgroup ………………………………..……….…46
CSMEP – Hydrosystem Action Effectiveness Subgroup.…………………………..…………58
CSMEP – Habitat Action Effectiveness Subgroup………………………….…….……..…….74
CSMEP – Hatchery Action Effectiveness Subgroup……………………………….……....….85
King County Long-Term Water Quality Monitoring Program…………………….………......92
Northwest Power and Conservation Council DRAFT RME Program (excerpt)…...……..…...98
Washington Governor’s Forum on Monitoring.…………………………………….….……..111
California Coastal Monitoring Program……….…………………………………….……......114
Integrated Status and Effectiveness Monitoring Program (ISEMP) John Day
Pilot Project…………………………………….……………………………………..………118
Integrated Status and Effectiveness Monitoring Project (ISEMP) Salmon River
Pilot Project.…………………………………….………………………………….………..130
Federal Caucus RM&E Program.……………………………………………………..…........140
Monitoring Strategy from the Oregon Plan for Salmon and Watersheds.……….……..…….145
Aquatic and Riparian Effectiveness Monitoring Program (AREMP) Overview………..........151
Ecosystem Management Decision Support Intro……………..…………………………..….155
PACFISH/INFISH Biological Opinion (PIBO): Effectiveness Monitoring
Program (excerpt)……..………………………………………………………….………....157
ii
ii
Workshop Agenda
Thursday, March 16th (9:00- 5:00)
Time
9:00
Topic
Welcome & Introduction
9:10-9:35
Plenary Presentation: The Role of Coordinated Research,
Monitoring and Evaluation in Aquatic Monitoring for the Pacific
Northwest
9:35-10:00
Plenary Presentation: Preliminary results of CSMEP & PNAMP
surveys of regional entities’ relative priorities for addressing
different management questions
10:00-10:30
Policy Panel Responses:
• What’s your perspective on issues of coordination?
• What’s your perspective on the survey results?
• Where do you think the Pacific Northwest needs to head in
terms of RME priorities?
• What guidance do you have for technical participants at this
workshop regarding RME priorities?
10:30-10:45
10:45-11:15
BREAK
Plenary Presentation: Insights from scientists.
• Do the survey results align with what you think policy
makers need to hear?
• What things do scientists need to know from policy makers
to design cost-effective, reliable RME?
Moderated General Discussion:
• Comments from panelists on scientists’ insights
Questions from the audience
11:15-12:
00
Proposed
Speaker(s)
Frank Young
CSMEP/CBFWA
& Jen Bayer
PNAMP/USGS
Angus Duncan,
Bonneville
Environmental
Foundation
Jim Geiselman,
BPA and Claire
McGrath,
University of Idaho
Rob Walton,
NOAA; Bill
Towey, Colville
Tribes; Bruce
Crawford, WA
Governor’s Forum
on Monitoring
Jeff Rodgers,
ODFW
All
LUNCH (provided on site)
1:00-3:00
Concurrent Technical Session 1 – Monitoring the Status and
Trend of Fish and Habitat
Breakout groups
with facilitator
and recorder
Workgroups: 1A) Fish in Watersheds; 1B) Habitat Conditions in
Watersheds; 1C) Fish & Habitat in Large Rivers with Dams; and
1D) Fish & Habitat in Estuary/Nearshore Areas
1
Time
3:00-3:20
3:20-4:20
4:20-5:00
5:00
Topic
Proposed
Speaker(s)
Basic Theme: Given what decision-makers need to know about the
status and trends of fish and habitat, how do we best provide that
information? What are the major challenges, and how do we build
on existing work to overcome them?
BREAK
Continue Technical Session I
Summarize Workgroup Conclusions
End DAY 1 – Dinner on your own. Workgroup facilitators meet
to consolidate results of Technical Session 1 and review plans for
Day 2, including composition of workgroups.
Friday, March 17th (8:30 – 4:00)
Time
8:15-11:00
Topic
Concurrent Technical Session 2 – Action Effectiveness
Monitoring
Workgroup
Composition
Break-out groups
with facilitator
and recorder
Workgroups: 2A) Habitat Actions; 2B) Harvest Actions; 2C)
Hatchery Actions; and 2D) Hydro Actions.
11:00 –
11:15
11:15 to
noon
Basic Theme: Given what decision-makers need to know about the
effectiveness of habitat, harvest, hatchery and hydro actions, how do
we best provide that information, building on a foundation of status
and trend RME? What are the major challenges, and how do we
build on existing work to overcome them?
BREAK
Concurrent Technical Session 3 – Synthesis and Integration
Workgroups: Multiple breakout groups address the same
questions, with membership structured to afford a diversity of
perspectives and expertise, (i.e. each session 3 workgroup should
have at least one member from workgroups 1A, 1B, 1C, 1D, 2A, 2B,
2C, and 2D).
Break-out groups
with facilitator
and recorder
Basic Theme: In the context of what decision-makers need to
know, what are key insights from attempts to develop integrated
RME programs including both Status & Trend and Action
Effectiveness RME? What are the major challenges, and how do we
overcome them? What are the recommendations for PNAMP and
CSMEP for next 1-2 years?
2
12:00-1:00
1:00-2:30
2:30-2:45
2:45-4:00
4:00-5:00
LUNCH
(provided on site, groups can continue discussion over lunch)
Continue Concurrent Technical Session 3 – Synthesis and
Integration
BREAK
Closing Plenary Session—Session reports.
• Report out of Technical Session 3 on Synthesis and Integration
• Brief summary of workshop results
o Progress report on survey results
o Status of on-the-ground work
o Recommended strategies
• Next Steps
o Completion of workshop products
o Communicate findings to PNAMP, CSMEP, and others
to coordinate implementation of recommendations
o Plan for Third Annual Workshop
Meeting Adjourns
Facilitators and Recorders Meet to Discuss Reporting
3
Workshop Objectives:
1. Share, review and discuss results of recent surveys conducted by CSMEP and
PNAMP of the relative importance of different resource management questions and
information needs in the Columbia Basin (CSMEP) and Pacific Northwest (PNAMP).
2. Share, review and discuss current advances in RME (e.g., indicators, analytical
approaches to evaluation, sampling designs, monitoring protocols, integrated M&E
programs).
3. Share, review and discuss on-the-ground implementation issues from 2004 and
2005 field seasons with an eye towards learning aimed at increased scientific rigor,
efficiency and standardization of efforts and approaches and how to improve
coordination across regional monitoring efforts.
4. Host concurrent technical work sessions to assess current monitoring activities,
approaches and methods, gaps, and critical problems associated with status and trend
RME, action effectiveness RME, and their integration.
5. Provide feedback and recommendations to PNAMP and CSMEP and the entities
involved in these groups.
Decision
Makers &
Program
Managers
Management Decisions: e.g. Are stocks recovering enough to de-list?
How should we manage harvest, hydrosystem and hatcheries? What
habitat restoration actions should we implement, where and when?
How sure do we need to be when making each of these decisions?
How much can we spend on RME?
Quantitative
Scientists
Evaluation Designs: What type of data and data analyses do we
need to clarify management decisions? At what scales? What level of
confidence and precision do we need for each type of decision?
Biometricians
Sampling Designs: Where and when should we sample to get data at
the spatial and temporal scales required for management decisions?
Field Biologists
Monitoring Protocols: How should we sample at those places and
times to make the most cost effective use of available resources, and
meet target levels of precision and confidence across regional scales?
Figure 1. The workshop is intended to stimulate dialogue and integration across all
necessary RME components and people. The three scientific roles shown at left
(quantitative scientists, biometricians, field biologist) need to be involved jointly in all
4
three of the associated RME components (evaluation designs, sampling designs and
monitoring protocols), which serve the needs of decision-makers and program managers.
Rationale for the Workshop and its Structure: Across the Pacific Northwest, many
federal, state, tribal, local and other entities require RME to make sound decisions on the
management of fish populations and their habitats (Figure 1). While the mandates,
regulatory drivers and information priorities of these entities understandably differ, there
are also many common elements and associated information requirements. The ‘RME
toolbox’ of concepts, designs, methods, recommendations and plans is growing steadily.
Interests in monitoring select appropriate elements from the RME toolbox to meet their
specific needs. At the same time, the Pacific Northwest as a whole could benefit from
maximizing the consistency of RME approaches so that results can be meaningfully
aggregated to address monitoring questions and policy needs at multiple scales. This
workshop is intended to stimulate dialogue on common management needs, clarify recent
RME advances, make progress on how to best apply available RME approaches toward
identified needs, and discuss next steps that will best improve regional coordination, as
well as acceptance and eventual adoption of reliable, relevant, cost-effective, and
consistent RME approaches.
For example, the Pacific Northwest Aquatic Monitoring Partnership (PNAMP) and the
Collaborative Systemwide Monitoring and Evaluation Project (CSMEP) endeavor to
provide tools and recommendations in support of programs to meet multiple aquatic
resource monitoring, data management and analysis needs across the Pacific Northwest.
These two groups have recently conducted surveys of resource management entities on
their RME priorities, and will present preliminary results from these surveys at the
workshop.
Workshop Products:
1. Report preliminary results of surveys of resource management agencies’ ranking of
management questions
2. Report on the current status of technical products and on the ground implementation
for status and trend RME, action effectiveness RME and data management systems
development.
3. Recommend strategies for improving dialogue to maximize the relevance and rigor of
RME, as well as coordinating future work on regional RME activities and implementation
processes to maximize the regional consistency of RME programs and activities.
Planning Committee: Dave Marmorek (CSMEP; ESSA Technologies), Jennifer Bayer
(PNAMP; USGS), Keith Wolf (KWA/Colville Tribes), Frank Young (CBFWA), Jim
Geiselman (BPA), Russell Scranton (NOAA Fisheries), Steve Leider (WA GSRO), Steve
Waste (NPCC), Scott Downie (CDFG), Jennifer O’Neal (Tetra Tech EC)
5
Pacific Northwest Aquatic
Monitoring Partnership
The purpose of the Pacific Northwest Aquatic Monitoring Partnership (PNAMP) is to
provide a forum for coordinating state, federal, and tribal aquatic habitat and salmonid
monitoring programs.
Improved communication, shared resources and data, and compatible monitoring efforts
provide increased scientific credibility, cost-effective use of limited funds and greater
accountability to stakeholders.
PNAMP provides leadership through the development and the advancement of
recommendations and agency level agreements that are considered for adoption by the
participating agencies.
PNAMP has adopted the following goals:
•
•
•
•
•
Improve communication between monitoring programs across state, tribal, and
federal organizations.
Improve scientific information needed to inform resource policy and management
questions and decisions.
Seek efficiencies and cost-effectiveness across monitoring programs through
compatible and cooperative monitoring efforts.
Promote science-based credibility of monitoring and assessment efforts.
Share resources and information between monitoring programs across state, tribal,
and federal organizations.
In addition to adopting a monitoring coordination structure with Steering Committee
guidance, PNAMP has identified and developed workgroups for five key elements of
monitoring: watershed condition monitoring, effectiveness monitoring, fish population
monitoring, estuary monitoring, and data management.
PNAMP has adopted a Charter to formalize the agreement among federal, state, and tribal
entities to participate in the coordination of scientific monitoring programs; 19 entities are
signatory to the Charter to date. PNAMP has also developed a coordination plan to
facilitate aquatic monitoring in the Pacific Northwest titled, “Strategy for Coordinating
Monitoring of Aquatic Environments in the Pacific Northwest.” These documents along
with the current year’s work plan can be accessed at
www.reo.gov/PNAMP.
6
FY2006 initiatives
Summary of PNAMP 2006 work tasks
• Conduct Monitoring Activity Inventory to facilitate coordination by
identifying where aquatic monitoring activities occur in the region and what
protocols are in use. Due summer 2006.
• Complete Side-by-side Protocol Comparison Test for comparison of
watershed monitoring protocols. Due May 2006.
• Complete universal status monitoring design template. Due fall 2006.
• Complete the publication of recommended procedures for the application
of common methods of capturing and counting salmonids and identify gaps
and protocols needing more formal comparison. Due April 2006.
• Develop a standardized field method training manual format. Template to
be used in training and institutionalizing standardized “field methods.”
Standardized format will ensure consistent implementation of recommended
protocols and clarify the definitions of key indicators and variables. Due
June 2006.
• Develop a coordinated approach to telemetry, tagging and marking
juvenile migrants and adult fishes. Statistical design, cost, and feasibility
will dominate the FY06 work. Due fall 2006.
• Develop standard data dictionary for documenting data collection
protocols and metadata. Due fall 2006.
• Implement, publish and publicize strategy for placing intensively
monitored watersheds (IMWs) throughout the Pacific Northwest to monitor
and evaluate cause and affect relationships between habitat restoration
actions and changes in fish abundance. Plan is complete. Need to define
next steps for implementation.
• Develop a regional strategy and recommendations for testing effectiveness
of habitat restoration projects at the reach scale using acceptable science
standards. Due summer 2006.
• Develop a list of habitat restoration project categories and protocols based
on the results of the project effectiveness monitoring testing work above.
Due fall 2006.
• Inventory existing habitat restoration projects across the region with
ongoing monitoring meeting recommended experimental designs. Due fall
2006
7
• Conduct Management Questions survey and follow up Practitioners’
Workshop. Due March 2006.
• Recommend core set of high-level indicators that could be used on a
regional scale to describe landscape level response to management. Due
March 2006.
• Conduct workshops for the collective evaluation and interpretation of
current and new knowledge regarding monitoring issues. Planned for FY06:
Remote Sensing II April 2006, Monitoring Practitioner’s Workshop March
2006, Monitoring for Aquatic Invasive Species TBD
Next Steps:
• PNAMP is seeking partners to become signatory to the Charter as well
as additional technical participants for all topics.
• Emerging technical topics for PNAMP include estuary monitoring and
large river monitoring.
The PNAMP annual workplan in detail may be found at
www.reo.gov/PNAMP or by contacting Jennifer Bayer at
[email protected].
Pacific Northwest Aquatic Monitoring Partnership (PNAMP) Members
Tribal Entities
Columbia River Intertribal Fish Commission
Confederated Tribes of the Colville Reservation
Northwest Indian Fisheries Commission
State Agencies
California Department of Fish and Game
Oregon Watershed Enhancement Board
Washington Interagency Committee for Outdoor Recreation
Washington Department of Ecology
Washington Governor’s Salmon Recovery Office
Federal Agencies
Bonneville Power Administration
National Oceanic and Atmospheric Administration Fisheries
U.S. Army Corps of Engineers
US. Bureau of Land Management
U.S. Bureau of Reclamation
U.S. Environmental Protection Agency
U.S. Forest Service
U.S. Geological Survey
Regional Entities
Columbia Basin Fish and Wildlife Authority
Pacific States Marine Fisheries Commission
8
Pacific Northwest Side-by-Side Protocol Comparison Test
Steve Lanigan – Watershed Monitoring Workgroup Lead
3/16/06
Summary
Eleven state, tribal, and federal agencies participated during summer 2005 in a side-by-side comparison
of protocols used to measure common in-stream physical attributes to help determine which protocols
are best for determining status and trend of stream/watershed condition. This protocol comparison was
sponsored by the Pacific Northwest Aquatic Monitoring Partnership as part of an ongoing effort to
enable different agencies to be able to share data and determine best measurement techniques. Field
sites were located in the John Day Basin, eastern Oregon, in mountain channels that provide critical
habitat for threatened and endangered salmonids. Twelve streams were examined, representing a range
of alluvial channel types (pool-riffle, plane-bed, and step-pool) and a range of channel/habitat
complexity (simple, free-formed channels vs. complex wood-forced ones). Study sites had bankfull
channel widths of 3-15 m, slopes of about 1-7%, and median substrate sizes of 9-154 mm. Channel
features of interest included reach-average width, depth, gradient, sinuosity, substrate characteristics
(median size, percent fines), wood characteristics (number, size), pool characteristics (residual depth,
area of pools), and channel entrenchment. Field crews from the USDA Forest Service Rocky Mountain
Research Station determined “true” channel characteristics using a dense array of cross sections spaced
every half bankfull width over stream lengths of 40-80 bankfull widths. A total station was used for
surveying channel cross sections and the longitudinal profile of the stream bed, while Wolman pebble
counts were used to sample substrate at each cross section. The sites were then inventoried by 1 to 3
field crews from 11 participating agencies using their measurement protocols. Preliminary results
indicate that some monitoring group protocols performed better than others for particular attributes, no
one monitoring group’s protocols performed best for a majority of the attributes compared.
Monitoring groups that provided survey crews:
Aquatic and Riparian Effectiveness Monitoring Program
California Fish and Game
Colville Confederated Tribes
Oregon Department of Environmental Quality
Oregon Department of Fish and Wildlife
Northwest Indian Fisheries Commission Timber, Fish, and Wildlife Monitoring Program
PacFish/InFish Biological Opinion Monitoring Program
Upper Columbia River Monitoring Program
USDA Forest Service Level II Stream Survey Program
USDA Forest Service, Rocky Mountain Research Station
US Environmental Protection Agency Environmental Monitoring and Assessment Program
Washington Department of Ecology
9
PNAMP Contact Information
PNAMP Coordinator:
Jennifer Bayer
509.538.2299 x 273
[email protected]
PNAMP Assistant
Debra Niemann
503 808-2267
[email protected]
Workgroup Leads:
•
Watershed Condition Monitoring
Steve Lanigan
503.808.2261
[email protected]
•
Effectiveness Monitoring
Bruce Crawford
360.902.2956
[email protected]
•
Fish Population Monitoring
Keith Wolf
425.788.3402
[email protected]
Jennifer O’Neal
425.482.7779
[email protected]
Data Management
Stewart Toshach
206 860-3495
[email protected]
Joy Paulus
360 902-2954
[email protected]
Estuary Monitoring
Julia Bos
360 407-6674
[email protected]
Cathy Tortorici
503 231-6268
[email protected]
•
•
10
CSMEP - Collaborative Systemwide Monitoring & Evaluation Project
The following material is extracted from the recent CSMEP proposal for FY07-09, focusing on the
issue of RME coordination. The complete CSMEP proposal is available from the CBFWA
website.
A. Overview
The Collaborative Systemwide Monitoring and Evaluation Project (CSMEP) is a
coordinated multi-agency effort to improve the quality and consistency of fish population
and habitat data to answer key monitoring and evaluation (M&E) questions relevant to
major decisions in the Columbia River Basin. CSMEP has made considerable progress in
improving access to subbasin data as well as in the collaborative design of improved M&E
methods. Data work products include metadata inventories of fish data for subbasins in
Washington, Oregon and Idaho, rigorous assessments of the strengths and weaknesses of
these data for addressing key questions about fish populations, and web-accessible
databases for both the metadata and assessments. Design work products have been
developed through a rigorous Data Quality Objectives process. This process has generated
sampling, response and evaluation designs which improve the reliability of management
decisions related to the status and trends of fish populations and to the evaluation of the
effectiveness of habitat, harvest, hatchery and hydrosystem recovery actions, building on
the subbasin data assessments. CSMEP’s design work to date has focused on a pilot project
for Idaho’s Snake River Basin, which has fed into both the NOAA-F/BPA Salmon River
Basin Pilot Study and the Lemhi River HCP. CSMEP is evaluating the tradeoffs associated
with alternative designs for the Snake Pilot, and proposes additional pilot projects in
Oregon and Washington, catalyzing implementation of improved M&E throughout the
Columbia Basin. CSMEP and StreamNet will continue metadata inventories for additional
Columbia River subbbasins, using these results to test the applicability of pilot project
monitoring designs to salmon, steelhead, bull trout and other resident fish species of
concern. CSMEP will continue to collaborate with PNAMP and other RME entities to
ensure that CSMEP’s analytical expertise is effectively utilized within ongoing monitoring
programs.
D. Relationship of CSMEP to other projects
CSMEP has worked very closely with StreamNet on data inventory and data management issues,
and several of the work elements we describe in section F depend on funding for StreamNet from a
separate contract (Table F1). CSMEP has also coordinated closely with the State of Salmon
initiative to inventory salmon data sets across the Pacific Rim, encouraging them to build on the
inventory work that CSMEP has completed.
Subbasin Planning
Subbasin plans developed for the NWPCC Fish and Wildlife Program vary widely in the
scope of their aquatic research, monitoring and evaluation (RM&E) components. Many
subbasin plans have concentrated on a ‘bottom-up’ approach, in accordance with the
11
initiative provided in the Technical Guidance for Subbasin Planners (NWPCC 2001)
which treats M&E largely at the project scale e.g., in support of individual habitat projects.
However, a number of the subbasin plans have moved beyond this (e.g. Salmon, John Day,
Grande Ronde, etc.) and are taking a more ‘top-down’ approach to coordinate RM&E
efforts at the regional or programmatic level. These plans have recognized that ‘bottom-up’
M&E undertaken within the subbasins will have a higher likelihood of generating
meaningful results if they reflect regional scale M&E strategies. This is more consistent
with the current NWPCC guidance to move M&E from project to regional and
programmatic scales.
Approaches being developed within the federal pilot projects and a suite of comprehensive
state and tribal monitoring initiatives allow broader integration and synthesis of M&E
information. General guidelines required to develop this ‘top-down’ RM&E framework are
evolving through the Pacific Northwest Aquatic Monitoring Partnership (PNAMP).
CSMEP is working with PNAMP to resolve the multitude of technical design elements
required to make this framework a reality, and allow effective integration of subbasin plan
monitoring into the broader regional framework. CSMEP’s ongoing Snake Basin Pilot
Project overlaps with five Columbia subbasins (Salmon, Grande Ronde, Imnaha, Asotin
and Clearwater), providing an initial test opportunity to employ information from existing
subbasin plans to help develop broader integrated M&E designs for Status & Trends and 4
H effectiveness monitoring across multiple subbasins.
NOAA Fisheries and Action Agencies Pilot RME Projects
CSMEP participants include many key individuals involved in planning/coordination of
the Wenatchee, John Day and Upper Salmon federal Pilot Projects. These individuals are
providing insights from development and implementation of these pilot projects that
directly affect CSMEP M&E designs. The Wenatchee Pilot Project is an experimental
application of the Upper Columbia Monitoring Strategy that provides CSMEP analysts a
working example of an integrated status/trends and effectiveness monitoring program at
the subbasin scale. The Wenatchee pilot is providing opportunities to determine the most
powerful indicators for effectiveness monitoring at a range of spatial and temporal scales,
as well as representing a real-world example of use of EMAP designs for status and trends
monitoring (i.e., interaction between the theoretical and the field application). The John
Day Pilot Project provides an opportunity to explore alternative statistical designs for
effectiveness monitoring and is a unique testing ground for the use of remote sensed
approaches (e.g. GIS analyses employing FLIR, LIDAR, TMDL models, etc.) to monitor
habitat changes at multiple spatial scales resulting from mitigation actions. The Upper
Salmon Pilot Project (to which CSMEP has been directly contributing) provides an
opportunity to explore approaches for both status and trends monitoring (SF Salmon River)
and effectiveness monitoring (Lemhi River). The SF Salmon work is not only exploring
different methods and designs for status/trends monitoring but is also exploring how to
integrate M&E programs across species, across agencies, and across monitoring tiers. The
effectiveness monitoring in the Lemhi is intended to inform how monitoring data collected
across subbasin restoration projects can best be linked into adaptive management planning.
12
PNAMP and the Federal RME Program
In the last four years an increasing number of entities have become engaged in work related to
monitoring and evaluation. Hence, definition of CSMEP’s niche, as distinct from such entities as
PNAMP and the Federal RME Plan, is very important to avoid confusion or duplication of effort.
Table D1 outlines the various policy, programmatic and technical roles involved in developing and
implementing monitoring programs. CSMEP operates entirely in the technical domain, but
interacting with programmatic and occasionally policy levels. By contrast, the Federal RME
Program and PNAMP have decision making authority at both the policy and programmatic levels.
The PNAMP charter offers a tremendous opportunity for distributing CSMEP work products, and
obtaining interagency buy-in across the Pacific Northwest for consistent and effective monitoring.
Federal RME and PNAMP scientists also work on technical products (e.g. PNAMP review of fish
monitoring protocols, tests of habitat monitoring protocols, watershed condition work). The
participating and charter entities in PNAMP have a strong interest in land management (Figure
D1), and have therefore taken the lead on habitat monitoring technical work, whereas CSMEP has
focused on fish monitoring. While the overarching goals of PNAMP and CSMEP are similar
(Table D2), there are considerable differences in the work elements and products incorporated into
annual and quarterly work plans. To avoid duplication of effort, all of CSMEP’s proposals and
work plans, including this proposal, are closely coordinated with PNAMP and the Federal RME
Plan. Methods of coordination include: overlapping membership of technical fisheries scientists
across the three entities; coordinated development of quarterly work plans; annual workshops for
presentation of results; and joint technical meetings. These 4 steps ensure no duplication of effort
in work products, despite an obvious and healthy overlap in goals. All of these groups are working
towards common objectives; there’s more than enough work to go around.
13
Table D.1 Definition of policy, programmatic and technical roles in developing and
implementing monitoring programs. The second row lists examples of entities fulfilling
each of these roles (not comprehensive). CSMEP is focused on the tasks in the technical
column, as well as interactions with programmatic entities. The roles and tasks are not
listed in order of priority. Source: Adapted from CSMEP FY04 Work Plan.
Policy / Leadership
PNAMP Executives; Federal RME
Program; WA Governor’s Forum on
Monitoring; Senior Policy Levels of
federal, state, and tribal fish and wildlife
agencies; Northwest Power and
Conservation Council
•
Identify and prioritize management
decisions and associated information
needs.
•
Make well-informed decisions about
M&E issues, acting within purview
of each agency, but with knowledge
of all entities’ needs)
•
Do reality check on what is
achievable/realistic; scope
•
Set goals driven by outcomes (not
too proscriptive).
•
Assure consistency in methods used
to evaluate success of M&E
•
Identify and secure appropriate
sources of funding.
•
Perform conflict resolution and
make final decision for issues
elevated from programmatic level.
•
Formalize/endorse programmatic
level agreements.
⇔
⇔
⇔
Programmatic
Program managers within
federal, state and tribal fish and
wildlife agencies, including
PNAMP agencies, Federal RME
Program
•
Provide guidance on
management priorities to
technical level based on
dialogue with policy level
•
Select and implement
preferred sampling designs
and monitoring protocols
based on technical level
evaluations of options.
•
Define population
management units and
scales of interest for
monitoring information.
•
Identify RME issues
requiring management
decisions, e.g.
- Performance metrics
- Action effectiveness
hypotheses
- Critical uncertainties to
be evaluated
•
Oversee timely management of
programmatic group’s deliverables
•
•
Assess ongoing work for
gaps.
Ensure implementation of accepted
sampling designs and monitoring
protocols, maximizing consistency
while recognizing agency
jurisdictions
•
Define options for
scope/resource
management.
•
Do project management.
•
Establish peer review
protocol.
⇔
⇔
Technical
CSMEP fish biometricians
and scientists within federal,
state and tribal fish and
wildlife agencies; PNAMP,
federal, state and tribal
scientists
1.
⇔
2.
3.
4.
5.
6.
7.
Coordinate with other
M&E programmatic
and technical entities.
Catalog existing work.
Make datasets available
to others.
Assess strengths and
weaknesses of existing
data.
Explore and evaluate
improved monitoring
protocols and sampling
designs for
consideration at
Programmatic Level.
Implement sampling
design and collect data
following
Programmatic Level
approval.
Evaluate monitoring
program results;
perform data analyses
for programmatic team
interpretation
14
Figure D1. Relationship of CSMEP to PNAMP partner agencies.
NOAA Pacific Coast
Salmon Recovery Fund
Program
Federal Caucus
RME Program
NPPC and BPA
Columbia Basin
F&W Program
Oregon Water Enhancement
Board Monitoring Program
USFS and BLM
Monitoring Programs
USFWS Bull Trout
RME Program
PNAMP
Pacific NW Aquatic
Monitoring
Partnership
EPA National and State Level
Monitoring Programs
California NW Forest Plan
Monitoring Program
Tribal Monitoring
Programs
Lower Columbia River
Estuary Program
BOR Habitat Monitoring
Program
UC, JD, Salmon River, and
Columbia Estuary Pilot
Projects Intensively
Monitored Watersheds
Washington Salmon Recovery
Fund Monitoring Program
CSMEP Collaborative
Systemwide Monitoring
and Evaluation Project
Watershed Condition
and Fish Monitoring
Protocols Test
Projects
15
Table D2. Comparison of PNAMP, CSMEP, and FRMEP. Differentiation of M&E niches occurs through development of distinct work plans and products.
Attribute
Pacific Northwest Aquatic Monitoring
Partnership (PNAMP)
Collaborative Systemwide Monitoring and
Evaluation Project (CSMEP)
Federal RME Plan (FRMEP)
Foundation of
Initiative
Federal Caucus RME Coordination needs
and a request by Governors of WA, OR,
ID, and Montana to develop a regionally
coordinated monitoring system. Prior
AREMP effort under the North-west
Forest Plan for monitoring on Federal
lands.
Northern CA, WA, OR, ID
All fish species?
Response to NMFS/USFWS document (Feb
2002) describing needs for M&E.
Recommended for funding by ISRP,
NWPCC, CBFWA, BPA for FY04-f06 under
Mainstem/Systemwide Solicitation.
Federal Caucus All-H Salmon Recovery
Strategy and NOAA 2000 Biological Opinion
on FCRPS (Being updated in 2006 to include
Recovery Plan RM&E, 2004 FCRPS BiOp
and Remand Process)
Columbia Basin
FY04-06: Salmon, steelhead, bull trout FY0709: Salmon, steelhead, bull trout and other
resident fish species of concern
Scientists within CBFWA, WDFW, IDFG,
ODFW, NOAA, USFWS, CRITFC, Yakama,
Colville, Nez Perce, BPA, StreamNet; ESSA,
Eco Logical Research, Quantitative
Consultants, PER, KWA.
Columbia Basin
Salmon and steelhead (2006 update to include
bull trout)
Thorough inventory and assessment of
existing fish monitoring data. Rigorous
development of M&E designs to improve fish
monitoring information for important regional
decisions on 4 H’s and ESA listings.
Coordination with PNAMP and other entities
to encourage implementation of these designs.
Development and implementation of Federal
Caucus Agency RM&E programs that support
Columbia Basin Salmon and Bull Trout
Recovery and FCRPS Biological Opinions.
Includes regional coordination and integration
with other federal, state and tribal monitoring
programs.
Geographic
Scope &
Species /
Habitats
Composition
Primary Goals
Scientists, managers and executives
within USFS, BLM, NWPCC, FRMEP,
WA SRFB, OWEB, CA NW Forest Plan
Monitoring Program, CSMEP, BOR,
EPA , PCSRF, NED
Coordinate and improve consistency of
approaches and protocols used by
member entities for monitoring
watershed condition, fish populations and
project effectiveness. Share advances in
M&E among member entities, and
promote adoption of improved M&E
methods to inform resource management
decisions.
Scientists, managers, and executives within
the Federal Caucus including NOAA,
USFWS, USFS, BLM, EPA and Action
Agencies (BPA, BOR, A COE), and technical
consultants
16
Attribute
Partnership (PNAMP)
Organizational
obligations
Open participation, non-binding
commitment to coordinate
Funded entities agree to complete tasks
developed in annual and quarterly work plans.
Implementation of developed M&E designs
by member organizations is voluntary.
Statutory mandates of Federal Caucus
Agencies. All-H Salmon Recovery MOU.
Policy Level
Responsibilities
Executive partners provide policy
direction and support to PNAMP through
the Steering Committee
No policy responsibilities. But get policy
makers required input to M&E designs (e.g.
relative importance of different species,
questions, scales; risk tolerance; cost)
Decide on scope and funding of Federal
Caucus M&E needs.
Programmatic
Level
Responsibilities
No programmatic responsibilities. But
Measure progress toward recovery of ESAlisted anadromous fish populations.
Develop and Maintain a coordination and present M&E designs for consideration and
feedback to Programmatic Levels in PNAMP
management structure for PNAMP
Identify and prioritize actions that are the most
and CSMEP entities.
effective towards meeting fish population
Coordinate peer review of PNAMP
performance objectives.
technical products.
Implement RME identified in the Federal
Provide Executives with coordinated
Caucus RME Plan which includes Federal
programmatic approaches that
Recovery Plan needs, the Action Agencies
integrate watershed and fish
Annual Implementation Plans under the
monitoring, action effectiveness
FCRPS BiOp, and other Federal Agency
monitoring, and data management.
BiOps.
Select and implement preferred
sampling designs and monitoring
protocols in the fresh water aquatic
environment.
Coordinate regional efforts in watershed
status/trend monitoring, effectiveness
monitoring and data management.
Methods of coordination include: overlapping membership of technical fisheries scientists across the three entities; coordinated
development of quarterly work plans; annual workshops for presentation of results; and joint technical meetings. These 4 steps ensure no
duplication of effort in work products, despite an obvious overlap in goals.
Coordination of
Technical
Responsibilities
17
Attribute
Partnership (PNAMP)
Technical Level
Responsibilities
Coarse scale inventories of existing
watershed and fish population
monitoring.
Develop and adopt a standardized set of
reporting metrics, monitoring protocols
and sampling designs to assess watershed
condition and fish population status and
trends
Develop standardized regional fish
population monitoring efforts
Develop and adopt a standardized set of
fish population metrics and compatible
protocols for sampling designs and data
collection
Develop and implement pilot projects for
testing monitoring approaches.
Detailed inventory of existing fish monitoring
data; metadata available on Internet.
Assess strengths and weaknesses of existing
data for answering key questions related to
major decisions;
Develop improved M&E designs for making
important decisions, building on strengths of
existing M&E and results of ongoing pilot
projects;
Systematic evaluation of alternative M&E
designs across multiple objectives, species
and scales
Implement sampling and collect data
following Programmatic approval;
Evaluate monitoring program results and
perform data analyses for programmatic
levels.
Develop alternative M&E designs and
coordinated programs for meeting Federal
Caucus RME responsibilities including
tributary monitoring, habitat action
effectiveness, hydro action effectiveness, and
data management.
18
E. CSMEP Project history (Oct 2003 to Dec 2005)
CSMEP is focused on the issue of systemwide monitoring and evaluation of fish status,
addressing requirements of NOAA-F and USFWS biological opinions and recovery plans
as well as the NWPCC Fish and Wildlife Program. This has involved an integrated,
collaborative effort by fisheries scientists and biometricians within CSMEP to fulfill
seven objectives:
1. Interact with federal, state and tribal programmatic and technical entities
responsible for monitoring and evaluation of fish and wildlife, to ensure that
quarterly work plans developed and executed under this project are well
integrated with ongoing work by these entities.
2. Collaboratively inventory existing monitoring data that bear on the problem of
evaluating the status and trend of salmon, steelhead, bull trout and other species
of regional importance across the Columbia Basin, including the Okanagan Basin
in Canada.
3. Work with existing entities (e.g. StreamNet, NOAA Fisheries, NWPCC) to make
a subset of existing monitoring data available through the Internet, recognizing
the continuing evolution of data management in the Columbia Basin.
4. Critically assess the strengths and weaknesses of existing monitoring data and
associated evaluation methods for answering key questions at various spatial
scales concerning the state of ecosystems and fish habitat, as well as fish
distributions, stock status and responses to management actions.
5. Collaboratively design improved monitoring and evaluation methods that will fill
information gaps and provide better answers to these questions in the future, by
providing state and tribal fish agency participation and work products for multiagency development of regionally coordinated monitoring programs.
6. Coordinate state and tribal participation and work products for regionally
coordinated, multi-agency implementation of pilot projects or large scale
monitoring programs.
7. Participate in regional forums to evaluate new monitoring program results, assess
new ability to answer key questions, propose revisions to monitoring approaches,
and coordinate proposed changes with regional monitoring programs.
Progress on each of these seven objectives is described below:
E1. Objective 1: Communicate and coordinate monitoring and evaluation activities
Since project initiation CSMEP participants have collaboratively developed work plans in
close consultation with other programmatic and technical entities to ensure that analyses
and monitoring designs explored as part of the project are consistent with the overarching
objectives of Columbia Basin monitoring agencies. CSMEP representatives have also
regularly participated in PNAMP meetings and workshops, as well as meetings with
NOAA Habitat, NOAA-AA RME, EPA-EMAP, AREMP, OWEB and TRT groups. A
19
number of CSMEP participants are also PNAMP members. CSMEP also includes
members of the bull trout Recovery Monitoring and Evaluation Group (RMEG) (see
presentation and 5 page summary of RMEG activities), which ensures development of
consistent monitoring approaches for this listed species. In FY 2006 CSMEP will further
integrate with federal, state and tribal programmatic and technical entities responsible for
monitoring of fish and wildlife in the Columbia Basin. CSMEP quarterly workplans will
be developed in conjunction with BPA and PNAMP representatives to ensure that
CSMEP analyses synchronize with, supplement and support broader regional RME
needs. CSMEP products will be provided to the ISAB for review.
E2. Objectives 2 and 3: Inventory existing monitoring data/make it accessible
CSMEP subbasin inventories describe, in a systematic manner, the kinds of information
currently available on the abundance, productivity, spatial distribution and diversity of
salmon, steelhead and bulltrout. To evaluate the range of data quality that exists within
the Columbia Basin, CSMEP has selected pilot subbasins that included both data rich and
data poor areas. For each of these pilot subbasins, StreamNet staff and CSMEP biologists
jointly completed an inventory of the information available for each of the key
performance measures for each of the target fish species. In FY 2004 CSMEP, with the
assistance of StreamNet staff, completed inventories of existing fisheries monitoring data
for six pilot subbasins in Washington (Lewis, Yakima), Oregon (Lower Columbia,
Imnaha) and Idaho (South Fork Salmon River and Selway River drainages). During FY
2005, CSMEP biologists and StreamNet conducted detailed inventories of fish data for
seven additional pilot subbasins selected in Washington (Okanagan, Methow, Kalama),
Oregon (Deschutes, Grande Ronde) and Idaho (Middle Fork Salmon, Upper Fork
Salmon). The locations of these CSMEP pilot subbasins (as well as the Federal RME
pilot subbasins) are provided in an overview map on the CSMEP website.
The results of the Tier 2 subbasin metadata inventories (C1 tables) undertaken by
CMSEP to date are now served up through a web-based meta-database developed,
hosted, and maintained by the Oregon Department of Fish and Wildlife's Natural
Resource Information Management Program (NRIMP). As of the end of FY 2005 there
were more than 1450 fish inventory records on this CSMEP data server and there have
been over 36,000 hits on the web server to date. This CSMEP meta-database is proving
useful to fisheries biologists operating within the confines of the pilot subbasins
inventoried to date. It is also providing information to a broader range of analysts who
can query the growing CSMEP dataset to develop overviews and comparisons of the
varied monitoring techniques, design approaches, analytical assumptions etc. that are
being applied for M&E across the region. A recent StreamNet report outlines further
steps that could improve overall data management of the CSMEP inventories. CBFWA
has also developed a publicly accessible CSMEP Website for coordination amongst
CSMEP members, communication of CSMEP goals and products to a larger audience,
and the storage of important reference materials.
20
In FY 2006 CSMEP will undertake metadata inventories for new subbasins still to be
selected in the three member states (ID, OR, WA). StreamNet will continue to maintain
the CSMEP web data server and will work with CSMEP to improve data delivery by
developing expanded hyper-links to the data sources identified in the subbasin
inventories, as well as supporting materials identified in the QA/QC and Strengths and
Weaknesses assessments of those data (e.g., reports, databases, maps showing where data
collected).
E3. Objective 4: Assess the strengths and weaknesses of existing monitoring data.
Throughout FY 2004 and FY 2005 CSMEP biologists critically assessed the strengths
and weaknesses of each pilot subbasin’s meta-data and associated evaluation methods for
answering the key CSMEP questions (at various spatial scales) concerning the state of
ecosystems and fish habitat, as well as fish distributions, stock status and responses to
management actions (CSMEP B2 tables – see template). CSMEP biologists reviewed the
strengths and weaknesses of these data for addressing Tier 2 status and trend questions,
and considered opportunities for using these data to answer Tier 3 action effectiveness
questions (see Appendix 1). The strengths and weaknesses reviews (Table E.2)
completed to date are identifying areas where fish monitoring is being done well, in
addition to uncovering inferential weaknesses and data gaps that will be important to
address in CSMEP’s monitoring design work. A supporting document “Comparative
summary of the statistical and cost properties of different methods for estimating CSMEP
fish performance measures” was developed within CSMEP to support this task. Though
excellent fish population monitoring does exist in many subbasins, a common weakness
is the fact that sampling sites were not typically chosen through a rigorous process that
allows generalization to larger spatial scales. A preliminary overall synthesis of strengths
and weaknesses across the pilot subbasins is available on the CSMEP website. The
strengths and weaknesses overview tables for spring chinook and sockeye are also
available on the CSMEP website. This synthesis will be further developed in FY 2006
(i.e., are there strengths and weaknesses in regards to monitoring of particular fish
performance measures that are common across the subbasins?)
Table E.2. Data strengths and weaknesses analyses completed in FY 2004/05 by subbasin and
species (hyperlinked to the Table B2 summaries on the CSMEP website).
State
Subbasin
Species
Idaho
South Fork Salmon River
spring/summer chinook
Clearwater, Selway River
chinook (spring, summer)
steelhead (summer)
bull trout
Imhaha
chinook (spring)
steelhead (summer)
Lower Columbia
fall chinook
Lewis
chinook (spring, tule and bright fall)
steelhead (summer, winter)
Oregon
Washington
21
State
Subbasin
Species
Yakima
coho
fall chinook
spring chinook
steelhead (summer)
Methow
Chinook (spring, summer)
Steelhead (summer)
In FY 2006 CSMEP will complete the Strengths and Weaknesses assessments of all subbasins
inventoried in the three years of the project, and finalize the synthesis of the general “lessons
learned” as to the effectiveness of current M&E protocols across the Basin. This synthesis report
will describe the implications of the CSMEP Strengths and Weaknesses assessments for M&E
design in the Columbia River Basin, and will be reviewed by PNAMP and the ISAB.
E4. Objective 5: Collaboratively design improved M&E methods
Significant progress has been made on CSMEP’s goals of collaborative design of
improved M&E methods. Six multi-agency monitoring design workshops have been held
(three workshops in FY 2004 and three workshops in FY 2005) to explore how best to
integrate the most robust features of existing monitoring programs with new approaches
(e.g., Federal RME pilot studies, EPA EMAP). (see Design Workshop summaries: 07/0911/2004, 08/21-22/2004, 08/20-21/2005, all PowerPoint presentations from Design
Workshops are also available on the CSMEP Website). CSMEP is exploring the ability of
these approaches to answer the questions in Appendix E1, and is attempting to lay out a
structured approach to evaluating the costs, benefits and tradeoffs of different M&E
strategies. Through the use of EPA’s Data Quality Objectives process (DQO) CSMEP is
developing general ‘design templates’ for monitoring the status and trends of fish
populations and the effectiveness of habitat, harvest, hatchery and hydrosystem recovery
actions within the Columbia River Basin. The CSMEP design process is outlined in the
document “Proposed evaluation and design of preliminary design templates” available on
the CSMEP website.
CSMEP’s focus on developing its M&E designs employing EPA’s DQO process is
intended to emphasize iterative learning within an adaptive management loop. CSMEP’s
overall design process involves the following steps:
1. initial problem assessment to make explicit our current understanding of the system,
clarify our understanding of management goals in the Columbia Basin, and identify
the key uncertainties in evaluating agency management actions;
2. careful design of monitoring to evaluate management actions and reduce the key
uncertainties;
3. monitoring of key performance measures to test key management hypotheses and
assess progress towards management goals;
4. evaluation of monitoring results against the goals defined in the assessment phase;
and
5. adjustments in our understanding of the system and the effects of management
actions; and proceed back to step 1.
22
In FY 2005, five CSMEP subgroups (Status and Trends, Habitat, Harvest, Hydro and Hatcheries)
have been applying the 7-step EPA Data Quality Objectives (DQO) process to develop a set of
robust M&E designs for evaluating both the status and trends of fish populations and the
effectiveness of habitat, harvest, hatchery and hydrosystem recovery actions in the Columbia
Basin. As a pilot example of this design process, CSMEP has focused their efforts to date
principally on the Snake River Basin spring/summer/fall chinook ESU (see map of the Columbia
River subbasin areas encompassed by the CSMEP pilot). This pilot exercise is however intended
to illustrate the collaborative processes that will be required for further development of an
integrated monitoring program across the entire Columbia River Basin (see “Guidance in
applying EPA’s DQO process to CSMEP’s FY 2005 Design Task”).
E4.1 Status and Trends Subgroup
CSMEP’s Status and Trends Subgroup has focused on identifying monitoring design
elements necessary to adequately address one of the most important management
decisions in the Snake River Basin: has there been sufficient improvement in population
status of a listed Snake River S/S Chinook ESU to justify delisting and allow removal of
ESA restrictions? This decision is based on the abundance, productivity and spatial
structure & diversity of SRSS chinook salmon over the prior 10 years (IC-TRT 2005).
The Status and Trends Subgroup’s summary of the design elements (DQO steps 1-7) for
status and trends monitoring that are required to answer this question is provided on the
CSMEP Website. A full description of the Status and Trends Subgroup’s work on DQO
steps 1-5 for the Snake River Pilot is presented as a chapter in Marmorek et al. 2005. A
brief PowerPoint presentation describing the Status and Trends Subgroup’s DQO steps 15 is also provided on the CSMEP website.
In FY 2005 as part of its work on DQO steps 6 and 7, the Status and Trends Subgroup
began development of a simulation model that can be used for evaluating alternative
designs for monitoring fish at the population, major population group and ESU scales;
this tool will be further refined in FY 2006. These design alternatives are intended to
describe: 1) the location and temporal pattern of measurements (“sampling design”); 2)
the specific types of measurements that are to be made (“response design”); and 3) the
analyses to be performed to make a decision (“evaluation design”). Alternative design
templates will be compared in terms of cost (dollars/yr) and probability of error in
decisions that are associated with individual templates. The immediate objective of this
simulation is to evaluate alternative design templates for determining the status of SRSS
Chinook salmon. The ultimate objective is to develop a tool that can be adapted for
monitoring designs in other basins and for other species. A draft document outlining the
Subgroup’s DQO steps 6-7 approach is provided on the CSMEP Website, as is a
preliminary version of the alternative design spreadsheet that will inform their model.
Viability datasets for Idaho have been assembled to assist in the task. Maps of alternative
monitoring designs for the Snake Basin pilot area (current vs. CSMEP low, medium, high
designs) are available on the CSMEP website. PowerPoint presentations on the
subgroup’s approach to DQO steps 6-7 (Presentation1, Presentation2) are provided on the
CSMEP website.
23
E4.2 Hydro Subgroup
CSMEP’s Hydro Subgroup took on a subset of hydro management questions across
several scales: individual projects, survival by different passage routes through the
hydrosystem, and overall life cycle survival. These different scales relate to a variety of
decisions: operations at individual projects (e.g. spill, bypass, removable spillway weirs);
overall operations (e.g. when to transport fish within season, compliance with
hydrosystem biological opinions), longer term hydrosystem decisions (e.g. flow
management, effectiveness of transportation over multiple years, system configuration);
and adequacy of hydrosystem operations for stock recovery. The choices that are
available to improve the quality of information for hydrosystem decisions, and reduce the
risks of making incorrect decisions, include: the number of years of data collected, the
magnitude of tagging effort, the number of stocks that are monitored, the ability to filter
out year to year natural variation and isolate the signal of management actions, and
implementation of deliberate manipulations of hydrosystem operations to reduce
uncertainty in effectiveness evaluations. For many of these questions, CSMEP has
developed low, medium and high alternative designs and explored the strengths and
weaknesses of each approach.
The Hydro Subgroup’s summary of the design elements (DQO steps 1-7) for hydro
monitoring are provided on the CSMEP website. A full description of the Hydro
Subgroup’s work on DQO steps 1-5 for the Snake pilot is presented as a chapter in
Marmorek et al. 2005. A full report on the Hydro Subgroup’s current progress on DQO
steps 6-7 is provided on the CSMEP Website, as are PowerPoint presentations for the
Hydro Subgroup’s DQO steps 1-5 and DQO steps 6-7 (Presentation1, Presentation2)
E4.3 Habitat Subgroup
Habitat actions are considered a cornerstone of recovery strategies for Columbia River
Basin fish stocks but there is a need to more clearly determine the effectiveness of these
actions for increasing salmonid survival rates and production. Monitoring designs for
evaluating the effectiveness of habitat actions must be able to reliably detect two linked
responses:
1. the effect of habitat actions on fish habitat; and
2. the effect of changes in fish habitat on fish populations.
The Habitat Subgroup’s summary of the general design elements (DQO steps 1-5) for
Habitat monitoring that can help address these questions is provided on the CSMEP
website. A full description of the Habitat Subgroup’s current work on DQO steps 1-5 for
the Snake River Pilot (at both intensive and extensive scales) is presented as a chapter in
Marmorek et al. 2005.
The Habitat Subgroup has recognized, however, that there are serious challenges to the
development of a generic template for habitat effectiveness monitoring. These include:
1. Habitat conditions vary greatly across subbasins in terms of their natural
biogeoclimatic regimes, the status of their fish populations, the degree of human
24
impact and management, and the number and nature of restoration actions that
have been implemented, or are being considered for implementation within them.
2. Habitat effectiveness questions encompass different scales of inquiry, which
imply different scales of monitoring.
The Subgroup is instead attempting to develop a consistent “question clarification
process” that can be applied to development of individual monitoring designs dependent
on the particular situation. They are piloting this approach within the Lemhi River
Subbasin. A summary of the Habitat Subgroup’s detailed DQO steps 1-7 for the Lemhi
River Subbasin is provided on the CSMEP website, as is a full Habitat report for the
Lemhi River Subbasin. PowerPoint presentations for the Habitat Subgroup’s DQO steps
1-5 and DQO steps 6-7 (Presentation1, Presentation2) are also available on the CSMEP
Website.
E4.4 Hatchery Subgroup
Throughout the FY 2005 contract period, the Hatchery Subgroup identified a number of
questions important to the evaluation of hatchery management, and has reviewed
numerous existing and proposed hatchery research, monitoring, and evaluation (RME)
plans within the Columbia River Basin. Following this review, the Hatchery Subgroup
has concluded that existing and proposed hatchery RME plans (if fully implemented) are
likely to address the majority of the management questions identified by the Subgroup.
However, the Hatchery Subgroup has also concluded that a number of questions
regarding the effectiveness of hatcheries as a class of actions are unlikely to be adequately
addressed by existing and proposed hatchery RME. These hatchery effectiveness
questions (identified in the Hatchery Subgroup’s Summary on the CSMEP website) will
likely be efficiently and comprehensively addressed only through the implementation of a
stratified and representative study design that spans the entire Columbia River Basin.
With appropriate stratification, this diversity can be leveraged to identify the mechanistic
linkages of individual programs to broader monitoring questions that evaluate the overall
effectiveness of hatchery strategies at the regional scale. These broader-scale hatchery
program effectiveness questions (as opposed to individual hatchery operation questions)
will become the focus of CSMEP designs intended to address larger scale multi-hatchery
questions that can be stratified across the region.
The Hatchery Subgroup has focused much of their initial efforts on developing
alternative monitoring designs that could help answer two of these critical questions
relating to hatchery effectiveness:
1. What is the magnitude and distribution of hatchery strays into natural populations,
and;
2. What is the relative reproductive success of naturally spawning hatchery fish and
natural origin fish?
Insights into approaches gained from the CSMEP analyses required to address these two
questions will provide a foundation for tackling additional hatchery questions in a
prioritized manner in FY 2006. The Hatchery Subgroup’s summary of the design
25
elements (DQO steps 1-7) for hatchery monitoring that are required to answer these and
other questions are provided on the CSMEP website. A full description of the Hatchery
Subgroup’s work on DQO steps 1-5 for the Snake pilot is presented as a chapter in
Marmorek et al. 2005. A report on the Hatchery subgroup’s current progress on DQO
steps 6-7 is provided on the CSMEP website, as are PowerPoint presentations for the
Hatchery Subgroup’s DQO steps 1-5 and DQO steps 6-7.
E4.5 Harvest Subgroup
Targeted fisheries on salmon are managed by setting allowable catch, catch allocations
and open periods for each fishery prior to opening a fishery (considering escapement
goals and preseason/updated run predictions) and then adjusting those regulations as runs
develop. However, both mark-selective and non-selective fisheries can exert mortality on
non-targeted stocks of anadromous, adfluvial, and resident species that are incidentally
intercepted. Removal of fish in fisheries can potentially affect spawners, life history
diversity and the spatial structure of populations. The Harvest Subgroup has therefore
been focused on developing alternative monitoring designs that can answer two general
classes of Harvest questions:
1. What are the inseason estimates of run size and escapement for each stock
management group (target and non-target) and how do they compare to preseason
estimates?
2. What is the target and nontarget harvest and when is it projected to reach
allowable levels?
The Harvest Subgroup’s summary of the design elements (DQO steps 1-7) for harvest
monitoring required to answer these questions is provided on the CSMEP website. A full
description of the Harvest Subgroup’s work on DQO steps 1-5 for the Snake pilot is
presented as a chapter in Marmorek et al. 2005. A report on the Harvest Subgroup’s
current progress on DQO steps 6-7 is provided on the CSMEP Website, as are
PowerPoint presentations for the Harvest Subgroup’s DQO steps 1-5 and DQO steps 6-7
(Presentation 1, Presentation2).
E4.6 Integration of monitoring across the CSMEP subgroups
A CSMEP Monitoring Integration Group has been formed to explore the integration of
the individual RME component parts within a larger monitoring framework (i.e., generate
improved efficiencies through integrated designs) for the Snake River Basin pilot design.
This integration effort across scales and monitoring efforts is a challenge faced by all
subbasins; hence the results will be of general benefit basin wide. The Integration Group
has begun to develop a matrix of shared performance measures and data
interdependencies across the different CSMEP subgroups. This evolving Looking
Outward Matrix (LOM) is available on the CSMEP Website. The matrix is providing a
starting foundation for identifying the priority performance measures for monitoring and
the relevant spatial scale(s) of these data for varied subgroup monitoring needs. The
Monitoring Integration Group is also pursuing a simulation analysis to assess the
cost/benefit of a large integrated PIT-tagging program designed to address a range of key
monitoring questions across the subgroups. The ultimate intent is to evaluate what
26
intensities of basin-wide PIT-tagging (and at what life-stages) would/would-not-be
sufficient to achieve adequate statistical power and at reasonable cost to address the suite
of Subgroup questions at various spatial scales. Initial analyses for this exercise are
presented as a draft report (PIT tag V4 12-14-05.doc) on the CSMEP Website. The
Integration Group will be working to further quantify this analysis in FY 2006 and
intends to extend this approach into other sampling protocols that have the potential for
integration across the monitoring subgroups.
In FY 2006 CSMEP design subgroups will consolidate their work on the Snake River
pilot designs and refine the tools that can be used to explore tradeoffs across alternative
M&E design options (e.g.. low, medium, high) for different questions, at different scales
of interest (e.g., project, population, MPG, ESU, CRB). CSMEP has also begun to
explore the full integration of EPA’s new EMAP Idaho Master Sample (see example map
of pre-selected EMAP sample points for the South Fork Salmon River MPG) into their
design work for the Snake Basin Pilot Project, beginning with a targeted technical
workshop on this topic in early FY 2006. Subgroup products will be summarized in a
report on M&E recommendations for the CSMEP Snake River Basin Pilot Project, which
will be externally reviewed and shared with Federal Action Agencies and NOAA pilot
project workgroups. CSMEP will begin to expand from the Snake River Basin pilot area
and start to develop broader M&E recommendations that can be applied to other
subbasins in the Columbia River Basin. Development of these expanded design analyses
are intended to follow a range of directions (e.g., applying Snake River Basin derived
design templates to CSMEP inventoried subbasins or NOAA pilot projects to test how
transferable designs are; consideration of high-level integration of M&E across multiple
subbasins (e.g., incorporation of contrasts in stock status and productivity); evaluating
options for designs across the entire CRB by building on level of current monitoring
infrastructure in different subbasins (as determined by Subbasin Plan descriptions and
CSMEP metadata inventories). Analytical results from CSMEP products in FY06 are
intended to help provide general M&E guidance that can feed into the NWPCC Rolling
Review process.
E5. Objective 6: Assist implementation of pilot projects/large scale monitoring
programs
CSMEP’s ongoing work on the Snake Basin Pilot Project is directly feeding into the
NOAA-F/BPA Salmon River Subbasin Pilot Study, and has assisted in the early
development of M&E designs for the Lemhi River Subbasin Habitat Conservation Plan
(HCP).
In FY06 CSMEP intends to convert recommendations from their Snake River Basin Pilot
Project into a practical plan for the Salmon River Subbasin Pilot Study (cross-fertilization
as CSMEP people are involved in this Pilot Study). CSMEP has also been tasked
(tentatively) by the Council’s Members Advisory Group (MAG) to facilitate the
development of improved multi-agency approaches to hydrosystem monitoring for
application to fall chinook in the Columbia Basin.
27
E6. Objective 7: Evaluate new monitoring program results and propose revisions to
monitoring approaches
CSMEP workshops throughout FY 2004 and FY 2005 have provided continuing
opportunities for biologists and biometricians from across the region to meet and discuss
recent advances in M&E approaches (e.g. EMAP sampling frames, results from federal
pilot projects, IMW strategies, etc.). CSMEP thus represents a unique forum for the
cross-fertilization of M&E ideas among federal, state and tribal fish agency staff. Ideas
expressed at these workshops are being incorporated into the developing CSMEP M&E
designs. Workshop presentations given by participants at the CSMEP monitoring design
workshops in FY 2004 and FY 2005 are provided on the CSMEP Website. An equally
important focus of CSMEP has been our efforts to work closely with managers to bridge
the gap between science and policy, and support better management decisions
CSMEP has grown into a cohesive team of analysts and biometricians with a clear
understanding of the issues involved in developing more efficient, integrated approaches
to M&E for fish and wildlife across the Columbia River Basin and with the technical
expertise to make real progress in this regard. The suite of hyperlinks within this
document provides an indication of the range of technical products directed towards the
improvement of regional M&E that have already been developed by CSMEP in the first
two years of the project (FY 2004 and FY 2005). Progress in FY 2006 is expected to
continue in all areas of the project and will build from the strong foundation established
during the first two years. CSMEP’s PISCES Statement of Work Report for FY 2006 is
available on the CSMEP Website.
F. CSMEP Plans for fy07-09
These are summarized in Table F1. More details are available in the full CSMEP
proposal for fy07-09.
28
Table F1. Summary of proposed CSMEP work elements and products by objective for
fy07-09, including collaborating entities. Programmatic and policy input required for
objectives 1, 5, 6 and 7.
Objectives and Work
Elements
Description of Work Products
1. Develop Work Plans / Interact with Programmatic Entities
1.1 Develop CSMEP
Collaboratively prepared quarterly
Quarterly Workplans
workplans to maximize integration and
efficiency, avoid duplication of effort.
1.2 Quarterly Progress
Quarterly reports by objective and work
Reports
element to ensure close contract
monitoring.
1.3 Preparation of Draft and
3-level annual reports: 2-pg. exec.
Final Annual Reports
Summary; ~25-pg. overview w ~75-pg.
appendices; hyperlinked detailed reports
1.4 CSMEP conference calls, Biweekly Calls: Track progress, review
meetings and workshops
products, coordinate efforts.
Workshops: Present results, get
technical/programmatic feedback,
brainstorm next steps in subgroups
1.5 Coordination w PNAMP
Joint PNAMP/CSMEP workshop in spring
on joint activities and work
each year with managers and policy
products
makers; Annual work planning; Synthesis of
work products
1.6 Present CSMEP progress Give presentations to inform region CSMEP
at various Columbia Basin
outcomes and products, and to integrate
forums
with others' efforts
2. Inventory existing data
relevant to questions
2.1 QA on StreamNet
Review StreamNet's metadata inventories
Inventory Work for ID, WA,
for CSMEP pilot areas in ID, WA and OR,
OR pilot projects
including salmon, steelhead, bull trout and
other high priority resident fish
2.2 Inventory Sockeye; Bull
Update metadata inventory of Okanagan
Trout; Other Resident Fish of
sockeye, add Wenatchee and Redfish Lake
Special Concern
stocks. Improve inventory coverage of bull
trout and resident fish of high concern.
Develop common inventory data standards
for Sockeye stocks.
3. Organize subset of data into accessible form
3.1 Continue to improve
CSMEP web-based metadata
application
3.2 CSMEP website
improvement
3.3 Database design
Add hyper links from CSMEP database to
on-line databases housed by agencies
collecting/maintaining data
Provide user-friendly summaries of CSMEP
products in hierarchical form to
communicate to multiple audiences
Develop standards for performance
measures, data types, and data sharing
protocols, starting w Snake pilot project
(chinook and steelhead) and gradually
expanding to other areas and species.
Timing
Entities Collaborating w
CSMEP
quarterly
PNAMP, BPA, NWPCC,
StreamNet
quarterly
BPA
annual
internal
biweekly
calls;
workshops
3X/yr
Calls: internal
Workshops:
Programmatic and Policy
Feedback for part of
meeting
PNAMP Steering
Committee
annual joint
workshop;
planning
mtgs 2X /yr
2-3 times /
yr
NWPCC, TRTs, PNAMP,
WA DOE, AFS, EPA
FY07-FY09
StreamNet (separate
contract); fed/state/tribal
agencies with data
FY07
DFO, Okanagan Nation
Alliance, StreamNet,
CBFWA Resident Fish
Committee
FY07-FY09
StreamNet (separate
contract); fed/state/tribal
agencies with data
CBFWA web master Amy
Langston
FY07-FY09
FY07-FY09
StreamNet, NED, PNAMP
29
Objectives and Work
Elements
Timing
CSMEP
Detailed review of CSMEP inventories and
strengths & weaknesses assessments; GIS
overlays of existing sampling sites with
EMAP master sample to refine sampling
designs for status & trend
As above for work element 4.1 for areas of
OR pilot (Grande Ronde and Imnaha)
mostly in
FY06; some
work in
FY07
USFS, BLM, OR Aquatic
Inventory, OWEB, ID
DEQ, PCSRF, BoR, EPA,
others
FY07 and
FY08
Area of WA pilot includes salmon recovery
regions in Lower, Middle and Upper
Columbia.
FY07 and
FY08
USFS, BLM, OR Aquatic
Inventory, OWEB, OR
DEQ, PCSRF, BoR, EPA,
others
USFS, BLM, WA SRFB,
WA DEQ, PCSRF, BoR,
EPA, DFO, others
Demonstrate cost-precision and other
tradeoffs associated with alternative
integrated designs that attempt to meet
information needs for key decisions in
recovery assessment and 4 H's.
mostly in
FY06; some
work in
FY07
Develop and present three ESU / provincial
scale fish and habitat M&E plans that
integrate across issues, questions, species
and agencies to meet identified priorities.
Involve CBFWA Resident Fish Committee
in all three projects. Work towards
implementation of these plans.
Generalize qualitative and quantitative tools
developed in CSMEP DQO process for use
throughout CRB. Present general
implications to managers and scientists.
ID (FY0607);
OR (FY0709); WA
(FY07-09)
Extend tools for assessing M&E designs to
detect recovery status (more VSP criteria,
species); publish/present results.
Complete DQO steps 6-7 for hydrosystem
decisions and publish/present results for
feedback. Extend findings to other regions
(i.e. Mid/Upper Columbia).
Work with restoration managers across
multiple watersheds on implementation /
M&E methods that will maximize learning
on restoration effectiveness at various
scales.
Complete M&E designs for large scale
hatchery / supplementation questions (i.e.
hatchery straying into wild populations,
relative reproductive success) and present
recommended plans
FY07-09
4. Evaluate ability to answer key questions with existing data
4.1 Organization of existing
data for Snake Basin pilot
design
data for OR pilot design
data for WA pilot design
5. Collaborative monitoring
program design
5.1 Consolidate Snake River
Pilot M&E design and
PrOACT tradeoff analysis
5.2 Complete three pilot M&E
projects at provincial / ESU
scales in ID, OR, WA
5.2a ID pilot project
5.2b WA pilot project
5.2c OR pilot project
5.3 Extend application of
CSMEP insights and tools to
other parts of CRB and
PNAMP entities
5.3a Status & Trends
5.3b Hydro
5.3c Habitat
5.3d Hatchery
FY07-09
Interact with
programmatic/policy
entities with responsibility
for decisions in recovery
assessment and 4 H's to
fine tune design, costs.
as for work element 5.1.
Build on insights gained
from NOAA-AA pilot
projects, WA GSRO,
CSMEP survey of M&E
priorities.
Market existence of tools
and results through
CBFWA, PNAMP,
NWPCC websites and
newsletters
Technical Recovery
Teams, ISRP/ISAB
FY07-09
AFEP, NOAA-Hydro,
PUDs, ISRP/ISAB
FY07-09
PNAMP entities,
restoration managers,
NOAA-AA pilot projects,
ISRP/ISAB
FY07-09
Hatchery managers,
ISRP/ISAB
30
Objectives and Work
Elements
5.3e Harvest
Timing
5.4 Feed M&E results into
NWPCC Provincial Review
Process
5.5 Get feedback from CRB
entities on various M&E
designs
6. Multi-agency implementation of monitoring programs.
FY07-09
CSMEP
Harvest managers (e.g.
TAC, US v. OR, PST,
CTC), ISRP/ISAB
NWPCC, CBFWA
Complete M&E designs for improving data
for harvest pre-season and in-season
decisions
Interact with CBFWA / NWPCC managers
to provide insights for project approval and
M&E guidelines
As described above under 5.2 and 5.3, but
beyond immediate areas of pilot studies
FY07-09
FY07-09
As for work element 5.1.
6.1 Provide input to
Participate in collaborative efforts building
conceptual plan for M&E
on knowledge developed from CSMEP,
implementation across CRB
NOAA and other pilot projects.
7. Multi-agency evaluation of results of new monitoring pgms.
FY07-09
As for work element 5.1.
7.1 Collaborative review of
federal RME projects, WA
SRFB Effectiveness
Monitoring Projects, and other
recent pilot projects
FY07-09
NOAA-AA pilot projects
Review results of Wenatchee, John Day
and Salmon pilot projects as they are made
available and incorporate into next set of
designs.
31
CSMEP - Status and Trends Subgroup
CSMEP’s Status and Trends subgroup has focused its DQO efforts on identifying
monitoring design elements necessary to address one of the most important management
decisions in the Snake Basin: has there been sufficient improvement in status of Snake
River S/S Chinook to justify delisting the ESU and allow removal of ESA restrictions?
This decision is based on the abundance, productivity, spatial structure and diversity of
SRSS Chinook salmon over the prior 10 years (IC-TRT 2005). Appendix A provides the
subgroup’s summary of the design elements (DQO steps 1-7) for Status and Trends
monitoring that are required to answer this question.
In FY05 as part of its work on DQO steps 6 and 7, the S &T subgroup began
development of a simulation model that can be used for evaluating alternative designs for
monitoring fish at the population, major population group and ESU scales (Figure 1); this
tool will be further refined in FY06. Monitoring designs are described as high, medium,
and low templates, in reference to the levels of accuracy and precision in data that are
collected using each template. Alternative design templates will be compared in terms of
cost ($/yr) and probability of error in decisions that are associated with individual
templates.
Sensitivity analysis will be used to examine how the rate and direction (type I versus type
II) of decision error responds to uncertainty in monitoring data. Specific monitoring
activities that will achieve high, medium, or low accuracy and precision may vary on a
site by site basis. To determine how specific monitoring activities influence uncertainty
in data, monitoring data must be compared to a performance standard (usually a census
dataset) using calibration studies. Calibration studies can provide information on the
accuracy and precision achieved by varying 1) the location and temporal pattern of
measurements (“sampling design”); 2) the specific types of measurements that are to be
made (“response design”); and 3) the analyses to be performed to make a decision
(“evaluation design”). Sensitivity analyses will identify where uncertainty in monitoring
data most strongly affects error in final decisions; thus, sensitivity analyses can identify
monitoring activities for which calibration studies are justified.
The immediate objective of this model is to evaluate alternative design templates for
determining the status of SRSS Chinook salmon, while limiting risk in the decision to
acceptable levels. The ultimate objective is to develop a tool that can be adapted for
monitoring designs in other basins and for other species. Figure 1 presents maps of the
current design of monitoring in the Snake Basin pilot vs. examples of alternative CSMEP
low, medium and high designs that will be evaluated in FY06.
32
Figure 1.
Process for developing a simulation model that can be used for evaluating
alternative status and trends designs for monitoring fish at the population, major
population group and ESU scales.
33
Figure 2.
Mapping of current and alternative CSMEP monitoring designs (low, medium, high) for TRT populations in the Snake River
Basin Sp/S chinook ESU.
34
Appendix A. CSMEP’s Status and Trends DQO Summary
DQO STEPS
SNAKE RIVER BASIN PILOT
1. State the Problem
Problem:
Delisting of Snake River S/S Chinook ESU
Stakeholders:
States—Washington, Oregon, Idaho
Tribes—NPT, SBT, CTUIR, CTWSR, YIN
Federal—NOAA, USFWS, USFS, BPA, USACOE
Intergovernmental—Columbia River Compact, CBFWA, CRITFC, PFMC, PSC, NPCC
Other—Idaho Power, conservation groups, fishers (tribal, commercial, sport),
landowners, upland land users (ranchers, farmers, municipalities, state and county
governments), water users (agricultural, industrial, municipal)
Interagency coordination, fiscal constraints, legal constraints, land ownership and access
Non-technical
Issues:
Conceptual Model: Life history models
2. Identify the Decision
What is the ESA listing status for Snake River S/S Chinook salmon?
Principal
Questions:
Alternative
• If status is “listed,” then recovery strategies (i.e., more restrictive management
Actions:
strategies at one or more points in the life history model).
• If status is “delisted,” then recovery or sustainable harvest strategies.
• If status is “recovered,” then sustainable harvest strategies
Decision
• Has there been sufficient improvement in population status of Snake River S/S
Statements:
Chinook ESU to justify delisting and allow removal of ESA restrictions?
• Are additional management actions required for regional, ESA recovery and NPCC
SAR goals?
3. Identify the Inputs
Information
Required:
Spatial
Information required
Abundance
Productivity
Diversity
structure
Abundance of
9
9
9
spawners
Abundance/distribution
9
9
9
9
of redds
Origin of spawners
9
9
Age-structure of
9
9
9
spawners
Sex ratio of spawners
9
9
Abundance/distribution
9
of juveniles
Juvenile survival
9
Policy
Inputs1
(9)
9
9
Sources of Data:
State, tribal, and federal programs and NGSs identified in CSMEP metadata inventories
Quality of Existing Data varies in level of precision and bias. Major issues:
Data:
• Abundance of spawners: 10 of 31 populations have weirs in combination with redd
counts, 21 of 31 populations rely on redd abundance as a surrogate for spawner
35
DQO STEPS
New Data
Required:
Analytical
Methods:
abundance. Weir data provide precise information on abundance of spawners but no
information on spawner distribution; redd data provide less precise information on
abundance but also provide information on distribution of spawners.
• Abundance/distribution of redds: populations vary in spatial and temporal extent and
resolution. Fixed index sites are used instead of probabilistic methods for site
selection.
• Origin of spawners: mark quality is high, but sample sizes are low especially during
years of low abundance.
• Age-structure of spawners: because they are based on carcass recoveries estimates are
imprecise and likely biased. As an alternative, application of a basin-wide estimate is
also imprecise and likely biased at the population-level.
• Sex ratio of spawners: same as for age-structure data
• Abundance/distribution of juveniles: 15 of 31 have juvenile traps, 22 of 31 populations
have snorkel . Trap data is more precise for abundance but give no information on
distribution; snorkel data are imprecise for abundance but provide high quality
information on distribution.
• Survival of juveniles: PIT-tags can provide precise estimates but sample sizes are low
in less productive populations.
• Probabilistic sampling strategies reduce bias and likely will improve information
available for spatial structure and diversity.
• Existing methods may require calibration and validation to improve their utility.
• Analysis of available data may indicate performance measures for which higher
quality data are needed to evaluate decisions
IC-TRT rules and criteria for combining measures of abundance, productivity, spatial
structure, and diversity.
Policy
Inputs1
(9)
9
4. Define the Boundaries
Snake River S/S Chinook Salmon
Target
Populations:
Spatial Boundaries Population, MPG, and ESU levels for S/S Chinook salmon within Snake River basin.
(study):
Temporal
Status data evaluated over generations from annual abundance data, generational
Boundaries
productivity data, and spatial structure and diversity data collected at unspecified
(study):
intervals. Data on historical distribution and productivity also are needed.
Practical
Legal and logistical issues with access, interagency coordination across jurisdictional
Constraints:
boundaries.
Spatial Boundaries Delisting decision made at level of ESU.
(decisions):
Temporal
IC-TRT rules for abundance and productivity require historical data, and 10 year series
Boundaries
of annual data. IC-TRT rules require spatial structure and diversity data collected at
(decisions):
unspecified intervals.
9
9
9
5. Decision Rules (IC-TRT Rules)
Critical
Two metrics (A/P and SS/D) are used to assess the status of each population. A/P
Components and
combines abundance and productivity VSP criteria using a viability curve. SS/D
Population
integrates 12 measures of spatial structure and diversity.
Parameters:
9
36
DQO STEPS
Critical Action
Levels (Effect
Sizes):
Risk categories are assigned at the population level for A/P using a 5% risk criterion to
define viable populations. Populations scored as moderate or high risk in A/P criteria
cannot meet viable standards, while populations at high risk for the 12 SS/D measures
cannot be considered viable.
If-Then Decision
Rules:
IC-TRT Draft
MPG-level Viability Criteria:
Low risk (viable) MPGs meet the following six criteria:
1. One-half of the populations historically within the MPG (with a minimum of two
populations) must meet minimum viability standards (Section X).
2. All populations meeting viability standards within the ESU cannot be in the
minimum viability category (Figure X); at least one population must be
categorized as meeting more than minimum viability requirements.
3. The populations at high viability within an MPG must include proportional
representation from populations classified as “Large” or “Intermediate” based on
their intrinsic potential.
4. Populations not meeting viability standards should be maintained with sufficient
productivity that the overall MPG productivity does not fall below replacement
(i.e. these areas should not serve as significant population sinks).
5. Where possible, given other MPG viability requirements, some populations
meeting viability standards should be contiguous AND some populations meeting
viability standards should be disjunct from each other.
6. All major life history strategies (i.e. adult “races,” A-run/B-run, resident and
anadromous) that were present historically within the MPG must be present and
viable.
ESU-level Viability Criteria:
1. All extant MPGs and any extirpated MPGs critical for proper functioning of the
ESU must be at low risk.
2. ESUs that contained only one MPG historically must meet the following criteria:
a. Two-thirds or more of the populations within the MPG historically must meet
minimum viability standards; AND
b. Have at least two populations categorized as meeting more than minimum
viability requirements.
Incorrectly concluding that delisting criteria have been achieved:
• Decisions to relax ESA restrictions increase risks to the ESU
Incorrectly concluding that delisting criteria have not been achieved:
• Minimal biological impact given that decisions do not relax ESA restrictions
• May over-invest in intensity of monitoring efforts
• Unnecessary listing and restrictive measures
• Loss of harvest opportunity
Consequences of
Decision Errors:
Policy
Inputs1
(9)
9
9
1
Policy Inputs - indicate with a check steps where group really needs policy feedback, presentation will
elaborate on what feedback is required
Steps 6 and 7.
Optimizing the Design (examples)
Evaluation Design (How data will be analyzed to answer a question)
Sampling Design (Where and When data will be collected)
Response Design (What and How data are collected)
Refer to following matrices.
37
"Current" Design Template:
Abundance / B1
spatial
distribution B2
of redds
B3
Cen-cen
B4
Stat-cen
B5
Stat-multi
B6
Stat-once
B7
Fixed-cen
B8
Fixed-multi
B9
Fixed-once
C1
Tags
C2
Hard parts
C3
Length at age
C4
Basinwide estimate
Age
structure of
spawners
Origin of
spawners
D1
D2
D3
Sex ratio of
spawners
E1
E2
Abundance /
F1
spatial
distribution
F2
of smolts
F3
Survival of
juveniles
G1
1
1
Cen-once
1
Hatchery marks,
handle fish at weirs
Hatchery marks,
remotely sense
Hatchery marks on
carcasses
1
1
1
1
1
1
1
1
1
1
Up. Salmon Up. Mainstem
Valley Creek
Yankee Fork
East Fork Salmon River
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Carcassper
survey
Female
redd
expansion
1
1
1
1
1
1
1
1
Juvenile trap
1
1
1
1
1
1
1
1
1
1
1
1
1
Electrofishing
Snorkel survey
Mark recapture
1
1
1
Lemhi
1
1
Cen-multi
1
Up. Salmon Low. Mainstem
Pahsimeroi River
Panther - extirpated
North Fork
Marsh Creek
Bear Valley Creek
Sulphur Creek
Upper MF Mainstem
Camas Creek
Loon Creek
1
1
Upper Mainstem Salmon
Chamberlain Creek
1
Big Creek
1
Lower MF Mainstem
Secesh
1
EFSF
1
Middle Fork
SF Mainstem
Up. Grande Ronde River
1
Little Salmon
Lostine
1
Imnaha Mainstem
SF Salmon
Minam
Big Sheep Creek
Catherine Creek
Lookingglass - extirpated
Wenaha
Method / Description
A1 Census weir
A2 Weir with MR
A3 Weir without MR
A4 MR survey, no weir
Tucannon
Data need
Abundance
of fish
Asotin
Low. Snake Grand Ronde-Imnaha
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
38
"High" Design Template:
Pahsimeroi River
Lemhi
Yankee Fork
Valley Creek
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Wenaha
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Carcass survey
Female per redd
expansion
1
1
1
1
1
1
Juvenile trap
Electrofishing
Snorkel survey
1
1
1
1
1
1
1
1
1
Mark recapture
1
1
1
1
Tags
Hard parts
Length at age
Hatchery marks,
Hatchery marks,
remotely sense
Hatchery marks on
carcasses
1
1
1
1
1
1
1
1
1
North Fork
1
Marsh Creek
1
Bear Valley Creek
1
Sulphur Creek
1
Upper MF Mainstem
1
Camas Creek
1
Loon Creek
1
Chamberlain Creek
1
Big Creek
1
Lower MF Mainstem
1
Secesh
1
EFSF
1
SF Mainstem
1
Little Salmon
1
1
Lostine
1
Imnaha Mainstem
1
Minam
1
Big Sheep Creek
1
Catherine Creek
1
1
Tucannon
E2
Abundance / F1
spatial
F2
distribution F3
Survival of
juveniles
G1
1
1
Asotin
D3
E1
1
1
1
1
Basinwide estimate
Sex ratio of
spawners
1
1
1
1
C4
D2
1
1
1
Cen-cen
Cen-multi
Cen-once
Stat-cen
Stat-multi
Stat-once
Fixed-cen
Fixed-multi
Fixed-once
D1
1
1
1
B1
B2
B3
B4
B5
B6
B7
B8
B9
C1
C2
C3
1
1
Abundance /
spatial
distribution
of redds
Middle Fork
1
1
A1 Census weir
A2 Weir with MR
A3 Weir without MR
Origin of
spawners
SF Salmon
1
Data need
Abundance
of fish
Age
structure of
spawners
39
"Medium" Design Template:
G1
1
Valley Creek
1
Yankee Fork
1
1
Lemhi
E2
F1
F2
F3
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Pahsimeroi River
1
1
1
1
Abundance /
spatial
distribution
Survival of
juveniles
D3
E1
1
North Fork
1
1
Marsh Creek
1
Hatchery marks,
Hatchery marks,
remotely sense
Hatchery marks on
carcasses
1
Bear Valley Creek
1
1
Sulphur Creek
1
1
Upper MF Mainstem
1
1
Camas Creek
1
1
Loon Creek
1
1
Chamberlain Creek
1
1
Big Creek
1
1
Lower MF Mainstem
1
1
Secesh
Tags
Hard parts
Length at age
1
1
EFSF
1
SF Mainstem
1
Little Salmon
1
Middle Fork
1
1
Lostine
1
Basinwide estimate
Sex ratio of
spawners
Imnaha Mainstem
1
Minam
1
C4
D2
Big Sheep Creek
1
B1
B2
B3
B4
B5
B6
B7
B8
B9
C1
C2
C3
D1
Catherine Creek
Tucannon
Cen-cen
Cen-multi
Cen-once
Stat-cen
Stat-multi
Stat-once
Fixed-cen
Fixed-multi
Fixed-once
Abundance /
spatial
distribution
of redds
Wenaha
Asotin
1
A1 Census weir
A2 Weir with MR
A3 Weir without MR
Origin of
spawners
SF Salmon
1
Data need
Abundance
of fish
Age
structure of
spawners
1
1
1
1
1
1
1
1
1
1
Carcass survey
Female per redd
expansion
Juvenile trap
Electrofishing
Snorkel survey
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Mark recapture
40
"Low" Design Template:
Big Sheep Creek
Minam
Imnaha Mainstem
Lostine
Little Salmon
SF Mainstem
EFSF
Secesh
Lower MF Mainstem
Big Creek
Chamberlain Creek
Loon Creek
Camas Creek
Upper MF Mainstem
Sulphur Creek
Bear Valley Creek
Marsh Creek
North Fork
Pahsimeroi River
Lemhi
Yankee Fork
Valley Creek
Abundance /
spatial
distribution
of redds
B1
B2
B3
B4
B5
B6
B7
B8
B9
C1
C2
C3
Cen-cen
Cen-multi
Cen-once
Stat-cen
Stat-multi
Stat-once
Fixed-cen
Fixed-multi
Fixed-once
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Tags
Hard parts
Length at age
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
C4
Basinwide estimate
Hatchery marks,
Hatchery marks,
remotely sense
Hatchery marks on
carcasses
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Carcass survey
Female per redd
expansion
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Age
structure of
spawners
Origin of
spawners
D1
D2
Sex ratio of
spawners
Abundance /
spatial
distribution
Survival of
juveniles
D3
E1
E2
F1
F2
F3
G1
A1 Census weir
A2 Weir with MR
A3 Weir without MR
Data need
Abundance
of fish
Catherine Creek
Wenaha
Middle Fork
Tucannon
SF Salmon
Asotin
Juvenile trap
Electrofishing
Snorkel survey
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Mark recapture
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
41
"Current" Design Template:
Weir without MR
A4
MR survey, no weir
Abundance /
B1
spatial
distribution B2
of redds
B3
Age
structure of
spawners
Origin of
spawners
Asotin
1
Tucannon
1
Big Sheep
Creek
Minam
Lostine
Wenaha
Up. Salmon
Up. Mainstem
Valley Creek
Yankee Fork
Pahsimeroi
River
Up. Salmon
Low.
Mainstem
East Fork
Salmon River
Lemhi
North Fork
Upper MF
Mainstem
Marsh Creek
Bear Valley
Creek
Sulphur Creek
Pistol Creek
Loon Creek
Camas Creek
Lower MF
Mainstem
1
Stat-multi
Stat-once
B7
Fixed-cen
B8
Fixed-multi
B9
Fixed-once
C1
Tags
C2
Hard parts
C3
Length at age
1
1
1
1
C4
Basinwide estimate
1
1
1
1
1
1
E2
Abundance /
F1
spatial
distribution F2
of smolts
F3
Survival of
juveniles
G1
1
1
1
1
1
B6
D3
Big Creek
1
1
Cen-once
Stat-cen
E1
Chamberlain
Creek
Little Salmon
EFSF
1
Cen-multi
B5
D1
1
1
Cen-cen
B4
D2
Sex ratio of
spawners
1
Imnaha
Mainstem
A3
1
Low. Snake
Grand Ronde-Imnaha
Up. Grande
Ronde River
Census weir
Weir with MR
Catherine
Creek
A1
A2
Little SCham Middle Fork
SF Salmon
Secesh
Abundance
of fish
SF Mainstem
Data need
Hatchery marks,
Hatchery marks,
remotely sense
Hatchery marks on
carcasses
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Juvenile trap
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Electrofishing
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Snorkel survey
1
1
1
1
1
1
Carcass survey
Female per redd
expansion
Mark recapture
1
1
1
1
1
1
1
1
1
1
42
1
"High" Design Template:
Age
structure of
spawners
Origin of
spawners
Pistol Creek
Sulphur Creek
Bear Valley
Creek
Marsh Creek
Upper MF
Mainstem
North Fork
Lemhi
Yankee Fork
Valley Creek
Up. Salmon
Up. Mainstem
Wenaha
Lostine
Minam
Catherine
Creek
Up. Grande
Ronde River
Imnaha
Mainstem
Big Sheep
Creek
Tucannon
Asotin
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Pahsimeroi
River
Up. Salmon
Low.
Mainstem
East Fork
Salmon River
Loon Creek
1
1
1
MR survey, no weir
Cen-cen
Cen-multi
1
1
Cen-once
B4
Stat-cen
B5
Stat-multi
B6
Stat-once
B7
Fixed-cen
B8
Fixed-multi
B9
Fixed-once
C1
Tags
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
C2
Hard parts
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
C3
Length at age
C4
Basinwide estimate
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Carcass survey
Female per redd
expansion
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Juvenile trap
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Snorkel survey
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Mark recapture
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
D1
D2
D3
Sex ratio of
spawners
Camas Creek
A4
Abundance /
B1
spatial
distribution B2
of redds
B3
Lower MF
Mainstem
Weir without MR
Big Creek
A3
Chamberlain
Creek
Census weir
Weir with MR
Low. Snake
Grand Ronde-Imnaha
Little Salmon
A1
A2
EFSF
Abundance
of fish
SF Salmon
Secesh
SF Mainstem
Data need
E1
E2
Abundance /
F1
spatial
distribution F2
of smolts
F3
Survival of
juveniles
G1
Hatchery marks,
Hatchery marks,
remotely sense
Hatchery marks on
carcasses
Electrofishing
43
"Medium" Design Template
Age
structure of
spawners
Origin of
spawners
Weir without MR
Asotin
Tucannon
Big Sheep
Creek
Imnaha
Mainstem
Up. Grande
Ronde River
Catherine
Creek
Minam
Lostine
Up. Salmon
Up. Mainstem
Valley Creek
Yankee Fork
Pahsimeroi
River
Up. Salmon
Low.
Mainstem
East Fork
Salmon River
Lemhi
North Fork
Upper MF
Mainstem
Marsh Creek
Bear Valley
Creek
Sulphur Creek
Pistol Creek
Loon Creek
Camas Creek
Lower MF
Mainstem
Big Creek
Wenaha
1
1
Stat-cen
B7
Fixed-cen
B8
Fixed-multi
B9
Fixed-once
C1
Tags
C2
Hard parts
C3
Length at age
C4
Basinwide estimate
E2
Abundance /
F1
spatial
distribution F2
of smolts
F3
Survival of
juveniles
G1
1
Cen-once
Stat-multi
E1
1
Cen-multi
Stat-once
D3
Sex ratio of
spawners
1
Cen-cen
B6
D2
1
Low. Snake
Grand Ronde-Imnaha
MR survey, no weir
B5
D1
Chamberlain
Creek
Weir with MR
A3
B4
1
Census weir
A2
A4
Abundance /
B1
spatial
distribution B2
of redds
B3
Little Salmon
A1
EFSF
Abundance
of fish
SF Salmon
Secesh
SF Mainstem
Data need
Hatchery marks,
Hatchery marks,
remotely sense
Hatchery marks on
carcasses
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Carcass survey
Female per redd
expansion
Juvenile trap
1
1
1
1
1
Electrofishing
Snorkel survey
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Mark recapture
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
44
"Low" Design Template:
Age
structure of
spawners
Origin of
spawners
1
1
1
1
1
1
1
Asotin
1
Tucannon
1
Big Sheep
Creek
1
Imnaha
Mainstem
1
Up. Grande
Ronde River
1
Catherine
Creek
1
Minam
1
Lostine
1
Wenaha
1
Up. Salmon
Up. Mainstem
1
Valley Creek
1
Yankee Fork
1
Pahsimeroi
River
Up. Salmon
Low.
Mainstem
East Fork
Salmon River
1
Lemhi
1
North Fork
1
Upper MF
Mainstem
Bear Valley
Creek
1
Marsh Creek
Sulphur Creek
1
Pistol Creek
1
Loon Creek
1
Low. Snake
Grand Ronde-Imnaha
1
1
1
1
Cen-cen
Cen-multi
Cen-once
B4
Stat-cen
B5
Stat-multi
B6
Stat-once
B7
Fixed-cen
B8
Fixed-multi
B9
Fixed-once
C1
Tags
C2
Hard parts
1
1
1
C3
Length at age
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
C4
Basinwide estimate
Hatchery marks,
Hatchery marks,
remotely sense
Hatchery marks on
carcasses
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Carcass survey
Female per redd
expansion
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Snorkel survey
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Mark recapture
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
D1
D2
D3
Sex ratio of
spawners
1
Camas Creek
MR survey, no weir
Abundance /
B1
spatial
distribution B2
of redds
B3
Lower MF
Mainstem
Weir without MR
A4
Big Creek
Weir with MR
A3
Chamberlain
Creek
Census weir
A2
Little Salmon
A1
EFSF
Abundance
of fish
SF Salmon
Secesh
SF Mainstem
Data need
E1
E2
Abundance /
F1
spatial
distribution F2
of smolts
F3
Survival of
juveniles
G1
1
1
1
1
Juvenile trap
Electrofishing
45
CSMEP – Harvest Action Effectiveness Subgroup
Targeted fisheries on salmon are managed by setting allowable catch, catch allocations and open
periods for each fishery prior to opening a fishery (considering escapement goals and
preseason/updated run predictions) and then adjusting those regulations as runs develop.
However, both mark-selective and non-selective fisheries can exert mortality on non-targeted
stocks of anadromous, adfluvial, and resident species that are incidentally intercepted. Removal
of fish in fisheries can potentially affect spawners, life history diversity and the spatial structure
of populations. The Harvest Subgroup has therefore been focused on developing alternative
M&E designs that improve data for harvest pre-season and in-season decisions. Such designs
must be able to answer two general classes of harvest questions:
1. What are the inseason estimates of run size and escapement for each stock management
group (target and non-target) and how do they compare to preseason estimates?
2. What is the target and nontarget harvest and when is it projected to reach allowable
levels?
Appendix A provides the Harvest Subgroup’s summary of the design elements (DQO steps 1-7)
for harvest monitoring that are required to answer these questions.
46
Appendix A. CSMEP Harvest DQO Summary
DQO STEPS
(Snake River spring/summer chinook ESU)
Policy
Inputs1
(9)
Problem:
Targeted fisheries on Chinook, steelhead, coho, and (in some years) sockeye are managed by setting allowable
catch and catch allocation limits and open periods for each fishery prior to opening a fishery (considering
escapement goals and preseason and updated run predictions) and then adjusting those openings and limits as runs
develop and catches are totaled.
Both mark-selective and non-selective fisheries exert mortality on non-targeted stocks of anadromous, adfluvial, and
resident species that are incidentally intercepted. Removal of fish in fisheries can potentially affect the number of
mature adults that spawn in natural and artificial production areas on a seasonal basis and potentially affect diversity
and spatial structure of population components on a longer term basis if removals are selective of phenotypes (e.g.,
size, sex or age). Fishing opportunity in areas with mixed stocks and species inevitably results in bycatch of nontargeted species or stocks. Because such bycatch counts towards the harvest of the bycaught species, it must be
accounted for. If the bycatch in a particular non-targeting fishery exceeds allowable catch or impacts set for that
fishery or some other pre-specified limit, then management actions will come into play. The type of action will depend
on the fishery, on the bycaught species and on management agreements in place.
Take includes direct harvest, indirect harvest (released fish that die or non-target landed fish). It may also be worth
considering the impact on fitness of catch and released fish.
Stakeholders:
State agencies and tribes that co-manage fisheries impacting anadromous fish populations:
•
•
•
•
•
•
•
•
•
•
Confederated Tribes of the Umatilla Indian Reservation
Confederated Tribes of the Warm Springs Reservation
Yakama Nation
Idaho Department of Fish & Game
National Oceanic and Atmospheric Administration Fisheries
Nez Perce Tribe of Idaho
Oregon Department of Fish & Wildlife
Shoshone-Bannock Tribes of Fort Hall
U.S. Fish & Wildlife Service
Washington Department of Fish and Wildlife
Non-technical
Issues:
Impacts to fish from other “H’s” and changes to fish marking programs are both technical and policy issues; other
non-technical issues are changes to artificial production schedules, consumer market demands and health concerns
(toxins).
Conceptual
Model:
Track components of run size and the methods of estimating them through the Columbia River tribal, commercial
and sport fisheries from the lower estuary to the tributaries of the Snake River basin for Snake River spring-summer
chinook salmon.
For example, natural origin Snake River spring Chinook salmon can be intercepted in mark-selective commercial and
sport fisheries downstream of Bonneville Dam, selective sport fisheries between Bonneville and McNary dams, and
in the lower Snake River; in traditional Treaty fisheries between Bonneville and McNary dams; and in terminal
selective sport and Treaty fisheries in Snake Basin tributaries. Snake River natural spring /summer chinook are
assumed by managers to have very low impact rates in ocean fisheries.
Principal
Questions:
What are the inseason estimates of run size and escapement for each management group (target and non-target)
and how do they compare to preseason estimates?
What is the target and nontarget harvest and when is it projected to reach allowable levels?
9
Alternative
Actions:
Open or close various fisheries;
Increased or decreased harvest opportunities for fishers.
9
Decision
Statements:
Open Fishery X during periods a, b, and c subject to the catch not exceeding Y for target species M and Z for nontarget species M.
Once the bycatch is projected to reach to quota, then the fishery would be halted, postponed, or reshaped.
9
47
DQO STEPS
Policy
Inputs1
(9)
Information
Required:
Catch, Effort, CPUE, stock identification (for mainstem spring season fisheries, the only stock identification used in
season is for below Bonneville fisheries where separation between Willamette and Upriver stocks are made.
Age-specific estimates of the numbers of each management unit (stock) in the escapement. Age specific data is only
used in forecasting, not in in-season fishery management.
Sources of Data:
Mainstem commercial, subsistence, ceremonial, and sport fisheries, Hatcheries, dams, previous fisheries, natural
spawning estimates, and mark samples.
Quality of Existing
Data:
The main source of uncertainty is statistical sampling error and perhaps bias due to assumption violations, such as
error in assumptions regarding release mortality rates. Decision making is typically based on point estimates of take
and addresses the uncertainties by adopting conservative actions.
(See Table at back of handout)
New Data
Required:
More data, more research on mortality rates resulting from interceptions with varying gears (e.g., hook and release,
tangle net release, and others). There is considerable uncertainty in tangle net release mortality rates. There have
also been wide ranging estimates made for hook and release mortality rates. There are debates about whether hook
location, barbed vs. barbless or environmental (i.e., temperature) are the most important determinant of release
mortality. There are no estimates currently in use for net dropout rates for Columbia River net fisheries.
Expansion of Genetic Stock Identification (GSI) baseline sampling and in-season sampling of catch and escapement.
There has been only limited GSI sampling of fish from harvest.
Analytical
Methods:
Stock-recruitment relationships have been used to set escapement goals for some Columbia Basin salmon
populations. Cohort analyses have been used to develop pre-season expectations. There are no agreed to
escapement goals for the entire Snake River spring/summer chinook ESU. There are some tributary return goals, but
these have not been useful for mainstem fishery management.
Target
Populations:
ESA listed salmonids including Snake Basin Chinook, sockeye, and steelhead ESUs. It should be noted that the
Snake River spring/summer ESU includes all naturally spawning spring/summer chinook populations in the Snake
Basin except from the Clearwater. This fact complicates management below the Mouth of the Clearwater, because
estimates of natural origin Snake River spring/summer chinook below the Clearwater are a mix of listed and nonlisted fish.
All anadromous populations impacted by mainstem and tributary fisheries during the time that Snake River spring
Chinook are migrating upstream.
Spatial
Boundaries
(study)
The Columbia River below Priest Rapids dam, Snake River to the WA, ID border, as well as terminal fisheries in
Snake River tributaries.
Temporal
Boundaries
(study)
Annual
March through June for all mainstem fisheries
May through July for Snake Basin tributary fisheries.
Practical
Constraints:
Budget; time required to analyze sample data in-season.
Spatial
Boundaries
(decisions):
The Columbia River below the mouth of the Snake River, Snake River to the WA, ID border, as well as terminal
fisheries in Snake River tributaries.
Temporal
Boundaries
(decisions):
Annual and in-season when data are available.
5. Decision Rules
48
DQO STEPS
Policy
Inputs1
(9)
Critical
Components and
Population
Parameters:
Harvest number, harvest rate, age and stock composition, escapement by stock.
9
Critical Action
Levels (Effect
Sizes):
Varies by return size. Formulas set by compact and treaty requirements.
9
If-Then Decision
Rules:
9
1. If the catch of upriver spring chinook and Snake River spring or summer Chinook approaches X% of the total
upriver spring chinook and Snake River spring summer chinook run size at the Columbia River Mouth in the
mainstem Columbia River tribal spring management period Zone 6 fishery, then the fishery will be closed. X%
depends on the allowed harvest rate in the management agreement that is based on the updated river mouth run
size. There is a stepped harvest rate schedule in the current mainstem management agreement.
2. If the catch and/or handling mortality of wild upriver spring chinook and Snake River spring/summer Chinook
approaches X% of the wild run size in the mainstem Columbia River non-tribal commercial or select area fishery,
then the fishery will {decision type here – in-season adjustments to effort level, gear type, duration, etc.}. The
decision will depend on if the sport/commercial allocation limit is being approached or if the overall wild impact
limit is being approached.
3. If the catch and/or handling mortality of wild upriver spring chinook and Snake River spring /summer Chinook
approaches X% of the cumulative run in the mainstem Columbia River recreational fishery, select area sport
fishery, or Washington Lower Snake River sport fishery, then the fishery will {decision type here – in-season
adjustments to effort level, gear type, duration, etc.}. The decision will depend on if the sport/commercial
allocation limit is being approached or if the overall wild impact limit is being approached.
4. If the catch and/or handle of the Snake River spring or summer Chinook approaches X% of the cumulative run in
the terminal area tribal fishery in any part of the Snake River Basin, then the fishery will {decision type here – inseason adjustments to effort level, gear type, duration, etc.}. The actual harvest limits in any terminal fishery
depend both on the allowed ESA take if any and the state tribal allocation agreements and escapement
objectives that may be in place in any year.
5. If the catch and/or handle of the Snake River spring or summer Chinook approaches X% of the cumulative run in
the terminal area non-tribal fishery of the Columbia River, then the fishery will {decision type here – in-season
adjustments to effort level, gear type, duration, etc.}. The actual harvest limits in any terminal fishery depend both
on the allowed ESA take if any and the state tribal allocation agreements and escapement objectives that may be
in place in any year.
Consequences of
Decision Errors:
Management is dependent on point estimates rather than on hypothesis testing so a discussion of precision is
relevant as opposed to a discussion of Type I and Type II error. There is no defined precision criteria except that a
20% sample is the goal for species composition.
1Policy
9
Inputs - checked steps indicate where group really needs policy feedback
Steps 6 and 7.
Optimizing the Design (examples)
Evaluation Design (How data will be analyzed to answer a question)
Sampling Design (Where and When data will be collected)
Response Design (What and How data are collected)
Refer to following table
49
From Tables A1 and C1.
For each population:
General
Specific
method to
Performance method to
Level of
Measure
estimate PM estimate PM Resolution Applicability Strengths
Additional cost (e.g.,
access, logistics,
# units per equipment needs,
year (sites addl spls needed not
covered in data
x
Cost per
unit of data replicates) already available )
Stock
composition
Visual Stock
Identification
Biological
sampling for
phenotypic
characteristic
s (light vs.
dark faces).
Above
Bonneville
ESU's as a
whole vs.
Willamette
PIT tags
Individual
Tagging
program and fish,
recovery
program (with
tag
detectors).
Coded-wire
tags
Tagging
program and
recovery
program.
Need fin clip
or wands to
detect CWT
presence so
CWT can be
removed for
later analysis.
Tag group
(typically
10,000200,000
fish)
To Spring
Chinook
below
Bonneville
fisheries or
other stockselectivity
based on run
timing
Weaknesses
quick,
low resolution; subjective,
inexpensive, uncertain accuracy, limited
non-invasive, applicability
Qualitative
assessmen
t of relative Dataset
precision required
L
Biosampling
of catch
List of
hatchery and
wild PIT
tagging
programs
(Jeff)
H
tagging expense, cannot
tag all sizes, tagging
mortality higher compared to
CWT, tags lost if fish
cleaned before sampling,
detection requires
investment in detectors (no
visible mark); not enough
wands in sampling
programs to take advantage
of the existing PIT tags.
Marks
released by
origin and
recoveries
from
appropriatel
y sampled
catch
List of
hatchery and
wild CWT
tagging
programs.
Willamette
Springs
(Jeff/Earl
Trial, Tom OR
Annette WA)
recovery relatively
expensive, data not typically
quickly available, batch
mark, need external mark or
detector to recover tag.
Invasive recovery.
Marks
released by
origin and
recoveries
from
appropriatel
y sampled
catch
Total cost
for data
acquisitio
n
Accuracy
and
Precision Bias of
of Input Input
Data
Data
Accuracy
and
Precision
of Method
Bias of
Method
High
High
High
Low
High
High
High
Low
50
General
Specific
method to
Level of
Measure
access, logistics,
x
covered in data
Cost per
Scale-Pattern Laboratory
Recognition analysis of
scale growth
patterns
(either visual
or
measuring).
Generally
hatchery/wil
d, has
limited ability
to identify
specific
origin
Typically used
to distinguish
H:W and is
also used to
differentiate
the two
primary
Columbia
basin sockeye
stocks.
Weaknesses
Qualitative
assessmen
precision required
Limited application, visual
M/L
scale analysis may be
subjective, training for
proper scale sampling and
interpretation, scale
preparation expense,
usually a delay between
sampling and analysis, need
baseline dataset if scale
analysis is conducted by
measuring scales (and lab
time for this process).
Total cost
for data
acquisitio
n
Accuracy
and
Precision Bias of
of Input Input
Data
Data
Accuracy
and
Precision
of Method
Bias of
Method
Scale
growth
patterns
from
appropriatel
y sampled
catch as
well as
known-stock
samples.
High
Low
Dependent Low if
on degree of correctly
differences used
in growth
characteristi
cs. Method
often works
well for
small
numbers of
stocks (2-3)
but poorly
for larger
numbers of
stocks (5 or
more).
Lab work on
appropriate
tissue
samples
from
sampled
catch
High
High
High
Run
reconstructio
n
Otolith
Research
Needed
Laboratory
analysis of
otolith growth
patterns.
Invasive recovery mark. No M/H
external features to
distinguished marked fish.
Sample preparation time.
Need trained people to
recover otoliths and to
analyze patterns.
Can be used Hatchery
to
rearing
differentiate programs
properly
marked
hatchery
stocks.
Annette talk
with Steve
Schroeder;
Tom talk
with Curt
Melcher
Genetic Stock Micro-satellite Annette talk
Identification technology; with Sewall
DNA analysis to describe
of
status quo.
frequencies in No age
catch
information.
optimized
against
known
frequencies
of donor
populations.
Depends on
coverage of
baseline
information.
Samples
cheap and
simple to
collect from
every fish,
non-invasive
Baseline needs
development (? Talk with
Sewell); data processing
expensive
M
Low
51
General
Specific
method to
Level of
Measure
access, logistics,
x
covered in data
Cost per
Elemental
analysis of
hard body
parts
Incidental
mortality
(post-release
mortality of
unharvested
fish
encountered
in fishery).
Laboratory
analysis of
otoliths or
scales for
elemental
composition
Currently
hatchery
rearing
programs,
potential
applicability
for wild
stocks.
No external features to
distinguish marked fish.
Sample preparation time.
Need trained people to
recover otoliths. Expensive
(except for tetracycline
marks which fluoresce).
FDA concerns with regards
to adding elements to
food/water for hatchery fish.
Extensive research needed
regarding applicability for
wild fish.
Various batch Apply mark or Identify fish
tag and
marking
group.
recover by
methods
examining
fish.
Currently
used for small
programs for
research
purposes
Limited number of marks
available to distinguish
groups of fish. Often poorly
coordinated, difficult to find
origin of batch marked fish.
Angler
interviews/Ru
n
reconstructio
n
Size and
species
selective
fisheries
Release mortality rate is
unknown and must be
supplied from elsewhere
subjecting these estimates
to bias. Depends on
voluntary reporting or
reporting from memory on
releases and angler ability
to differentiate species and
run reconstruction
assumptions.
Interview
anglers
regarding
release rates
are used to
estimate the
number of
fish released
and run
reconstructio
n is then used
to divide the
released fish
into
species/stock
/age
categories
Can be used
to
differentiate
properly
marked
hatchery
stocks.
Some
potential
(research
required) to
differentiate
natural
origin stocks
using natural
differences
in
composition.
Weaknesses
Specific
fisheries and
resolution
subject to
same criteria
as under run
reconstructio
n with stock
composition
Qualitative
assessmen
precision required
M/H
Total cost
for data
acquisitio
n
Accuracy
and
Precision Bias of
of Input Input
Data
Data
Accuracy
and
Precision
of Method
Bias of
Method
Marks
released by
origin and
recoveries
from
appropriatel
y sampled
catch
52
General
Specific
method to
Level of
Measure
access, logistics,
x
covered in data
Cost per
Test fishing
Bycatch to
catch
encounters
(defining
conversion
rate)
Specific
fisheries and
resolution
subject to
same criteria
as under
stock
composition
Size and
species
selective
fisheries
subject to
data (tagging
programs) as
under stock
composition.
Double Index Hatchery
Tag Program programs
release
companion
marked/tagge
d fish and
unmarked/tag
ged fish.
MarkSpecific
fisheries and selective
stocks
fisheries
represented
by the DIT
program
Test fisheries
Time and
general
area.
marked to
unmarked
ratio using
double index
tags.
Post release
mortality
rates
Test fishing
for catch
characteristic
s subject to
same data
consideration
s as stock
composition
rates from
To the
paired
related tag
fisheries,
group.
release rates,
or
escapement
rates.
Weaknesses
Qualitative
assessmen
precision required
Total cost
for data
acquisitio
n
Accuracy
and
Precision Bias of
of Input Input
Data
Data
Accuracy
and
Precision
of Method
Bias of
Method
unknown and must be
to bias. Needs routine
application during run,
otherwise limited by time
and geography.
unknown and must be
to bias. Foregone harvest
on unmarked DIT group,
cost of doubling tagging
program; electronic
detection equipment,
sampling expense (must
sample all fish returning, not
just marked fish)
Size and
species
selective
Mark selective
fisheries
Captive
holding
All selective
To catch
method and fisheries
fishery.
Compare
recovery
rates of
between test
and control
groups of
fish.
Specific
fisheries
measured
on.
Commercial
fisheries,
research
needed to
apply to
recreational
fisheries.
53
General
Specific
method to
Level of
Measure
Weaknesses
access, logistics,
x
covered in data
Cost per
Fishing Effort Aerial
surveys
Count and
Daily to
categorize
weekly
nets or fishing
boats
Limited to
commercial
net fisheries
and
recreational
boat fisheries
Limited by geography,
visibility, safety, and budget
Interview
anglers
regarding
fishing effort
Daily to
weekly
Currently
used for
recreational
fisheries,
could be used
for other
fisheries
Lots of staff time required,
relies on honest answers,
need accessible anglers,
difficult to get a
representative sample
Mandatory
Interview
check station anglers at
check point
on fishing
effort.
Daily to
weekly
Generally
applied to
terminal
fisheries in
restricted
access areas.
Depends on honest answers
from anglers. Limited by
geography.
Landing
Tickets
By sale daily Commercial
fishery at a
licensed
buyer.
Recreational
Angler
interviews
Catch
(number of
harvested
fish)
Each
commercial
sale
generates a
receipt with
species and
weight.
Average
weight
during
fishery
period.
To zone and
Angler
Voluntary
day up to
Harvest
return of
sampling
Record Cards angler
harvest cards
validated
through
followup
survey
Commercial
fishery at a
licensed
buyer.
Widely
applicable,
but restricted
by
participation
Not all fish are sold to
commercial buyers.
Numbers of fish generally
not recorded.
difficult to
apply a
representative
design.
Qualitative
assessmen
precision required
Total cost
for data
acquisitio
n
Accuracy
and
Precision Bias of
of Input Input
Data
Data
Accuracy
and
Precision
of Method
Bias of
Method
Boat counts
categorized
by angler
type
H
Reported
landings
(pounds) by
species.
Number of
fish and
total weigh
of batch
Participation (return of cards L
not enforced). High potential
for bias by nonrepresentative card return.
Delay in processing;
historical record rather than
management tool.
Reported
harvest by
individual
angler
Catch per
Effort
54
General
Specific
method to
Level of
Measure
access, logistics,
x
covered in data
Cost per
Angler
interviews
Interviews of To zone and
anglers
day up to
stratified by sampling
angler type,
weekday/wee
kend/holiday.
Weaknesses
Qualitative
assessmen
precision required
Total cost
for data
acquisitio
n
Accuracy
and
Precision Bias of
of Input Input
Data
Data
Accuracy
and
Precision
of Method
Bias of
Method
Currently
used for
recreational
fisheries,
could be used
for other
fisheries
Mandatory
check station
"Over the
Bank" sales
monitoring
"On-board"
monitoring of
fishing
vessels
Harvest Rate Marked fish
(estimated
exploitation
proportion of estimate
stock that is
harvested in
specific
fishery)
Catch/run
size
Fishery
recoveries of
known
marked fish
population
F/(Ncr-Fo)
[Where: F =
Estimated
harvestrelated
mortality; Ncr
= Estimated
population
abundance at
Columbia
River mouth;
Fo = ocean
harvestrelated
mortality]
55
General
Specific
method to
Level of
Measure
access, logistics,
x
covered in data
Cost per
Age
Composition
of catch
Indicator
stock
monitoring
- Ocean
fisheries
assumed to
have similar
exploitation
rates on fish
from a group
of stocks or
rivers that mix
fully.
PIT tags
Recapture of
PIT tagged
individuals.
High
precision in
total age
assignment
for
individuals.
(May not
provide a
freshwater/sal
twater age
breakdown.)
Requires
recapture of
fish tagged as
known age
juveniles.
Weaknesses
Qualitative
assessmen
precision required
H
Recovery of
sufficiently
large
number of
PIT tags
from fish
that
adequately
represent
the catch.
Total cost
for data
acquisitio
n
Accuracy
and
Precision Bias of
of Input Input
Data
Data
High
Likely
high as
stocks
comprisi
ng catch
not PIT
tagged
proportio
nally
Accuracy
and
Precision
of Method
High, but
may not be
able to
separate
total age
into
freshwater
and
saltwater
age
Bias of
Method
Likely
high due
to catch
not PIT
tagged
proportio
nally
56
General
Specific
method to
Level of
Measure
access, logistics,
x
covered in data
Cost per
Coded-wire
tags
Recapture of
CWT
individuals.
High
precision in
total age
assignment
for
individuals.
(May not
provide a
freshwater/sal
twater age
breakdown.)
Requires
recapture of
fish tagged as
known age
juveniles.
Scale
analysis
Interpret
scale patterns
Weaknesses
Qualitative
assessmen
precision required
M
Scales
collected
from an
appropriatel
y sampled
catch.
Total cost
for data
acquisitio
n
Accuracy
and
Precision Bias of
of Input Input
Data
Data
Accuracy
and
Precision
of Method
Bias of
Method
High
Likely
high as
stocks
comprisi
ng catch
not
tagged
proportio
nally
High
Low if
Moderate, Moderat
properly depends on e
sampled reader,
availability
of knownage scale
samples,
scale
condition,
species, and
run
High, but
may not be
able to
separate
total age
into
freshwater
and
saltwater
age
Likely
high due
to catch
not
tagged
proportio
nally
57
CSMEP – Hydrosystem Action Effectiveness Subgroup
The existence and operation of the Federal Columbia River Power System (FCRPS) is one of the more important anthropogenic
factors influencing mainstem survival of listed and proposed ESUs. Decisions on FCRPS actions can directly or indirectly affect
survival of these stocks. Monitoring of the expected and actual effectiveness of these actions (e.g., juvenile collection, bypass, and
transportation; water management; offsite mitigation) is essential for reliable decisions (Figure 1).
Columbia
River
Ice
Harbor
Lower
Monumental
Little
Goose
Lower
Granite
Snake
River
McNary
The
Dalles
John
Day
Bonneville
In-river
Spawning
& Rearing
Estuary
Transported
Ocean
Figure 1. Illustration of the different scales of
interest for monitoring the effects of the hydrosystem on salmon survival rates: survival at individual projects, survival by different
passage routes through the entire hydrosystem, post-Bonneville survival of different groups of fish, smolt to adult return rates back to
Lower Granite Dam, and overall life cycle survival back to the spawning ground. Moving from smaller to larger scales results in more
58
confounding from other factors (ocean conditions, hatcheries, harvest), but also provides an assessment of indirect effects over a larger
portion of the life cycle. The CSMEP Hydro Subgroup is integrating monitoring designs with other CSMEP subgroups, particularly
related to PIT-tagging.
CSMEP’s Hydro Subgroup took on a subset of hydro management questions across several scales: individual projects, survival by
different passage routes through the hydrosystem, and overall life cycle survival. These different scales relate to a variety of decisions:
operations at individual projects (e.g., spill, bypass, removable spillway weirs); overall operations (e.g., when to transport fish within
season, compliance with hydrosystem biological opinions), longer term hydrosystem decisions (e.g., flow management, effectiveness
of transportation over multiple years, system configuration); and adequacy of hydrosystem operations for stock recovery. Moving
along these scales, the performance measures of interest change. Performance measures range from direct survival at and between
dams, to smolt-to-adult survival rates (e.g., smolts leaving Lower Granite Dam to adults returning there 2-3 years later) to inferences
about delayed mortality from contrasts in mortality patterns (contrasts in recruits/spawner or smolt-to-adult survival rates). At the
largest scales, other factors come into play (estuary, ocean, harvest, etc.) which make it harder to isolate the signal from the
hydrosystem. Appendix A provides the Hydro Subgroup’s summary of the design elements (DQO steps 1-7) for hydro monitoring.
The choices that are available to improve the quality of information for hydrosystem decisions, and reduce the risks of making
incorrect decisions, include: the number of years of data collected, the magnitude of tagging effort, the number of stocks that are
monitored, the ability to filter out year to year natural variation and isolate the signal of management actions, and implementation of
deliberate manipulations of hydrosystem operations to reduce uncertainty in effectiveness evaluations. Our analyses indicate that for
some questions, simply adding more tags won’t improve the quality of information unless one can filter out the effects of year to year
natural variation. CSMEP has developed, and is continuing to assess alternative methods of data analysis that filter out this variation.
For many of these questions, CSMEP has developed low, medium and high alternative designs and explored the strengths and
weaknesses of each approach.
59
Appendix A. CSMEP Hydrosystem DQO Summary
DQO STEPS
Policy
Needed
(9)
Problem:
The existence and operation of the Federal Columbia River Power System (FCRPS) is one of the more
important anthropogenic factors influencing mainstem survival of three ESUs of concern to this Snake River
[SR] pilot study: SR spring/summer chinook, SR fall chinook, and SR steelhead. ESA-listed bull trout are also
affected, but are not considered in this pilot study.
Decisions on FCRPS actions directly or indirectly affecting survival of these stocks are conducted under the
authority of the ESA. Information on the expected and actual effectiveness of these actions (e.g., juvenile
collection, bypass, and transportation; water management; offsite mitigation) is essential for reliable decisions.
There is a need to assess what quality of data are required to: 1) reliably detect the effects of FCRPS actions
on fish survival rates; and 2) reliably compare survival rates to pre-defined goals.
Stakeholders:
NMFS makes FCRPS management decisions under the ESA for the three ESUs considered in this preliminary
analysis. USFWS also assesses FCRPS effects on bull trout under ESA.
Other stakeholders affected by these decisions: state agencies and tribes that co-manage the fisheries
resource; federal fishery agencies that implement ESA and hydropower mitigation management; federal
agencies that operate and market electricity from the FCRPS and fund mitigation activities; commercial,
recreational and tribal fishers: power users.
Non-technical Issues
affecting M & E:
Funding, legal authority to handle and mark fish, legal authority to place detection structures in fishways,
decisions on dam operations that affect detection rates and/or influence contrasts among different groups
(e.g., volume of spill, bypass/barging operations); ongoing BiOp/Remand considerations.
Conceptual Model:
Assessment of hydrosystem impacts involves a suite of four sets of indices: 1) direct survival (project pathway,
reach, entire hydrosystem); 2) SAR overall; 3) SAR ratios (T/I, D, upriver / downriver stocks); and 4)
recruits/spawner (spatial/temporal patterns). Moving from 1 to 4 results in more confounding from other factors
(ocean conditions, hatcheries, harvest), but also provides an assessment of indirect effects over a larger
portion of the life cycle.
Decisions /
Alternative Actions
Hydro Action Effectiveness Questions [Section of Report on DQO steps 6-7; Status of Work to Date 1 ]
[Example feedback required]
Are SARs, and
important SAR ratios
relating to
effectiveness of
transportation,
meeting NPCC and
BiOp targets?
Is SAR sufficient for 1) NPCC goal 2 & 2) recovery goals? [6.1; A]
Is transportation more effective than in-river passage? [6.1; A]
What is the relative survival of transported fish post-BONN, compared to in-river fish (D)? [6.1; A] {no
regulatory target}
Has hydrosystem complied with performance standards set out in 2000 FCRPS BiOp? [6.2; A]
{If targets not met (by how much?), then may need to consider changes in FCRPS operations (e.g., when, how
much to transport) or configuration.}
9
Should FCRPS
change timing of
transportation of
some species within
season?
How does effectiveness of transportation change over the course of the season? [6.5; A] {Are Snake R wild
chinook equally important as wild steelhead? Are wild chinook more important than hatchery chinook?}
9
1
Status of Work to Date: A: Have posed testable hypotheses; explored precision of alternative
evaluation, sampling and response designs; still need to do more work and assess costs; B: Have clarified
questions and described evaluation design; need to do more work on forming hypotheses, assessing precision
of alternative sampling and response designs; C: Have only clarified questions
2
Pg. 13 of NPCC mainstem amendments of 2003-2004. www.nwcouncil.org/library/2003/200311.pdf ; interim goals of 2-6% SAR
60
DQO STEPS
Policy
Needed
(9)
Is the cumulative
What’s the incremental mortality of Snake R fish populations (passing 8 dams) as compared to lower Columbia
effect of hydrosystem stocks passing 1-3 dams? [6.3; B]
actions and estuary- What is the inferred delayed mortality of both in-river and transported fish? [6.4; B]
ocean conditions
leading to stock
recovery?
{No regulatory target}
9
Are current flow and What is the effect of different flow management actions in the hydrosystem on SAR and Sp/Sp ratios? [6.6; B]
spill management
What is the effect of different flow and spill management actions on in-river survival? [6.7; A-B]
actions meeting
{Need to confirm targets}
survival targets?
If not, should FCRPS
change these
actions?
9
Is offsite mitigation
Have freshwater habitat restoration actions been sufficient to compensate for hydrosystem direct and delayed
working?
mortality, as measured on the Snake R aggregate sp/sum chinook stock? [6.9; B]
{No regulatory target}
9
Are dam project
operations
maintaining desired
targets for fish
survival rates and
condition?
{FCRPS Biological
Opinion and other
targets}
9
What are the survival rates and condition of fish past turbines, spillway and bypass routes of passage? How
would RSWs change SARs and Sp/Sp? Would RSWs be an effective alternative to transportation? Would the
reduced spill associated with RSWs affect fish survival and condition? [6.8; C]
Review current operational targets for project survival, fish guidance, etc. with BPA, Army Corps
Decision Statements: Decision rules are within purview of agency with statutory authority. Logically, decision rules should:
• anticipate survival changes in the mainstem;,
• address management measures for all the Hs,
• incorporate adult and juvenile data (consider indirect effects);
• project stock abundance over many decades; and
• accommodate gradual improvements in habitat condition and habitat deterioration that could offset
hydrosystem effects.
9
Information
Required:
Estimates of direct survival rates and SARs for a contrasting range of: mainstem passage timings and routes
(transported vs. in-river; bypass vs. spillway vs. turbine); species (spring/summer chinook, fall chinook,
steelhead, sockeye); stock origins (upstream vs. downstream; wild vs. hatchery).
Estimates of estuary/ocean survival rates are required to assess delayed mortality; these are inferred from
estimates of in-river survival, SARs, recruits / spawner, and the proportion of fish below Bonneville Dam that
were transported.
Estimates of the feasibility of achieving survival improvements across all H’s need to be merged for evaluating
the most promising suite of actions for recovering populations.
Sources of Data:
Direct survival estimates through the hydro system and estuary/ocean through tagging and recapture: coded
wire tags, PIT tags, balloon tags, radio tags, hydro acoustic technologies. Other survival and recruitment
estimates used to estimate estuary/ocean survival, climate/ocean effects and delayed mortality: dam counts,
redd counts, carcass counts, age analysis from scales.
61
DQO STEPS
Quality of Existing
Data:
Data quality varies by the question of interest, species and stock origin:
• Precision of survival estimates greatly improved since the use of PIT tags;
• Precision of survival estimates varies with number of fish tagged: estimates for hatchery
spring/summer chinook and steelhead more precise than for wild fish; estimates for entire year’s
migration more precise than for within-year groups; poor estimates for fall chinook
• Tagging methods to determine the relative survival rates of fish through different dam passage routes
have various weaknesses
• Recruit/spawner estimates have various limitations but also have long time series; contrasts in SARs
provide better signal of hydrosystem effects
• Intensive studies of smolt health and estuary survival not yet linked to SAR data for various PIT-tagged
groups, to understand mechanisms of delayed mortality
New Data Required:
• Higher precision in SARs to improve reliability and speed of responses to key questions
• PIT-tag based SAR data for fall chinook (hatchery)
• Improved ability to assess differences in spring/summer chinook SARs across contrasts in passage
•
•
Analytical Methods:
Policy
Needed
(9)
route, timing, and stock origin
Better linkage of physiology studies below Bonneville with SAR data
See tables and charts at end of this handout
Methods for estimating precision of reach survival, SARs and SAR ratios (e.g., T:I, D) are well developed.
Methods for inferring delayed mortality are indirect, involve many inputs, and are less precise. Challenge is to
develop evaluation methods that filter out natural year to year variation (as well as other confounding factors)
to isolate hydrosystem effects and answer key questions in an acceptable timeframe.
Target Populations:
Snake River spring/summer chinook, fall chinook, steelhead and sockeye examined to date; bull trout also of
interest
Spatial Boundaries
(study)
From entrance into hydrosystem at Lower Granite to various points beyond (reach survival, Bonneville Dam,
estuary, return to Bonneville Dam, return to Lower Granite Dam, return to spawning ground)
Temporal
Boundaries (study)
Studies must be of a sufficient duration to detect the effect of contrasting actions. Thus the required duration of
monitoring depends on the hydrosystem action being evaluated, and the effect size of interest (longer for more
subtle effects). Time scales range from daily detections of PIT-tags, to seasonal contrasts of SARs, to annual
SARs, to decadal-scale contrasts in spawner-recruit data.
Practical Constraints: Difficult or impossible to: determine causes of mortality after fish pass into ocean; disentangle effects of
hatchery operations and # of dams passed on hatchery SARs (these factors covary); relate condition of PITtagged smolts in the estuary with their ultimate SAR.
Not enough wild fish in some years to obtain reliable estimates of mainstem survival rates or SARs; year to
year variation in survival means that ‘average effects’ may hide important information
Spatial Boundaries
(decisions):
Entire Columbia Basin. Decisions on hydrosystem operations and configuration have implications over the
scale of the electricity grid to which generated power is distributed.
Temporal
Boundaries
(decisions):
NOAA and USFWS Biological Opinions on FCRPS are released every 5 to 10 years. Analyses need to
consider actions such as habitat restoration which may take decades to become fully effective.
5. Decision Rules
Critical Components,
Population
Parameters and
Action Levels:
9
• Compare hydrosystem survival rates to NOAA FCRPS BiOp targets,
• Compare SARs to NPCC interim targets of 2-6% and other recovery goals; Compare T:I ratios to
assess transportation benefit (e.g., T/I > 1.0?);
• Compare D to level indicative of substantial delayed mortality of transported fish (e.g., D < 0.7?)
• Compare probability of extinction and probability of recovery to NOAA and USFWS targets
62
DQO STEPS
If-Then Decision
Rules:
If juvenile survival rates through the hydrosystem, SARs or the probability of recovery are consistently below
target levels,
and this can be clearly shown to be related to the direct and indirect effects of the hydrosystem,
then alternative mitigative actions will be considered (e.g., changes to hydrosystem operations, removal of
predators from reservoirs, changes to hydrosystem project structure or configuration)
Consequences of
Decision Errors:
Failure to make required changes in hydrosystem operation or configuration may result in extinction of fish
species, or failure to recover stocks.
Making ineffective changes in operations or configuration may waste significant amounts of money.
Steps 6 & 7.
Summary of
alternative M & E
designs for
assessing
effectiveness of
hydrosystem
actions
CSMEP developed and evaluated several sets of alternative designs for the questions outlined in steps 1-5.
This work is ongoing and is summarized in a detailed report on DQO steps 6-7. The following pages provide
tables, graphs and charts to illustrate some alternative designs (low, medium and high intensity) developed by
CSMEP for a subset of the questions described above under step 2. Increased levels of tagging would help to
improve the precision of answers to all hydrosystem questions. Ultimately these low, medium and high designs
need to be synthesized across all hydro questions, and then across the questions from all subgroups.
Policy
Needed
(9)
9
63
Table A-1. Alternative designs for hydro questions related to SAR, T/I, and D. Current M & E efforts
are shaded in grey. The low option is indicative of less M & E than current efforts, whereas
the high option involves increased effort. Charts A-1 to A-4 (end of this Appendix) describe
low, medium and high M & E options for spring/summer chinook and steelhead in greater
detail.
M & E Level Spring/summer Chinook
Fall Chinook
Steelhead
Sockeye
Low
Background level of PITtagging. Run reconstruction
SARs: partition wild and
hatchery smolts and adults at
upper dam (e.g., Petrosky et
al. 2001; Raymond 1988)
PIT tag hatchery group only
as surrogate for wild SAR??
Run reconstruction SARs:
partition wild and hatchery
smolts and adults at upper
dam (e.g., Marmorek et al
1998; Raymond 1988).
SARs: run reconstruction at
Redfish Lake; T/I and D
unsampled, assume response
similar to or worse than
spring/summer chinook??
Medium
PIT tag wild aggregate SARs
from opportunistic tagging
(current CSS); PIT tag SARs
major production hatcheries.
Estimate annual T/I and D for
wild and hatchery groups
(current CSS)
PIT tag wild fall Chinook
above LGR, treat same as
CSS fish; PIT tag SARs
hatchery and a surrogatenatural group (fall chinook
transport evaluation proposed
for 2006)
PIT tag wild aggregate SARs
from opportunistic tagging
(current CSS since 2002); PIT
tag SARs major production
hatcheries (proposed CSSnot funded)
SARs: PIT tagged migrants
from Redfish Lake, treated
same as CSS fish in FCRPS
(beginning 2005); T/I and D
not feasible because
extremely limited potential
sample sizes
High
PIT tag for wild SARs at level
of Major Population Groups
(MPG); all hatchery releases
represented in SAR, T/I and
D estimates
Same as medium??
PIT tag for wild SARs at level
of Major Population Groups
(MPG) (feasibility ??); all
hatchery releases
represented in SAR, T/I and
D estimates
Not possible to go beyond
Medium level.
Probability density functions of CSS control and transport
SARs of wild chinook for migration years 1994-2002
1.2
Relative Prob. density
1
Control (C0)
0.8
Transport (T0)
Target
Minimum
`
0.6
0.4
0.2
0
0%
1%
2%
3%
4%
5%
6%
7%
8%
9%
10%
SAR
Figure A-1.
Probability density functions of SARs for wild Chinook, for both control (blue) and
transported (red) fish, between 1994 and 2002, compared to the NPCC interim goal of 26% (green line at 2%). Most of the fish in both groups had an SAR lower than the
minimum end of the 2-6% desired range.
64
Table A-2. Alternative designs to assess if the hydrosystem has complied with performance standards set
out in 2000 FCRPS BiOp. More specifically, is inriver survival of wild Snake spring-summer
chinook and steelhead at or above the 2000 BiOp 50% (approx.) standard?
Background: The BiOp standard is that survival should be only 9% greater than pre-BiOp
estimates of survival from LGR to BON. There are four challenges in attempting to
monitor for compliance with this standard: 1) smaller changes in survival are easier for
managers to achieve and more “certain to occur”, but also harder for biometricians to
detect; 2) inter-annual variation is high and monitoring cannot change this fact; 3)
estimated survival for any given year (1998-2004) has had high error bounds (generally
+/- 10-15%); and 4) there are few data from before 2000 (Figure A-2).
M & E Level
Monitoring Activities
Low
Decrease tagging and detection efforts from current level (Medium option), wait 5-10 years for data to accumulate, and
compute a new, multi-year average survival rate. However, years with exceptionally low or high flow, temperature, etc.
(e.g., 1999, 2001, 2005) make it unlikely that the spread in the estimate will go down just because one has more years of
data. On the other hand, lower tagging effort would obviously decrease costs. Decreased spill would increase fish
detection and precision, but will also likely decrease survival rates.
Medium
Continue current tagging and detection efforts. With current tagging (e.g., 20,000 - 30,000 wild chinook tagged or detected
at LGR each year), estimates of LGR to BON in-river survival have wide error bounds. If one pools estimates for 19982004, survival rate on average was about 47%, +/- 9%.
High
Increase tagging substantially (e.g., 2-10 fold), and increase below-Bonneville trawls significantly (e.g., 2-5 fold). Continue
efforts to increase detection efficiency at the BONN corner collector, which diverts many fish away from turbines (good)
but detects few if any tagged fish (bad). Preliminary results suggest that these steps in combination would reduce the
spread in any given year’s survival estimates quite substantially. However, inter-annual variation in survival rates is clearly
a real phenomenon, and not just an artifact of sampling/tagging/detection efforts. This variation will of course continue into
the future. Obviously, this would result in increased costs for both tagging and (trawl) detections in the estuary below
BON.
Wild Snake spring-summer chinook inriver survival, LGR to BON
0.90
Survival Rate from PIT tags
0.80
0.70
0.60
0.50
0.40
0.30
0.20
0.10
Upper 95%
Lower 5%
Mean survival
Standard
0.00
1997
1998
1999
2000
2001
2002
2003
2004
2005
Migration Year
Figure A-2.
The 95% confidence intervals about estimates of mean annual in-river survival rates of
wild Snake spring-summer chinook generally bracket the 2000 FCRPS standard of
49.6%.
65
Alternative Designs for the question “What’s the effect of different in-season transportation
management actions on Post-Bonneville survival of transported fish?” These designs are likely to be
similar to those listed above in Table A-2 for compliance with BiOp survival targets.
Background:
Preliminary NOAA and CSS work suggests that Transport/In-river SAR ratios may vary
within a season. For Snake River wild spring-summer chinook – early migrants may do
better in-river, but later migrants do better in barges. In addition, it appears that hatchery
chinook, and both wild and hatchery steelhead, do not exhibit a strong seasonal pattern –
transport is generally better, both early and later in the spring. Hence there are tradeoffs
among species when making in-season transport decisions.
Wild spring-summer Chinook TIR's, 1995-2002
Transport-Inriver Ratio
4.0
TIR - 5th%
Average TIR
TIR - 95%
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0
1
2
3
4
5
Quartile (1-4) of passage season
WIld Snake Steelhead TIR's, 1995-2002
Transport-Inriver Ratio
4.0
TIR - 5th%
Mean Bootstrap TIR
TIR - 95%
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0
1
2
3
4
5
Quartile (1-4)
Figure A-3.
Transport to Inriver SAR ratios (TIRs) for wild spring summer Snake River chinook (top
graph, gradually improving ratio over the season) and wild Snake River steelhead
(bottom graph, little pattern).
66
Table A-3. Alternative designs for estimating upstream-downstream differential mortality and SAR
ratios. Current M & E efforts, or developed proposals, are highlighted in grey. SR=Spawner
Recruit.
M&E
Level
Spring/summer Chinook
Fall Chinook
Steelhead
Sockeye
Low
SR estimates, upstreamdownstream incremental
mortality, wild index stocks from
Snake River and downriver
regions (Deriso et al. 2001). No
program for SARs.
No SAR data yet for either wild or
hatchery; will be getting hatchery
SAR data in 2005 with NOAA
study. Snake R and Deschutes R
SR data.
SR estimates only. No program
for SARs.
n.a.
Medium
SR estimates for upstreamdownstream incremental
mortality, plus SARs from PIT-tag
studies for Snake River and John
Day wild stocks and SARs for
Snake River and downriver
hatchery stocks (Carson &
Leavenworth) (Figure A-4).
SR estimates from Snake and
n.a.
downriver stocks, SARs for Snake downriver stocks, SARs from PIT
tag studies for Snake River wild
and Deschutes wild fall chinook.
stocks (current CSS) plus
additional PIT-tag SARs from wild
downriver stocks (John Day).
High
Elements from medium level plus
more representative wild stock
composition for both SR and SAR
estimates, both regions.
(candidates: Warm Springs,
Yakima, others – monitor Major
Population Groups consistent with
Status and Trends subgroup)
downriver stocks, SARs for Snake
and multiple downriver wild and
hatchery stocks. Candidate wild
stocks: Deschutes, N Fk. Lewis,
Hanford Reach
*
n.a. *
downriver stocks, SARs for Snake
and multiple downriver wild and
hatchery stocks. Candidate wild:
John Day, Deschutes, Warm
Springs
There is no comparable downriver sockeye stock. One could include Wenatchee and Okanagan sockeye stocks to get a
general comparison basin-wide.
Delayed mortality of in-river fish
Delta Model for Snake River Spring Summer Chinook
1.0
0.8
0.6
0.4
0.2
0.0
1970
1975
1980
1985
1990
1995
2000
Brood Year
Figure A-4.
Updated estimates suggest delayed mortality of in-river fish has remained high even as
ocean conditions improved in the late 1990s.
67
Hatchery Chinook:
10,000 tags @Hat +
10,000 tags @traps
to get in-river survival
LOW:
Hatchery Chinook:
200,000 tags @ 5 Hat.
for CSS to get transport
& in-river SARs
MED: + +
HIGH:
Chart A-1.
L
O
W
Hatchery Chinook:
35,000 tags @ 3 Hat.
for NPT to get overall
SARs for sub-basins
Add 100,000 tags at
non-CSS/NPT hatcheries:
clwh, pahp, sawt, koos
Total in system above LGR
of approx. 355,000 tags
Increase in PIT-tagging effort for hatchery Chinook from baseline (LOW level) with
releases for in-river survival estimation purposes to current (MED level) with larger
releases for SAR estimation with transport and in-river migrants. The HIGH level
increases tagging effort at hatcheries not currently covered within MED level.
68
LOW:
Hatchery Steelhead:
16,000 tags @Hat +
12,000 tags @ traps
Hatchery Steelhead:
80,000 tags @ 2 Hat.
for CSS to get transport &
in-river SARs
MED: +
HIGH:
Chart A-2.
Coordinate with
LOW level to route
a proportion of fish
to get transport
SARs
Proportionally PIT-tag 3% of
hatchery production (300,000 tags)
to get transport & in-river SARs;
Total in system above LGR of
approx. 312,000 tags
Increase in PIT-tagging effort for hatchery steelhead from baseline (LOW level) with
releases for in-river survival estimation purposes to MED level adding two hatchery
releases for SAR estimation of transport and in-river migrants. The HIGH level increases
tagging to cover 3% of overall production, proportional to individual hatchery releases
number, across the basin.
69
Wild Chinook:
40,000 tags @ traps, etc.
LOW:
MED: + +
HIGH:
Chart A-3.
Wild Chinook:
26,000 tags @ traps
for CSS to get
transport &
in-river SARs
L
O
W
Wild Chinook:
20,000 tags @ 2 traps
for NPT to get overall
SARs for sub-basins
Add 100,000 tags across trap
sites in MED level to improve
SARs for both transport/in-river
and overall sub-basins;
Total in system above LGR of
approx. 186,000 tags
Increase in PIT-tagging effort for wild Chinook from baseline (LOW level) with releases
for in-river survival estimation purposes to current (MED level) with larger releases for
SAR estimation with transport and in-river migrants. The HIGH level increases tagging
effort at each trap covered within LOW and MED levels.
70
LOW:
MED: +
HIGH:
Chart A-4.
Wild Steelhead:
25,000 tags @ traps, etc.
Wild Steelhead:
1,400 tags @ 1 trap
for CSS to augment
LOW level
Coordinate with
LOW level to route
a proportion of fish
to get transport
SARs
May be impossible to increase
PIT-tagged wild steelhead
numbers over LOW and MED
levels above.
Difficult to increase PIT-tagging effort for wild steelhead from baseline (LOW level) due
to low numbers available for tagging, so emphasis in MED level has been a limited
increase in tagging effort with coordination of the routing of more available tagged wild
steelhead to transport for SAR estimation of transport and in-river migrants. A HIGH
level may not be reachable in the near future from the basin above LGR.
71
Hatchery Chinook:
Too few PIT-tags for
estimating SARs
LOW:
MED: +
HIGH:
Chart A-5.
Hatchery Chinook:
Lower Columbia stock
15,000 tags @Carson H for
Wind R
Hatchery Chinook:
Mid-Columbia stocks
15,000 tags @Leav. H
for Wenatchee R
40,000 tags CleElum H
for Yakima River
Add Lw Columbia stocks:
15,000 tags @Warm Spg H
for Deschutes R
15,000 tags @Irrigon H for
Umatilla R
Total system= 100,000 tags
Increase in PIT-tagging effort for hatchery Chinook from current (MED level) to HIGH
level by increasing tagging effort at hatcheries not currently covered within MED level.
MED and HIGH provide improved upstream-downstream contrasts.
72
Wild Chinook &
Wild Steelhead:
Too few PIT-tags for
estimating SARs
LOW:
Wild Chinook:
6,000 tags @John Day R
Wild Steelhead:
6,000 tags @John Day R
MED: +
Unknown Potential
HIGH:
Chart A-6.
Current PIT-tagging effort for wild Chinook and Steelhead in John Day River (MED
level) may be difficult to reproduce in other Lower Columbia River tributaries, so
potential for HIGH level is presently unknown. MED provides improved upstreamdownstream contrasts.
73
CSMEP – Habitat Action Effectiveness Subgroup
Habitat actions are considered a cornerstone of recovery strategies for Columbia River
Basin fish stocks but there is a need to more clearly determine the effectiveness of these
actions for increasing salmonid survival rates and production. Monitoring designs for
evaluating the effectiveness of habitat actions must be able to reliably detect two linked
responses :
1. the effect of habitat actions on fish habitat; and
2. the effect of changes in fish habitat on fish populations.
Appendix A1 provides the Habitat Subgroup’s summary of the general design elements
(DQO steps 1-5) for Habitat monitoring that can help address these questions.
CSMEP’s Habitat Subgroup has recognized, however, that there are serious challenges to
the development of a generic template for habitat effectiveness monitoring. These
include:
1. Habitat conditions vary greatly across subbasins in terms of their natural
biogeoclimatic regimes, the status of their fish populations, the degree of human
impact and management, and the number and nature of restoration actions that
have been implemented, or are being considered for implementation within them.
2. Habitat effectiveness questions encompass different scales of inquiry, which
imply different scales of monitoring.
The subgroup is instead attempting to develop a consistent “process” that can be applied
to development of individual monitoring designs dependent on the particular situation.
They are piloting this approach within the Lemhi Subbasin. Appendix A2 provides a
summary of the subgroup’s detailed DQO steps 1-7 for the Lemhi Subbasin.
74
Appendix A1. CSMEP Habitat DQO Summary
(General Assessment DQO Steps 1 – 5)
DQO STEPS SNAKE RIVER BASIN PILOT
Policy
Inputs
1
(9)
Problem:
Habitat degradation and loss of connectivity are considered key factors in the
decline of CRB anadromous and resident salmonid populations. Habitat
actions are considered a cornerstone of recovery strategies but there is a need
to more clearly determine the effectiveness of these actions for increasing
salmonid survival rates and production.
Stakeholde States—Washington, Oregon, Idaho
rs:
Tribes—NPT, SBT, CTUIR, CTWIR, YIN. Colville Tribes
Federal—NOAA, USFWS, BPA, USACOE
Other—NPPC, CBFWA, conservation groups, Tribal, commercial, sport
fishers, landowners & local soil conservation districts
Nontechnical
Issues:
Lack of funding; Landowner Permission; Uncoordinated processes; Ill-defined
scope and objectives; Jurisdictional overlap (e.g., state, tribal, federal,
international boundaries, local regulations); Legal constraints and adjudication
Conceptual Habitat actions will first increase habitat distributions and/or improve habitat
conditions, and the improved conditions will lead to increased habitat use,
Model:
improved fish condition, reach-scale abundance, and watershed scale fish
survival and productivity. Thus the problem has three components:
1) detect the effect of habitat actions on habitat,
2) detect the effect of changes in habitat on fish populations, and
3) detect the overall effect of habitat actions on fish populations.
The scale of effects of actions on habitat may vary, ranging from the local
target action area up to the entire watershed. Effects of actions on fish
populations could range from individual fish up to the larger population, the
extent of which may be dependent on a species life history characteristics.
Principal
1. Have specific habitat projects affected habitat conditions and local fish
Questions:
distribution, population survival, abundance or condition?
Includes
• What are the species, down to life-history type and gender, of interest?
the key
• To what factors do you want to be able to attribute the observed
policy
population response?
questions 2. Did groups of habitat projects within a subpopulation or sub watershed on
(numbered)
aggregate affect fish survival, abundance or condition in a larger
and the
demographic unit?
clarificatio 3. Are particular classes of habitat projects effective?
n elements
• Are there surrogate measures you can use to answer your questions?
(bulleted)
4. What are the mechanistic connections between habitat actions and fish
9
75
required for
population responses?
biologists
• What is the spatial boundary of the population for which inferences must
to translate
be made?
these
• Over what time period(s) do you want to describe this population
questions
response?
into M & E 5. Have habitat projects achieved the expected improvements in conditions?
in the field
• What is the population response variable you want to evaluate to
(e.g.,
determine whether a change has occurred?
Lemhi
• What is the reference and final condition (i.e., the change to be defined
example)
in the population response variable)?
• What size of change in population response do you want to be able to
detect?
• What tradeoffs between uncertainty, errors and costs are you willing to
accept
(DQO Steps 6 and 7)
Alternative Maintain current program and designs of habitat actions
Actions:
Make adaptive management changes to design of current habitat actions to
improve performance and increase benefits to fish populations.
Discontinue habitat actions as currently designed, adopt different strategy for
restoring fish populations
Decision
Is the current program of habitat actions achieving the objectives for improved
Statement: fish habitat and fish population performance measures so that program
modifications, expansions, or elimination are not required?
Policy
Inputs
1
(9)
9
9
Action
Levels
(critical
effect
sizes):
Quantitative performance standards need to be specified against which the
results of monitoring can be compared. At this time there are few examples of
such quantitative values available for the evaluation of habitat actions in the
Columbia River basin. Without this guidance it will be necessary to calculate
the ability of alternative designs to detect a range of effect sizes that bracket
important action levels. These action levels will vary with the scale of the
monitoring question (e.g., project level, subbasin, ESU). The minimum
detectable effect at different scales will often not be known and may need to
be guessed at by analysts until verified in the field.
9
Informatio Data/inventories of past, ongoing, and planned habitat restoration activities
n Required: (and their hypothesized effects and sequencing)
Data/Inventories of past, ongoing, and planned fish and habitat monitoring.
Sources of State, tribal and federal programs and NGOs identified in CSMEP meta-data
inventories (which may include information from the following sources: AA
Data:
& NOAA habitat action inventories; IAC/SRFB; IDQ; USFS (AREMP and
PIBO); USFWS
Quality of
Data available apply generally only to Snake spring-summer chinook, and
76
Existing
Data:
only to actions affecting parr-to-smolt or parr-per-spawner life stages.
Additionally, these data were collected through programs that were not
designed to evaluate habitat project effectiveness. CSMEP data inventories
found little information on programs that specifically collected data to assess
the effectiveness of habitat actions in the Snake River (a situation likely
common throughout the Columbia Basin)
New Data
Required:
New data and sampling approaches are required. There is a need for
statistically valid sampling of both fish and habitat. Better spawner and smolt
enumeration may be necessary to detect changes in these metrics due to
habitat actions. It appears feasible to obtain this in many locations in the
Snake (but not everywhere).
Analytical
Methods:
B-A, or BACI designs, where differences between before and after treatment
values of performance measures may be compared to Action Levels using a ttest or confidence intervals. To account for important covariates and
confounding factors, it may be necessary to apply more complex analytical
models.
Examples from recent work include linear regression models, non-linear
neural networks, and multivariate models. However, because these
applications are quite novel, with few published analyses completed to date, it
is difficult to predict exactly what methods will be required, especially for
detection of habitat action effects on fish survival and productivity.
Policy
Inputs
1
(9)
Target
Snake River Spring and Summer Chinook (current focus of Lemhi example)
Populations (with linkages to Upper Columbia summer Chinook due to lower river
:
harvest)
Redfish Lake Sockeye
Snake River steelhead
Bull Trout (also addressed in the Lemhi example)
Spatial
Watersheds within the lower Snake River ESU
Boundaries (Lemhi subbasin as focal example)
(study):
Temporal Monitoring duration: Until actions shown to be effective or not
Boundaries Update schedule: example - beginning in year x at 3, 5, 7, 12, and 15 year
(study):
intervals, and each 4 year period subsequent
Time scale over which the data vary: 3-12 years
Practical
Funding
Constraints Access to sample site, project locations, or data.
:
Statistical constraints such as feasibility of acquiring required data
Spatial
Watersheds/ESU within the lower Snake River
Boundaries
77
Policy
Inputs
1
(9)
(decisions):
Temporal Federal Recovery Plan is scheduled to be completed in April 2006
Boundaries Adaptive Management schedule (plan check ins are recurring intervals)
(decisions): 2004 FCRPS BiOp is a 15 (?) year plan with milestone and check-in
State Recovery Plans have 25 year planning cycle
Subbasin Plans have 15 year planning cycle
HCP is a 30 year plan
5. Decision Rules
Critical
Component
s and
Population
Parameters:
1)
2)
3)
4)
Changes in habitat quantity
Changes in habitat conditions (quality)
Change in smolts per spawner resulting from habitat actions
Changes in parr-to-smolt survival rates from actions
Critical
Action
Levels
(Effect
Sizes):
- these
need to be
clearly
defined
1)
2)
3)
4)
Changes in habitat quantity - X% increase
Changes in habitat conditions – X% goes from poor to good?
Change in smolts per spawner must be at least X%?
Changes in parr-to-smolt survival rates must be at least X%?
If-Then
Decision
Rules:
If the observed change in the critical population components between
treatment (project) and control locations, before and after the implementation
of the project is positive and greater than or equal to the critical action level
then do more of these project types in similar locations.
If an effect is not detected, then the process moves through the adaptive
management sequence to assess whether the monitoring and evaluation
program was sufficient to be able to detect such a change, or whether the
management action, or Action Level criteria need to be changed.
Consequen
ces of
Decision
Errors:
May continue/expand actions that have little beneficial effect (Type I error);
May discontinue actions that really do work (Type II error);
Undue or increased cost;
Continued loss of fisheries;
Negative impacts to state and local economies;
Federal trust responsibilities not met;
Adjudicated requirements not met
9
9
9
9
1
Policy Inputs - indicates with a check steps where group needs greater policy level feedback, presentation will elaborate on what feedback is
required
78
Appendix A2. CSMEP Habitat DQO Summary
Lemhi Basin Example
Specific Assessment DQO Steps 1 – 5
Detailed Monitoring Designs DQO Steps 6 - 7
DQO STEPS LEMHI BASIN EXAMPLE
Policy
Inputs
1
(9)
Problem:
As part of an extended Habitat Conservation Plan (HCP) to restore salmonid populations in
the Lemhi a number of water conservation projects are to be implemented in the basin,
primary of which are a series of approximately 10-16 actions to reconnect currently isolated
tributaries to the mainstem Lemhi River in combination with reestablishment of the
historical hydrograph. Evaluating the cumulative success of these actions across a range of
fish performance measures is the focus of Lemhi M & E efforts.
Stakeholders: IDFG, Shoshone Bannock Tribes, Local landowners, Office of Species Conservation,
Upper Salmon Basin Watershed Project, NOAA Fisheries, USFWS
Non-technical Landowner relationships, lack of funding, interagency coordination
Issues:
Conceptual
Model:
The underlying assumption of the HCP is that as habitat conditions are improved, fish will
respond and desired biological effects will be achieved. The conservation objectives are 1) to
provide adequate flow to remove or reduce migration barriers, 2) maintain or enhance
riparian conditions, and 3) improve instream conditions with respect to cover, temperature,
flow, and sedimentation. The desired actions are: 1) reconnect tributaries to the Lemhi River,
2) alter channel morphology to address fish passage, 3) minimize fish entrainment, 4)
enhance spawning and rearing habitat, 5) maintain minimum flows, 6) improve riparian
corridors, 7) mimic the natural hydrograph. Some are these actions will be quite local, while
others will address the entire Lemhi watershed.
Principal
Questions
Have the actions implemented under the Lemhi HCP:
• Expanded the distribution of rearing juvenile salmonids?
• Increased the density of rearing juvenile salmonids?
• Increased the number of chinook smolts leaving the Lemhi River?
• Caused any changes in seasonal migration pulses and size distribution of Chinook
smolts leaving the Lemhi River?
• Increased abundance of bull trout in reconnected tributaries?
• Increase parr-smolt survival of juvenile Chinook leaving the Lemhi?
• Increased returns of adult Chinook salmon to the Lemhi basin?
9
Alternative
Actions:
Maintain current Lemhi HCP program of habitat actions
Make adaptive management changes to design of current habitat actions to improve
performance of HCP habitat actions and increase benefits to fish populations.
Discontinue plans for HCP habitat actions as currently designed, adopt different strategy for
restoring fish populations
9
Decision
Statement:
Is the current program of habitat actions achieving the objectives for improved fish habitat
and fish population performance measures so that program modifications,
reductions/expansions, or elimination are not required?
9
79
Policy
Inputs
1
(9)
Information
Required:
Habitat Performance Measures:
1. Temperature – creation of summer refugia in reconnected tributaries & adjacent
main stem
2. Flow – increased ease of passage & survival of adults & juveniles
3. Substrate & channel characteristics – increase amount of optimal spawning/rearing
habitat
Fish Performance Measures:
1. Spatial distribution (chinook parr, steelhead parr/smolts, all bull trout)
2. Parr density (chinook)
3. Smolts per redd (chinook)
4. Migratory timing & size (chinook)
5. Population abundance (bull trout)
6. Parr-to-smolt survival (chinook)
7. Redd counts (chinook) – to account for effect of seeding level and changes in
spawning distribution.
8. Spawning adults (chinook) – weir counts, to account for effect of seeding level
Sources of
Data:
IDFG Chinook redd counts, IDFG snorkel surveys, IDFG juvenile screw traps, IDFG
tributary surveys (bull trout redd counts, electrofishing surveys, tissue sampling), IDFG PIT
tag detectors at diversion bypasses in Lower Lemhi, Idaho State University telemetry
tracking of bull trout in upper Lemhi and Hayden Creek, IDWR flow and temperature gauges
at several sites, USGS flow gauges, flow modeling by BoR and University of Idaho, IDEQ
FLIR flight of Lemhi mainstem, IDFG water temperature monitoring at remote sites in
mainstem and tributaries, baseline instream and riparian habitat inventory (1994) by multiagency group, PIBO reach inventories.
Quality of
Existing Data:
New Data
Required:
Analytical
Methods:
•
•
•
•
Long time series of consistently done single pass-pass chinook redd counts
Little hatchery influence on datasets (no hatcheries on Lemhi)
Estimates of outmigrating juveniles available from traps
Adult weir is needed on Big Timber Creek tributary to evaluate movements of fluvial
trout
• Expanded telemetry tracking of trout in Upper Lemhi
• Increased in the number and frequency of parr density surveys
• Systematic steelhead abundance estimates
• More information in general is required for other areas of the Lemhi watershed,
particularly Hayden Creek and the lower mainstem
B-A, or BACI designs, where differences between before and after treatment values of
performance measures may be compared to Action Levels using a t-test and confidence
intervals. To account for important covariates and confounding factors, it may be necessary
to apply more complex analytical models.
Preliminary designs divide the Lemhi into three Sections:
• Section A – mainstem Lemhi and tribs below Hayden Creek. Tentatively an additional
control area.
• Section B – mainstem Lemhi and tribs above Hayden Creek. Tentatively the Treatmen
area.
• Section C – Hayden Creek and tribs. Tentatively the Control area.
80
Target
Populations:
Sp/Summer Chinook
Bull Trout
Spatial and
Temporal
Boundaries
(study):
The sampling design (where, when, and for how long the protocols are activated) is
dependent on the spatial contrast and protocol of interest. For example, if a snorkeling
protocol were activated to address the effects of channel reconnection in the Lemhi
watershed, randomly selected sites could be snorkeled in treatment and control areas of the
Lemhi inter- and intra-annually for a period of 20 years. Five-year check-ins could be
included for progress evaluation. Alternatively, if a snorkeling protocol were activated to
address the effects of channel reconnection in tributary/mainstem junctions, snorkeling
would occur at fixed and random sites only within the treatment areas on an inter- and intraannual basis for a period of 20 years. For either question, sampling intensity (number and
size of the sample units) will be determined based on desired statistical attributes (accuracy,
precision, and power)
Practical
Constraints:
Funding
Access to sample sites, project locations, or data.
Statistical constraints such as feasibility of acquiring required data
Inherent variability of the Lemhi system
Spatial
Boundaries
(decisions):
Lemhi Basin
Temporal
Boundaries
(decisions):
Habitat Conservation Plan (HCP) for Lemhi Basin is 30 years duration)
Policy
Inputs
1
(9)
5. Decision Rules
Critical
Have the actions implemented under the Lemhi HCP expanded the distribution of rearing
Components juvenile salmonids within the basin and increased the density of rearing juvenile salmonids
and Population relative to average mainstem densities by X% over 30 years (with some precision) when the
Parameters
number of spawners, natural disturbances, climate indicators, and habitat conditions not(key
impacted by the actions have been accounted for?
examples):
Have the actions implemented under the Lemhi HCP produced at least a 100% increase in
the number juvenile spring Chinook salmon leaving the Lemhi River in 30 years (+/- X%)
when the number of spawners, natural disturbances, climate indicators, and habitat conditions
not-impacted by the actions have been accounted for?
Have the relative magnitudes of the seasonal migration pulses and size distribution of
migrating Chinook juveniles leaving the Lemhi River changed over the life of the Lemhi
HCP?
Have the actions implemented under the Lemhi HCP increased the abundance of bull trout in
reconnected tributaries relative to unconnected tributaries by X% over 30 years (with some
precision)?
Have the actions implemented under the Lemhi HCP increased parr-smolt survival (X% +/specified precision) of juvenile spring Chinook salmon leaving the Lemhi River in 30 years
when the number of spawners, natural disturbances, climate indicators, and habitat conditions
not-impacted by the actions have been accounted for?
Have the returns of adult Chinook salmon to the Lemhi basin increased X% (+/-specified
precision, see VSP criteria developed by ICTRT) of the life of the Lemhi HCP?
9
Critical Effect Have not been defined for the Lemhi HCP
Sizes:
9
81
If –Then
Decision
Statements:
These have not yet been defined for the Lemhi HCP (i.e., what would be the appropriate
response if the actions do/do not result in expected improvements in habitat/fish performance
measures)
Consequences May continue/expand actions that have little beneficial effect (Type I error);
of Decision
May discontinue actions that really do work (Type II error);
Errors:
Undue or increased cost
Policy
Inputs
1
(9)
9
9
82
Steps 6 and 7.
Optimizing the Design
(examples)
Evaluation Design
(How data will be analyzed to answer a question)
Question 1: Has the spatial
distribution of juvenile chinook
changed as a result of tributary
reconnections?
L: ($ 285K/year; price includes snorkeling, seining, Hayden Creek (section C), Upper
electroshocking, tagging efforts)
Lemhi (section B), and lower Lemhi
Compare connected and unconnected tribs within each (section A), including tributaries –
several times per year for each site.
of sections A, B, C using BACI like design
w/covariates
Snorkel surveys to estimate parr
numbers, several times per year for
each site. Verify detection rates
with multi-pass electro-shocking.
M: ($354K/year; price includes snorkeling, seining, Add more sites to Hayden Creek (C),
Upper Lemhi (B), and lower Lemhi
Compare connected and unconnected tribs within each (A), including tributaries – several
times per year for each site
of sections A, B, C using BACI like design
w/covariates
with multi-pass electro-shocking
H: ($421K/year; price includes snorkeling, seining,
Same number of sites as M design but
more effort towards tagging in Hayden
Compare connected and unconnected tribs within each Creek ( C), Upper Lemhi (B), and
lower Lemhi (A), including tributaries
of sections A, B, C using BACI- like design
– several times per year for each site
w/covariates.
with multi-pass electro-shocking
L: (60K/year; Cost assumes Question 1 work paid for Hayden Creek (Section C), Upper
PIT tagging of all fish captured + cost of screwtrap
Lemhi (Section B), and lower Lemhi
operation)
(Section A) once per year.
Compare between sections A, B, & C using BACI-like
design w/covariates.
Collection and tagging of parr
during parr surveys plus collection
of parr in screwtraps and detection
of tagged smolts at Lower Granite
dam.
M: ($60K/year; Cost assumes Question 1 is paid for Hayden Creek (C ), Upper Lemhi (B),
tagging of all fish captured with PIT tags + cost of
and lower Lemhi (A) once per year.
screwtrap operation)
Compare between sections A, B, & C using BACI-like
design w/covariates.
This design provides more power than the low design
because more fish are tagged at more sites. There are
no additional tagging costs because these are included
in work to address Question 1.
of tagged smolts at Lower Granite
dam.
Question 2: Have the actions
implemented under the Lemhi
HCP increased parr-smolt
survival of juvenile spring
Chinook salmon leaving the
Lemhi River?
Sampling Design
(Where and When data will be
collected)
Response Design
(What and How data are collected)
83
Steps 6 and 7.
(examples)
Evaluation Design
Sampling Design
(Where and When data will be
collected)
Hayden Creek (C), Upper Lemhi (B),
H: ($69-97K/year; Cost assumes all fish captured
under Question 1 were tagged with PIT tags + cost of and lower Lemhi (A) and tributaries
multiple times a year.
screwtrap operation+ cost of either 3 mainstem PIT
tag detectors, 11 mainstem detectors, or mainstem and
tributary detectors)
Compare connected and unconnected tribs within each
of sections A, B, C using BACI- like design
w/covariates.
This is the most powerful design. The addition of
several PIT tag detectors, multiple treatment and
control sites and more tagged fish will allow better
estimation of large scale effects on fish survival and
distribution as well as the development and testing of
relationships between habitat changes and fish
survival.
Response Design
of tagged parr and smolts in
tributaries and mainstem Lemhi
and at Lower Granite dam.
84
CSMEP – Hatchery Action Effectiveness Subgroup
Throughout the FY 2005 contract period, the Hatchery Subgroup identified a number of questions important to the evaluation of
hatchery management, and has reviewed numerous existing and proposed hatchery research, monitoring, and evaluation (RME) plans
within the Columbia River Basin. Following this review, the Hatchery Subgroup has concluded that existing and proposed hatchery
RME plans (if fully implemented) are likely to address the majority of the management questions identified by the Subgroup.
However, the Hatchery Subgroup has also concluded that a number of questions regarding the effectiveness of hatcheries as a class of
actions are unlikely to be adequately addressed by existing and proposed hatchery RME. These hatchery effectiveness questions
(identified in Appendix A) will likely be efficiently and comprehensively addressed only through the implementation of a stratified
and representative study design that spans the entire Columbia River Basin. As such, the study designs to address these questions are
best developed within a collaborative process that can rely on the expertise of the multiple tribal, state, and federal agencies with
operational jurisdiction and familiarity with the facilities. This expertise exists within CSMEP and has been useful in understanding
the high level of diversity represented by individual programs. With appropriate stratification, this diversity can be leveraged to
identify the mechanistic linkages of individual programs to broader monitoring questions that evaluate the overall effectiveness of
hatchery strategies at the regional scale. These broader-scale hatchery program effectiveness questions (as opposed to individual
hatchery operation questions) will become the focus of CSMEP designs intended to address larger scale multi-hatchery questions that
can be stratified across the region.
The Hatchery Subgroup has focused much of their initial efforts on developing alternative monitoring designs that could help answer
two of these critical questions relating to hatchery effectiveness:
1. What is the magnitude and distribution of hatchery strays into natural populations? (For harvest augmentation hatcheries.)
2. What is the relative reproductive success of naturally spawning hatchery fish and natural origin fish? (For both F1 and F2
generations from supplementation hatcheries.)
Insights into approaches gained from the CSMEP analyses required to address these two questions will provide a foundation for
tackling additional hatchery questions in a prioritized manner in FY 2006. Appendix A provides the Hatchery Subgroup’s current
summary of the design elements (DQO steps 1-7) for hatchery monitoring that are required to answer these and other questions.
85
Appendix A. CSMEP Hatchery DQO Summary
DQO STEPS
Policy
Inputs1 (9)
Problem:
Artificial propagation is used extensively as a management tool for Pacific salmon in the Snake River
Basin. Hatchery programs are operated to contribute to three general management goals:
1. Harvest Augmentation: to provide fish for tribal, commercial, and recreational opportunity while
keeping impacts to natural populations within acceptable limits.
2. Supplementation: the use of hatchery fish to enhance the viability of natural populations while
keeping impacts to non-target populations within acceptable limits.
3. Genetic Conservation: maintain genetic resources of imperiled populations to allow for
reintroduction of supplementation in the future.
Considerable uncertainty remains, however, regarding benefits and risk of supplementation and
conservation hatchery programs, as well as, risks to natural populations for harvest augmentation
programs.
Stakeholders:
States—Washington, Oregon, Idaho
Tribes—NPT, SBT, CTUIR, CTWIR, YIN
Federal—NOAA, USFWS, BPA, USACOE
Other—NPPC, CBFWA, conservation groups, Tribal, commercial, sport fishers
Non-technical
Issues:
Appropriate people with enough time to assess ongoing M & E programs, develop integrated study plans,
and assess if uncertainties are being adequately addressed.
Lack of adequate funding for data collection for supplementation program M & E projects
Principal
Questions:
There is a large suite of monitoring questions required in evaluation of hatchery programs (the group
identified 11 questions for harvest augmentation, 25 for supplementation and 5 for conservation
hatcheries) Examples of some principal questions include:
Harvest Augmentation:
9
• To what degree does the hatchery program meet harvest objectives?
• What is the magnitude and distribution of hatchery strays into natural populations
Supplementation
• What is the ratio of R/S for hatchery produced and naturally produced fish
• What is the relative reproductive success of natural spawning hatchery and natural fish?
Conservation
• Questions within this category are being deferred until later date
For the purpose of initial CSMEP design work we focused on only a smaller subset of key Harvest and
Supplementation hatchery questions
Alternative Actions: Make adaptive management changes to hatchery programs to improve performance and reduce impacts
on natural populations.
Reduce the magnitude of reliance on hatcheries, including elimination of some hatchery programs.
Modify existing hatchery facilities to improve effectiveness.
Decision
Statements:
Is the harvest augmentation hatchery program achieving harvest contribution objectives and keeping
impacts to natural populations at acceptable levels so that program modifications, reductions, or
elimination are not required?
Is the supplementation hatchery program enhancing the viability of the target natural population and
keeping the impacts to non-target populations at acceptable levels so that program modifications,
reductions, or elimination are not required?
Is the genetic conservation hatchery maintaining the genetic resources of the imperiled population so that
program modifications, reductions, or elimination are not required?
9
9
86
DQO STEPS
Action Levels
(critical effect
sizes)
The level of change that would trigger an adaptive management action varies considerably between each
hatchery objective and the associated metrics. Unfortunately, the vast majority of hatchery programs do
not identify quantitatively “acceptable limits”. It would be difficult to describe generic action levels for each
metric that would trigger an adaptive change in a specific hatchery program or groups of programs.
Instead, decisions must be framed around two general factors, the perceived value of the particular
metric monitored as well as the variance or degree of confidence in this value.
Information
Required:
The range of information required to address the suite of hatchery questions is extensive. We identified a
need for information from 65 performance measures across hatchery, supplementation and conservation
hatcheries. Examples of some of the information needs include:
Harvest Augmentation Hatcheries:
Policy
Inputs1 (9)
9
Commercial, recreational, and tribal harvest contributions by fishery
Smolt-to-adult survival
Progeny-to-parent ratios
Stray rates
Proportion of natural spawners that are hatchery strays
Supplementation Hatcheries:
•
•
•
•
•
•
•
•
•
•
•
Recruits per spawner for hatchery/natural fish
Hatchery/Natural fish abundance
Hatchery/Natural fish harvest rates
Hatchery/Natural fish spawning distribution
Smolt to adult survival for natural/hatchery fish
Allelic richness
Sources of Data:
Data resides in a variety of sources. Each hatchery program has some data for some metrics. No
programs have data for all metrics. The temporal and spatial scales, as well as, the number of metrics for
which there are data varies considerably among programs.
Quality of Existing
Data:
Harvest augmentation hatcheries: catch contribution, catch distribution, and smolt-to-adult survival data
is available. In some cases, however, specific harvest objectives are not well defined. The data available
for stray rates, and particularly for stray impacts to natural populations, is limited.
Supplementation hatcheries: little data available because many programs are early in implementation
and the data sets are for a limited number of years. In addition, most supplementation hatchery programs
are not collecting data for all the important metrics, and there are no pre-treatment data in some cases.
There is a lack of data from control or reference streams that can be used for comparisons.
New Data
Required:
There is a need to conduct an audit of each hatchery program to determine what objectives and metrics
are being assessed and at what level of adequacy.
There is a need for additional years of data from currently well designed hatchery M & E programs, and
for greater sampling for programs which have inadequate monitoring designs (e.g., - more extensive
juvenile tagging; more extensive spawning survey efforts; development of comprehensive genetics
monitoring)
This needs to be a large multi-basin and multi-generation effort.
Analytical
Methods:
Given the wide range of hatchery objectives and associated metrics, a suite of analytical approaches are
required, including:
•
•
•
•
•
•
•
Pre-post/control comparisons (BA, BACI)
Relative performance comparisons between hatchery/natural fish
Standard parametric and non-parametric tests
Genetic analyses
Time series analyses
Stock recruitment analyses
EMAP designs
87
DQO STEPS
Target
Populations:
Snake River Spring/Summer Chinook ESU - 33 natural populations
Snake River Steelhead ESU - 25 natural populations
Snake River Fall Chinook ESU - 1 natural population
Snake River Sockeye ESU - 1 natural population
Policy
Inputs1 (9)
Spatial Boundaries The focus areas are in the Snake River Basin. However, ocean and in-river harvest downstream from the
(study)
Snake are included in the boundary, as are the locations of populations where Snake River fish stray.
Temporal
There will be variable temporal boundaries dependent on the specific question and associated metrics.
Boundaries (study) Temporal durations include life-stage specific, annual, generational, and multi-generational.
Practical
Constraints:
There are financial, logistical, and technical constraints. In addition, there is lack of pre-treatment or
control system data available around which to frame analyses.
Spatial Boundaries The first level of assessment is hatchery program-specific, the second level is the target natural
(decisions):
population, the third level is at the major population grouping, the fourth level is at the ESU, and the fifth
level will be outside the ESU.
Temporal
Boundaries
(decisions):
Adaptive decisions can be made annually for each program; however, the temporal scale is linked to the
timeframe for availability of adequate data. Adequate data, in some cases, is many salmon generations
out in the future.
5. Decision Rules
Critical
Components and
Population
Parameters;
Critical Action
Levels:
If-Then Decision
Statements;
While the group was successful at defining hatchery monitoring questions and associated information
needs, they were less successful at defining appropriate “decision rules.” Very few hatchery programs
provide quantitative guidance for the use of data in an adaptive management framework. For example,
many hatchery programs communicate their goals as “supplement natural populations while keeping
impacts to non-target populations within acceptable limits.” The vast majority of programs do not identify
quantitatively what “acceptable limits” are or define what their response would be if these acceptable
limits were exceeded. This is a complex issue, primarily for three reasons: 1) acceptable limits are not
purely scientific, since hatcheries have legal mandates and a complex societal basis; 2) the long and
short-term biological affects of impacts at various levels are largely unknown and are unlikely to manifest
uniformly across all hatchery programs; and 3) decisions regarding hatchery management are typically
made based on the interaction of multiple questions rather than the result of a single question, thus
creating a difficult decision matrix.
Consequences of
Decision Errors
May continue/expand hatchery actions with little beneficial effect or even detrimental effects on fish
populations;
May discontinue hatchery actions that do have beneficial effects on fish populations;
Unnecessary costs;
Negative impacts to fisheries harvest;
Continued loss of fisheries
9
1Policy
Inputs - indicates with a check steps where group needs greater policy level feedback, presentation will elaborate on what feedback is
required
88
Steps 6 and 7.
(examples)
Evaluation Design
Sampling Design
(Where and When data will be collected)
Response Design
Harvest Augmentation
Q: What is magnitude
and distribution of
hatchery strays into
natural populations?
L ($ 1,330,000)
38 new sites @ 35 k
- high tag rates (rotating); lower replication = longer
study duration
- sampled across CRB at sites selected by stratified
design (84 sites + 10 additional EMAP sites)
- sampled annually and also opportunistically
- multiple pass carcass surveys for CWT or PIT tagged
fish
M ($2,070,000)
69 new sites @ 30K
- high tag rates (rotating); mod replication = mod study
duration
fish
H ($2,425,000)
97 new sites @ 25k
- high tag rates (rotating); higher replication = shorter
study duration
fish
L) ($3,000,000/year-approx.)
($250,000 x 12 sites)
Inference via BACI design
Monitored at 3 strata x 2 replicates + reference = 12
sites,
Minimum of 2-3 generations
Abundance of Progeny/adults, change in productivity
(identical sampling infrastructure and tagging program
as for high option)
H ($1,800,000/year-approx.)
($300,000 x 6 sites)
Genetic parentage analysis
Monitored at 3 strata X 2 replicates = 6 sites,
Variable duration depending on the contrast
Genetic sampling of adults and progeny and
assignment of juveniles/adults to parents
Supplementation
Q: What is the relative
reproductive success of
natural spawning
hatchery and natural
origin fish?
89
Principle CSMEP Hatchery Questions to be addressed:
Throughout the FY 2005 contract period, the hatchery subgroup identified a number of
questions critical to the evaluation of hatchery management, and has reviewed numerous
existing and proposed hatchery Research, Monitoring, and Evaluation (RME) plans
within the Columbia River Basin. Following this review, the subgroup has concluded that
existing and proposed hatchery RME plans (if fully implemented) are likely to address
the majority of the management questions identified by the subgroup. However, the
subgroup has also concluded that a number of questions regarding the effectiveness of
hatcheries as a class of actions are unlikely to be adequately addressed by existing and
proposed hatchery RME. These effectiveness questions (listed below) will likely be
efficiently and comprehensively addressed only through the implementation of a
stratified and representative study design that spans the entire Columbia River Basin. As
such, the study designs to address these questions are best developed within a
collaborative process that can rely on the expertise of the multiple tribal, state, and
federal agencies with operational jurisdiction and familiarity with the facilities. This
expertise exists within CSMEP and has been useful in understanding the high level of
diversity represented by individual programs. With appropriate stratification, this
diversity can be leveraged to identify the mechanistic linkages of individual programs to
broader monitoring questions that evaluate the overall effectiveness of hatchery strategies
at the regional scale. These broader-scale hatchery program effectiveness questions (as
opposed to individual hatchery operation questions) will become the focus of CSMEP
designs intended to address larger scale multi-hatchery questions (listed below) that can
be stratified across the region.
Harvest Augmentation Hatcheries: To what extent can hatcheries be used to assist in
meeting harvest management goals while keeping impacts to natural populations within
acceptable limits?
Regional Question
Priority
1
What are annual harvest contributions and catch distribution of hatchery produced fish?
H
2
To what degree does the hatchery program meet harvest objectives?
H
3
What is the distribution of hatchery strays into natural populations?
H
4
What are the proportions of stray hatchery fish in non-target natural populations?
H
5
What is the relative reproductive success of hatchery origin adults relative to natural
origin adults?
H
6
What are the disease agents and pathogens in hatchery fish, to what degree are these
agents transmitted to natural fish, and what are the impacts of such transmissions?
H
7
What are the impacts of hatchery strays on non-target populations?
H
90
Supplementation Hatcheries: To what extent can hatcheries be used to enhance viability
of natural populations while keeping impacts to non-target populations within acceptable
limits?
Regional Question
Priority
1
What are the status and trends of habitat targeted by supplementation projects and what is/are the lifestage specific factors that limit productivity?
H
2
What is the reproductive success of naturally spawning hatchery fish relative to natural origin fish in
target populations?
H
3
What are the disease agents and pathogens in hatchery fish, to what degree are these agents
transmitted to natural fish, and what are the impacts of such transmissions?
H
4
What are the relative effective population sizes and genetic diversity of hatchery supplemented vs. unsupplemented populations?
H
5
What proportion of hatchery origin juveniles return as adults to target versus non-target populations?
H
6
Do hatchery origin juveniles from supplementation programs stray at a greater rate than their natural
origin conspecifics?
H
7
What are the proportions of natural spawning stray hatchery fish in non-target natural populations and
their impact on the viability of natural populations?
H
8
What is the reproductive success of naturally spawning hatchery fish relative to natural origin fish in nontarget populations?
H
9
What are the effects of hatchery supplementation on the productivity, abundance, and viability of nontarget natural and hatchery-influenced populations?
H
91
SUMMARY OF KING COUNTY LONGTERM WATER QUALITY MONITORING
PROGRAM
The long-term monitoring program in King County is designed to most effectively use
the public’s scarce dollars to monitor environmental health in the county. The county’s
long-term monitoring program is divided into several key components:
•
Land Use and Land Cover
•
Weather
•
Streams and Rivers
•
Large Lakes
•
Vashon-Maury Island
•
Puget Sound
Each of these is described below.
Land Use and Land Cover
Descriptions of land use and land cover are important components of any environmental
assessment in the county. How the land is used, and how that changes over time, are key
components of descriptions of habitat quantity and quality, and stormwater runoff
quantity and quality. The King County Department of Natural Resources and Parks
(KCDNRP) develops land use coverages for the county from LandSat satellite images
every two years. KCDNRP also works cooperatively with UW to assess UrbanSim
output for land use, and convert the land use to land cover via the land cover change
model.
Weather
Weather is monitored throughout King County by a large number of agencies,
jurisdictions, organizations, and individuals. Weather monitoring can range from simple,
manual precipitation gauges, to fully automated multi-parameter meteorological stations.
Having high-quality weather data is instrumental to the environmental assessments in
King County, since hydrological issues are tied directly to weather patterns. The longterm monitoring program strives to ensure that there are precipitation and temperature
gages throughout the county, and that full meteorological stations are placed in strategic
locations. Coordination with weather monitoring programs by NOAA, UW, WDOT, and
others is critical to avoid duplication of effort. County-collected weather data is available
at: http://dnr.metrokc.gov/wlr/waterres/hydrology/GaugeSelect.aspx.
92
Streams and Rivers
Stream and river monitoring includes several important factors, including flow, water
quality, sediment quality, and benthic macroinvertebrate monitoring. Stream and river
health is directly affected by land use changes in the county.
Routine and Storm Water Quality Monitoring
Water quality in rivers and streams in WRIAs 8 and 9 is monitored monthly at 54
locations for field parameters, conventionals, and microbiology. Metals and organic
chemicals are not monitored as part of this program.
Wet-weather water quality in rivers and streams is monitored at 17 of the 54 routine
monitoring sites. Wet-weather monitoring occurs up to six times per year and includes
all parameters measured during routine sampling, as well as low-level metals.
The stream monitoring program is designed to answer the following four questions:
Question #1: Has the water quality of the streams under storm and/or base flow changed
over time?
Question #2: Is there a spatial difference in the water quality of both storm and base flow
conditions?
Question #3: What is the human health risk from stream water quality?
Question #4. Do current (or future) King County facilities impact stream water quality?
The streams monitoring program is described at:
http://dnr.metrokc.gov/wlr/waterres/streamsdata/.
Continuous Gauging
Continuous stream or river flow measurements are made at 44 sites in King County.
Flow is recorded every 15 minutes. Additional project-specific flow measurements are
made at numerous additional sites in the county. Continuous streamflow and temperature
data are available at: http://dnr.metrokc.gov/wlr/waterres/hydrology/GaugeSelect.aspx.
Stream Sediment Chemistry
Stream sediment chemistry monitoring is conducted to assess the accumulation of
potentially toxic chemicals in stream sediments in WRIAs 8 and 9. Stream sediments are
collected annually (August) from sites near the mouths of 10 streams. In addition, stream
sediment chemistry characterized in 1 to 3 subbasins each year, with about 1 sample per
stream mile. All stream sediment samples are analyzed for conventional parameters,
metals, and organic chemicals.
This program is designed to answer the following five questions:
Question #1: How does sediment quality in streams compare to available sediment
guidelines or thresholds?
Question #2: Are there other chemicals present in stream sediments that do not have
guidelines?
Question #3: How does sediment quality change over time?
Question #4: Are there differences in sediment quality within a monitored stream basin?
Question #5: How is sediment quality different among monitored streams that have
similar sampling strata?
93
Stream sediment monitoring is further described at:
http://dnr.metrokc.gov/wlr/waterres/streamsdata/sediment.htm.
Stream Benthic Macroinvertebrate Monitoring
Stream benthic macroinvertebrate samples are collected annually from 10 randomlyselected sites in each of 15 subbasins in WRIAs 8 and 9.
This program is designed to answer the following four questions:
Question #1: Do different watershed sub-basins within the Greater Lake Washington
Watershed (e.g. Bear Creek, Issaquah Creek, Lower Cedar Tributaries Creek, etc.) differ
in terms of biological condition?
Question #2: Do different watershed sub-basins within the Greater Green-Duwamish (e.g.
Soos Creek, Newaukum Creek, Jenkins/Covington Creek, etc.) differ in terms of
biological condition?
Question #3: Is the biological condition improving (or declining) over time? Is the trend
significant?
Question #4: Do different land use patterns measured at the sub-basin level affect
biological conditions differently within the watershed?
The stream benthic macroinvertebrate monitoring program is described at:
http://dnr.metrokc.gov/wlr/waterres/Bugs/index.htm.
Large Lakes
Long-term monitoring of Lakes Sammamish, Washington and Union (including the ship
canal) is critical to ensuring that these important, regional waterbodies are maintained
into the future. Monitoring includes routine water quality monitoring, summertime
swimming beach monitoring, continuous temperature monitoring, water level monitoring,
sediment quality monitoring, sediment benthic macroinvertebrate monitoring, toxic
cyanobacteria monitoring, and phytoplankton/zooplankton monitoring.
Routine Water Quality Monitoring
Water quality in the large lakes is monitored monthly in winter, and twice monthly is
summer, at 25 locations. Water quality is monitored every 5 meters below the water
surface for conventional parameters, nutrients, bacteria, and chlorophyll.
This program is designed to answer the following four questions:
Question #1: Has the water quality of the large lakes changed over time?
Question #2: Is there a spatial difference in large lake water quality?
Question #3: What is the human health risk from large lake water quality?
Question #4. Do current (or future) King County facilities impact large lake water
quality?
The large lakes monitoring program is described at:
http://dnr.metrokc.gov/wlr/waterres/lakes/index.htm.
Freshwater Swimming Beach Monitoring
Bacteria samples are collected weekly during warm weather at designated swimming
beaches along Lakes Sammamish, Washington, and Union, and in Green Lake.
94
Approximately 20 beaches are sampled each year. This program is a coordinated
program between KCDNRP, SKCPH, and various local jurisdictions.
This program is designed to answer the following three questions:
Question #1: Is the swimming beach water quality safe for human recreation?
Question #2: Is there a spatial difference in swimming beach water quality?
Question #3: What is the source of any identified unsafe water quality?
The swimming beach monitoring program is described at:
http://dnr.metrokc.gov/wlr/waterres/swimbeach/default.aspx.
Continuous Temperature Monitoring
Temperature throughout the water column depth is continuously monitored at 10 sites
using a series of thermistor chains. This monitoring program is coordinated with
temperature monitoring conducted by the Seattle District US Army Corps of Engineers.
This program is designed to answer the following five questions:
Question #1: Has the temperature of the large lakes changed over time?
Question #2: Has the seasonal and diurnal lake stratification and destratification changed
over time?
Question #3: Is there a spatial difference in the temperature of the large lakes?
Question #4: Is there a spatial difference in the seasonal and diurnal lake stratification
and destratification?
Question #5: Is the lake temperature adequate for salmonid use?
The continuous temperature monitoring program is not yet described on the web.
Toxic Cyanobacteria Study
A short-term study of phytoplankton, zooplankton, and toxic cyanobacteria has been
implemented for the major lakes, as described in the Sampling and Analysis Plan - Phase
II For Toxic Cyanobacteria in Lake Washington, Lake Sammamish and Lake Union,
February 2005. Sample collection utilizes the combined efforts of the routine Major
Lakes Sampling Program and the Swimming Beach Monitoring Program (SAP in
review). The Major Lakes Sampling Program collects samples twice per month between
March and October. Swimming Beach Monitoring occurs weekly from mid-May through
mid-September. Utilizing both programs will provide for weekly sample collection
throughout most of the productive growing season and provide for better tracking of
microcystin production in the lakes. Additional focused sampling efforts will be made to
collect scums or accumulations of cyanobacteria if they are present within the visual
distance of routine lakes sampling sites or as part of the Small Lakes Monitoring
Program. A bloom will be defined as a visually observable accumulation of
phytoplankton in the water column or as a surface accumulation.
Samples are collected using the same site-specific collection method used for chlorophyll
a described in section 4.5. In addition, seasonal microcystin and quantitative
phytoplankton samples are collected at seven locators with discrete surface sample
collection and at all six locators with integrated composite samples. In situ profiling will
be conducted at every five meter depth (from 1 meter to just above the bottom) at one
selected locator.
95
Phytoplankton and Zooplankton Monitoring
Zooplankton sample collection in Lake Sammamish is designed to compliment the
zooplankton sampling program at the University of Washington. UW sampling is
focused on the south end of Lake Washington to provide data on food availability for
outmigrating salmonids and an estimate of seasonal grazing pressure on the
phytoplankton populations. Qualitative ring net tows are conducted at one location in
Lake Sammamish for live zooplankton sampling. Zooplankton analyses are conducted
by the University of Washington Zoology laboratory.
Large Lakes Sediment Chemistry Monitoring
King County monitors large lake chemistry every 10 years to assess possible
accumulation of toxic chemicals. The sediment sampling design will be detailed in the
annual workplan, and will generally follow existing sediment SAPs that address
SEDQUAL data requirements. Sediment samples will be collected for chemical testing
using standardized equipment and procedures. Protocols for sediment sampling in lakes
Sammamish, Washington and Union are defined in the Sampling and Analysis Plans and
standard operating procedures.
Major Lakes Benthic Sampling
The benthic community will be assessed using a variety of indices (e.g., abundance,
diversity and the abundance of stress-tolerant or intolerant species). Benthic sampling
will occur on an infrequent schedule defined during annual work planning.
Collection and analysis of benthic samples requires significant resources. In order for
benthic samples to be comparable across studies, all benthic samples should be collected
using standardized protocols. Standardization of bioassessment methods for determining
benthic species composition, taxa richness, diversity, evenness, trophic levels and major
taxonomic spatial and temporal patterns may be enhanced significantly by the
conventional use of a U.S. No. 30 sieve (APHA et al. 1998) during sample collection and
initial processing. All samples will be initially preserved with 10% buffered formalin.
Alcohol is not a satisfactory tissue fixative, especially for soft body parts. Fixation with
formalin stabilizes tissue proteins to retain characteristics of the soft body parts (APHA et
al. 1998). A sufficient volume of formalin will be added to sample containers such that
the entire volume of pre-screened sample shall be covered with formalin. Specifics of
benthic sampling design and collection for Lake Washington can be found in the Lake
Washington Baseline Sediment Sampling SAP.
Vashon-Maury Island Water Resources
An integrated water resources monitoring program is conducted on Vashon-Maury Island
in Puget Sound to ensure long-term sustainability of the quantity and quality of water for
people and the environment. This monitoring program includes groundwater level
monitoring, groundwater quality monitoring, weather monitoring, stream flow and
temperature monitoring, land-use assessments, and water resources modeling. This
program is described at: http://dnr.metrokc.gov/wlr/WQ/vashon-island/ and groundwater
data are available at: http://dnr.metrokc.gov/wlr/wq/groundwater-data.htm.
96
Puget Sound
King County conducts routine monitoring of Puget Sound water quality and sediment
quality. This monitoring is conducted in coordination with the Puget Sound Ambient
Monitoring Program (PSAMP).
Routine monthly monitoring is conducted at 12 offshore stations and 10 beach stations.
A water quality profile is obtained using a CTD (conductivity, temperature, depth)
instrument, and samples are analyzed at the lab for conventional parameters, nutrients
and bacteria.
97
EXCERPT FROM:
Draft Guidance for Developing Monitoring and Evaluation as a
Program Element of the Fish and Wildlife Program
March 2006
98
Table of Contents
I. Developing Monitoring and Evaluation in the Program
1
Why Monitoring is Important: Evaluating the Fish and Wildlife Program
Why Monitor at a Programmatic Scale?
Where the Program Stands Today
Developing a Regional Approach to Monitoring: Spatial Scales
Determining Biological Effectiveness: The Role of Adaptive Management
A Regional Scale Program Requires Regional Scale Monitoring
Collaborative Funding
Inventory
Data Management
1
1
2
3
4
6
6
7
8
8
II. A Road Map for Developing a Regional Monitoring Framework 11
Management Questions and the Need for Supporting Data
High Level Indicators: Monitoring for the Program
What Are High-Level Indicators?
Why Are High-Level Indicators Important?
11
12
13
14
How High Level Indicators Are Being Identified
15
Provincial Scale Objectives: When and Where to Monitor for the Program
Fish and Wildlife Program Projects: the Building Blocks of Regional Monitoring
15
15
III. Using the Project Selection Process to Implement Monitoring
19
What Does it All Mean? Paradigm Shift Ahead
Developing a Monitoring Component of the Fish and Wildlife Program
Making Long-Term Commitment to Monitoring
Conducting Large-Scale Field Experiments
Point of Departure: Getting Underway for Fiscal-Year 2007-2009
Current and Proposed Monitoring Components for the Program
19
19
20
20
20
21
IV. Program Evaluation Requires Broad Range of Monitoring
25
Monitoring and Action Effectiveness Research
Develop Common Protocols for Fish/Wildlife Population and
Environmental Status and Trend Monitoring
Population Status, Trends and Distribution
Develop Common Site Selection Procedures
Develop Models for Predicting Abundance
Habitat Monitoring
Action Effectiveness Research
Habitat Project Effectiveness
Intensively Monitored Watersheds
Estuary
Artificial Production Effectiveness
26
26
27
27
28
28
29
29
30
31
32
99
Hydro Related Research, Monitoring and Evaluation
Uncertainties Research
Project Implementation/Compliance Monitoring
33
33
V. References Cited
34
VI. Appendices
36
Appendix A. Regional Monitoring Framework
Appendix B. Definitions of Monitoring Terms
Appendix C. Categories of Monitoring Within the Regional Framework
36
49
50
100
I. Developing Monitoring and Evaluation in the Program
This document provides guidance on monitoring for two different audiences. The first
and second chapters provide the Northwest Power and Conservation Council (Council)
and other policy makers the rationale and context for this guidance. Chapter one explains
the importance of developing a regional approach to monitoring. Chapter two describes
the primary elements of a regional approach to monitoring, including high-level
indicators, provincial scale objectives, and project scale work, and how they are linked.
Chapters three and four are intended for practitioners of monitoring. Chapter three
provides guidance on how the Council will use the project selection process to help
develop programmatic scale monitoring in the region. Chapter four describes the
categories of monitoring under which work will be initiated to implement a programmatic
approach and includes specific guidance on tasks.
Why Monitoring is Important: Evaluating the Fish and Wildlife Program
The 2000 Fish and Wildlife Program (Program) establishes a basinwide vision for fish
and wildlife along with four overarching biological objectives:
•
A Columbia River ecosystem that sustains an abundant, productive, and diverse
community of fish and wildlife
•
Mitigation across the basin for the adverse effects to fish and wildlife caused by
the development and operation of the hydrosystem
•
Sufficient populations of fish and wildlife providing abundant opportunities for
tribal trust and treaty right harvest and for non-tribal harvest
•
Recovery of the fish and wildlife affected by the development and operation of
the hydrosystem that are listed under the Endangered Species Act
The principal vehicle for implementing these objectives are the restoration projects to
improve conditions for listed and non-listed anadromous fish, resident fish, and wildlife
that have been impacted by the hydrosystem in the Columbia River Basin. The central
question for the Council is whether or not the projects are in fact helping the Program
reach these objectives. This guidance will facilitate the development of a monitoring
component for the Program that over time will provide the basis for a quantitative
assessment of progress toward the Program’s overarching objectives. Specific provincial
scale objectives that will be developed in 2006
Why Monitor at a Programmatic Scale?
In the Pacific Northwest, natural resource management entities collect and analyze many
types of information for answering specific management questions to address the
objectives for which they are responsible. To effectively combine information to answer
101
management questions will require monitoring across multiple geographic and temporal
scales, including:
•
Tributaries with major projects, populations
•
Major population groups
•
Subbasins
•
Evolutionarily Significant Units
•
Major Population GROUPS
•
the Columbia River Basin
Developing standardized approaches, making and securing long-term funding
commitments, and coordinating with our regional partners are all essential to the success
of this initiative.
The Pacific Northwest Aquatic Monitoring Partnership or PNAMP, is working to develop
standardized protocols and methods for field data collection, data management, and
analytical processes, which, in widespread use, would change this data into a common
currency. This would enable data collected for an initial primary purpose to maintain
value for use by subsequent secondary users wishing to analyze aggregate data, or
“rolling-up” the data, following a set of universal guidelines. Similarly, there is value in
being able to combine information even within the same watershed or nearby watershed
to increase the inferences or statistical power of information.
While there are many potential analytical applications for aggregate data, this guidance
identifies the need to coordinate the collection of data in a manner that can support
evaluation and decision-making at higher-level spatial scales, for example subbasin plans,
Evolutionary Significant Units, and provincial scale objectives. By supporting this work
it will be possible to conduct basic assessment and evaluation work at the population
level, and at a regional scale.
This increases the potential for technical, policy and public organizations to communicate
using consistent language and processes and to provide accurate and unambiguous
information to the public, NGO’s, governments and their branches. Enabling operational
adaptive management and well-informed decision-making will be the principal results of
this initiative.
Where the Program Stands Today
Until now, monitoring in the Fish and Wildlife Program has primarily been conducted to
evaluate work at the project scale, across all subject areas. This approach has generated
monitoring information useful to individual restoration projects. However, monitoring
102
has not been developed into an element of the program that can provide a basis for
evaluating the program. Consequently, the Program must now apply limited resources to
developing a more programmatic approach, which can detect the cumulative effect of
restoration actions. By developing the ability to conduct such evaluations, the program
will be able to identify future actions that are more strategic. Identification of high-level
indicators and the development of provincial scale objectives will be required to support
this programmatic approach.
While monitoring of work at the project scale has intrinsic value, and will continue on a
reduced basis, it cannot substitute for the lack of a monitoring program of sufficient
scope to provide a basis upon which the program as a whole can be evaluated, and redirected. Monitoring must be conducted at different scales to:
•
Assess the performance of the program relative to biological and programmatic
objectives
•
Identify where and why there are performance problems
•
Identify the most effective actions needed to correct problems so that program
objectives can be achieved
Developing a Regional Approach to Monitoring: Spatial Scales
The absence of a regionally coordinated approach to monitoring and evaluation in the
Columbia River Basin has constrained restoration and planning efforts for decades. For
this reason, it is important that a more hierarchical approach be utilized with increased
emphasis on achieving useful outcomes from monitoring. Specifically, methods need to
be developed and implemented so that monitoring results can be combined at the same
scale, or rolled up to higher scales to provide scientifically defensible evaluations . For
example, to determine whether the status of fish and wildlife populations and/or
ecological condition of a subbasin, an ESU, or the Columbia River Basin as a whole is
improving or declining over time. This capability would be very useful to policy and
decision makers as they deliberate on future actions under the program that affect the
long-term, ecological health of the basin.
Shifting the focus of monitoring from project to larger spatial scales has both benefits and
challenges. One benefit of focusing on the population scale is that it’s a scale with direct
relevance to fish managers, who want to know if actions within a watershed can actually
improved a fish population’s production, for example smolts/spawner, in addition to
improving habitat conditions in the restored reaches. The population scale is also of great
interest to agencies like NOAA Fisheries charged with evaluating the status of listed
populations.
There are also some significant challenges in shifting monitoring to larger spatial scales.
Reliably attributing observed changes in fish survival or production to particular sets of
management actions requires careful monitoring design. Otherwise, one might
103
erroneously infer that observed changes were due to management actions when in fact
they were the result of natural variation in freshwater climate or ocean conditions.
Ideally, one would monitor both ‘treated’ areas (those with habitat restoration actions)
and nearby ‘reference’ areas (those without restoration actions), for several generations of
fish populations, both before and after implementation of actions, and measure other
explanatory variables simultaneously. One challenge is that it becomes increasingly
difficult at larger scales to establish the strong contrasts required to evaluate
effectiveness; that is areas and times with and without certain classes of restoration
actions. For example, adjacent subbasins will each have a variety of implemented
restoration actions, so that comparing fish production across these subbasins and over
time will not lead to any clear inferences on which actions (if any) were responsible for
any observed differences in trends over time. It will therefore still be necessary to
conduct effectiveness evaluations at finer spatial scales for a carefully selected subset of
restoration actions and locations.
Determining Biological Effectiveness: The Role of Adaptive Management
Adaptive management provides a valuable tool for ensuring that timely feedback from
program activities increases effectiveness by re-directing future work. In their seminal
work applying adaptive management in a hydropower context, Professor Kai Lee and
Jody Lawrence wrote:
Adaptive management encourages deliberate design of measures. This assures
that both success and failures are detected early and interpreted properly as
guidance for future action. Information from these evaluations should enable
planners to estimate the effectiveness of protection and enhancement measures on
a systemwide basis. Measures should be formulated as hypotheses. Measures
should make an observable difference. Monitoring must be designed at the outset.
Biological confirmation is the fundamental measure of effectiveness. (Emphasis
added.)
(From Adaptive Management: Learning from the Columbia River Basin Fish and Wildlife
Program, Environmental Law Vol.16:431-460, 1986.)
The National Research Council (NRC) reported several lessons learned about the
practicability of adaptive management and the institutional conditions that affect how
experiments on the scale of an ecosystem can be conducted (NRC, 1996), specifically:
•
Learning takes from decades to as long as a century. Patience is both necessary
and difficult, particularly in institutional settings such as government that work in
faster cycles
•
Systematic record keeping and monitoring are essential if learning is to be
possible
104
•
Cooperative management in the design and execution of experiments is
indispensable
•
Experimentation within the context of resource use depends on the collaboration
of resource users
•
Adaptive management does not eliminate political conflict but can affect its
character in important, if indirect, ways
Monitoring and evaluation is at the heart of the adaptive management because it provides
the information, data and analysis for identification of need and subsequent tracking the
progress of plans and population, or their lack of progress, for decision-makers and
resource managers. The success of the program depends on the consistent application of
well-designed research, monitoring and evaluation that can enumerate the information
required for different types of decisions at multiple scales, for example management of
harvests, the hydrosystem, and hatcheries; and, decisions on the protection and
restoration of habitat.
Another key element to an adaptive management experiment is providing a large enough
perturbation to a system so a detectable change in a response variable can be measured.
For example, by measuring responses to a limited range of spill and flow levels in the
Columbia River hydrosystem, it will be difficult to assess detectable changes over the
salmon and steelhead life-cycle and to contrast those changes in life-cycle survivals to
those for transported juvenile fish.
To be successful, adaptive management requires that "triggers" be established for
initiating adjustments or changes based on the results of monitoring. Monitoring without
triggers and adjustments does not constitute, and cannot support, adaptive management.
Triggers should be identified, required, and be more than checkpoints in time. For
example they could be related to performance standards, or achievement of deliberate
experimental design outcomes, or management targets. Thus, failure to attain
performance standards as expected, on the stated timeframe, should trigger the
appropriate review of what happened, why, and the determination of next steps. The
following steps to implementing adaptive management are portrayed in Figure 1:
•
Assessing limiting factors and critical uncertainties
•
Designing projects, programs and monitoring to maximize both on-the-ground
effectiveness and learning
•
Coordinated and documented implementation of projects
•
Consistent monitoring through standardized methods, protocols, and training
•
Timely and thorough evaluation of effectiveness
105
•
Overall guidance to the region to adjust plans and programs at the Province and
subbasin level
Figure 1. Sequence of steps in adaptive management.
Assess
Adjust
Design
Evaluate
Implement
Monitor
A Regional Scale Program Requires Regional Scale Monitoring
For over a decade the Program’s science review groups have been calling for the
“development and implementation of a system-wide monitoring and evaluation
program,” (SRG 93-2). The objectives and management questions of the Program
overlap those of many other regional entities and, local state, federal and tribal
governments. The costs of the monitoring and research needed to adequately address
these common management questions are more than one program can adequately support
or fund alone. Only through the combined efforts of multiple entities can a sufficient
level of information be developed to answer resource management questions through
coordinated, standardized and programmatic approaches to monitoring. There are a
number of existing efforts in the region to coordinate monitoring and evaluation but until
recently there has been a lack of an organizing principle or central forum to facilitate
these efforts.
The 2000 Fish and Wildlife Program, Basinwide Provision D.9, states that:
“The Council will initiate a process involving all interested parties in the region to
establish guidelines appropriate for the collection and reporting of data in the
Columbia River Basin.”
Another directive for developing a regional approach to monitoring was included in the
“Recommendations of the Governors of Idaho, Montana, Oregon and Washington for
Protecting and Restoring Columbia River Fish and Wildlife and Preserving the Benefits
of the Columbia River Power System,” issued in June of 2003. In response, Council staff
has joined and helped inaugurate the Pacific Northwest Aquatic Monitoring Partnership,
or PNAMP, chartered to provide such a forum (see www.pnamp.org). Through their
participation in PNAMP, the Council, Bonneville, and the fish and wildlife managers are
106
working to implement the Program within the context of a regional network of
monitoring efforts so that the shared monitoring needs and objectives of the program can
be achieved. Major accomplishments of PNAMP relevant to the development of a
regional approach to monitoring include the following several key operational
documents:
•
•
•
•
Draft Plan 2004, titled, “Recommendations for Coordinating State, Federal, and
Tribal Watershed and Salmon Monitoring Programs in the Pacific Northwest,” on
January 6, 2004.
Considerations for Monitoring in Subbasin Plans 2004
Strategy 2005
Charter 2005
In addition to providing staff support for this regional initiative, the Council has also
funded the Collaborative Systemwide Monitoring and Evaluation Project (CSMEP),
designed to facilitate implementation of monitoring within the Columbia Basin. CSMEP
is a three-year project funded under the Program that is working on several of the tasks
identified as priorities by the Fish Monitoring Workgroup of PNAMP, NOAA, USFWS,
and the Action Agencies. In close coordination with PNAMP, the CSMEP has been
working since October 2003 to develop rigorous approaches to monitoring and evaluation
that directly serve the needs of specific decisions, and build on the strengths of existing
monitoring infrastructure. PNAMP and CSMEP have been, and will continue to, work
closely together.
PNAMP is playing a key role in the development of coordinated approach to monitoring
at a regional scale. It provides a central forum for the discussion of policy and
management issues and sponsors workgroups comprised of monitoring practitioners
working to resolve technical issues. PNAMP is the key forum for implementing the
regional framework for monitoring described in Chapter II.
Collaborative Funding
In 2000, the Council shifted from an annual project funding cycle to a three-year cycle.
Because state and federal agencies remain on an annual funding cycle, it is difficult for
them to make long-term funding agreements. Consequently, formal arrangements such
as memoranda of agreement (MOAs) may be necessary to secure long-term funding
commitments for selected large-scale field experiments, for example the MOA between
Bonneville and the U.S. Forest Service. In regard to the Program, it is important to
acknowledge the difficulty inherent in reprogramming existing funds to support
additional research initiatives within the available direct-program budget.
Yet the important question is not how much investment in additional monitoring the
program might afford, but rather how to implement a comprehensive regional research
agenda that can be funded from multiple sources, sustained, and managed to mutually
endorsed outcomes. A more systematic and strategic approach to leveraging investment
by many parties is warranted. This guidance identifies critical uncertainties that need to
107
be addressed by multi-agency initiatives, cooperative funding agreements, and the
sharing of responsibility for implementation.
Some identified monitoring needs are currently, or should be more appropriately, the
requirement or shared responsibility of federal or state agencies other than Bonneville,
under mandates other than the Northwest Power Act. This point is particularly relevant
to ESA recovery planning and implementation research needs that are proposed for the
Columbia River Basin but have application coast-wide. Discrete elements of this
guidance present differing degrees of opportunities for regional coordination and shared
funding. To succeed, it is incumbent upon members of PNAMP to develop and
implement incentive strategies. Incentives may include funding, regulatory flexibility, or
recognition, all of which can work in combination. Thus, there is a need to work
cooperatively with entities that represent alternative funding sources and have
responsibilities that overlap those of the Council, for example the Trust for Public Lands
and others. The regional entities should recognize that all programs are limited by what
they can afford to sustain, but that by working together, all the programs could benefit
from focused, coordinated expenditures.
Inventory
It would be valuable for long-term planning of monitoring and evaluation to assemble a
comprehensive and detailed inventory of all monitoring activity in the Pacific Northwest
region. The inventory should be structured to provide a web-based, searchable database.
The benefit here is clear: knowing who is doing what, where, why and how, will enable
cost savings through elimination of duplicate work, and enable a higher level of
collaboration and communication. This inventory will also identify critical gaps and
areas where strong and weak data sets currently exist.
CSMEP has conducted detailed inventories, and assessments of the strengths and
weaknesses, of data within a dozen subbasins of the Columbia Basin. These very detailed
data inventories, which are available on an internet-accessible website, need to be
complemented by a broader, less detailed inventory to be developed by PNAMP.
Additional inventory work, on the scale of the entire Pacific region, is being conducted
by the State of Salmon project, a joint effort of Ecotrust and the Wild Salmon Center.
108
Data Management
A regional approach to monitoring will fail without the support of a data management
system that can provide regional access to the data sets developed through monitoring
efforts, on a timely basis, for analytical manipulation. To be successful a data
networking system must be able to assist scientists in the identification and development
of data standards as it relates to the monitoring of fish and wildlife populations and their
related habitats. This objective helps to identify solutions that improve access, sharing,
and coordination among different collectors and users of monitoring data for fish and
wildlife populations and their related habitats. It also provides a data reporting
foundation that will lead towards coordinated agency reporting, uniform monitoring
protocols, and improved data quality and quantity. Objectives include:
•
Develop a consistent data standards and protocols within and across each of the types
of monitoring
•
Establish a close working relationship for data consistency across the data sources
•
Identify and document the specific data needs of the region for watershed condition
monitoring, fish population monitoring, and effectiveness monitoring
•
Develop and recommend data collection standards and information to be shared
across the various monitoring programs
•
Share requirements and results with regional data networking entities to ensure
sharing of monitoring data
•
Test the collection protocols, sampling methods and data sharing mechanisms
•
Implement coordinated solutions within regional programs
•
Embed common analysis capabilities and reporting capacity
•
Provide public access sections or linked web sites for informational and collaborative
processes
There are many different interests and initiatives concerned with improving data
collection or management in the Columbia Basin and the Pacific Northwest. These
efforts involve many different constituencies, mandates, and obligations. At present,
there is no common regional data management network that links these interests and
initiatives. To address this situation, the Council has initiated a process for identifying
data needs in the basin, surveying available data, and filling any data gaps. The Council,
NOAA Fisheries, and other regional entities supporting this effort consider it imperative
to develop a regional data network that will:
•
Utilize existing databases, facilitate data management and sharing
109
•
Help subbasin planners
•
Underpin salmonid recovery efforts under the FCRPS Biological Opinion
•
Support the monitoring component of the Council’s Program
The Northwest Environmental Data Network or NED is leading this initiative. PNAMP
scientists, statisticians and biologists plan to develop data collection standards that can be
used across many different programs. NED could then support the consistent use of these
standards and in addition develop data sharing standards as a part of an overall regional
data network. NED will also work with other regional groups to support consistent
standards for their data collection programs. In this way NED will support the sharing
and integration of many different regional data sets including the PNAMP data.
There are important differences in the expertise needed for PNAMP and NED, while
PNAMP is primarily made up of scientists, statisticians and biologists, NED will be
supported by information network specialists whose role is to work with regional
scientists and others to make the information that they collect more consistent and more
readily available. This “corporate” approach to organization realizes the efficiencies and
benefits that can be gained from introducing standards to information systems while
supporting the flexibility that is needed by the many different data collection subdisciplines. It is also consistent with the recommendations that the region has received
on data management and science by the Council’s Independent Science Advisory Board
and the Independent Science Review Panel.
The methods and protocols used in the collection of data for use within the Fish and
Wildlife Program must be consistent with guidelines approved by the Council and
adopted by the region. It is important to note that while the ISRP checks these criteria, it
is Bonneville who must enforce the guidelines. Guidelines appropriate for the collection
and reporting of data at the project scale include:
•
The project must have measurable, quantitative biological objectives
•
The project must either collect or identify data that are appropriate for measuring
the biological outcomes identified in the objectives
•
Projects that collect data appropriate for secondary use, for example subsequent
higher scale evaluations, must make this data and accompanying metadata
available to the region in electronic form
•
Data and reports developed with Bonneville funds should be considered to be in
the public domain
•
Data and metadata must be submitted within six months of their collection
110
WASHINGTON GOVERNOR’S FORUM ON
MONITORING
The Governor's Forum on Monitoring Salmon Recovery and Watershed Health (FORUM) was created by
Executive Order 04-03 in August 2004, to coordinate monitoring consistent with the Comprehensive
Monitoring Strategy and Action Plan for Watershed Health and Salmon Recovery (CMS).
The FORUM is comprised of state and federal agencies, named by the Governor in the Executive Order,
involved in watershed health and salmon recovery. It also includes representation from the Northwest
Power and Conservation Council and the tribes.
According to the Executive Order, the FORUM is chartered to:
Provide a multi-agency venue for coordinating technical and policy issues and actions related to
monitoring Washington’s salmon recovery and watershed health.
We envision a coordinated network of state, federal, and tribal agencies able to share data and coordinate
spending effectively for developing and reporting the status and trends of Washington’s salmon and
watershed health and restoration efforts.
The tasks of the FORUM outlined in the Executive Order are as follows:
1.
2.
3.
4.
5.
6.
7.
Make recommendations on biennial reporting of monitoring results and progress in watershed health
and salmon recovery.
Foster integrated analysis and reporting of monitoring information.
Provide monitoring recommendations to the Salmon Recovery Funding Board (SRFB), the Governor's
Salmon Recovery Office (GSRO) and appropriate state agencies.
Develop a broad set of measures that will convey results and progress on salmon recovery and
watershed health in ways that are easily understood by the public, legislators and Congress.
The Forum is also encouraged to develop such indicators with federal, tribal, regional and local
partners working on salmon recovery and watershed health so that there is standardization of the
measures used.
Coordinate with local and regional watershed and salmon recovery groups, tribes, other states, the
Northwest Power and Conservation Council, U.S. Environmental Protection Agency, NOAA Fisheries,
U.S. Fish and Wildlife Service, and U.S. Forest Service.
By January 2006, and biennially thereafter, provide a report of its key activities and recommendations
to the Governor, the Legislature, and the SRFB.
In order to accomplish the various tasks, the FORUM formed six technical subcommittees to assist in a
multi-agency review of the major monitoring activities. The FORUM meets quarterly to address
monitoring issues. Figure 1 is a schematic representation of the FORUM structure and subcommittees
created.
111
Governor’s Forum On
Monitoring
State Partners
Federal, Tribal, Local
Biennial Forum Report to
Governor, Legislature, SRFB
Fish Abundance
Marine/Nearshore
Subcommittee
Monitoring Steering Committee
Barrier
Subcommittee
Effectiveness
Subcommittee
State of Salmon In
Watersheds Report
RCW 77.85.020
Habitat/Water
Quality
Status/Trends
Data Sharing
SWIMTAC
Figure 1. FORUM Structure
Statewide Measures of Salmon Recovery and Watershed
Health Status-Trend Monitoring
Statewide measures in order to have significance must be based upon long term status and trend
monitoring. Since April 2005, the FORUM has convened six subcommittees to assist in making
improvements to the statewide high level measures of salmon recovery and watershed health.
The following status and trend programs are considered most important for informing the public,
the Legislature, and the Governor:
Adult salmon abundance.
The WDFW and the tribes are working on showing harvest and adult spawners on the same charts so that
total marine production can be displayed compared to recovery targets. They are developing ways to
improve confidence limits and to display the results of monitoring populations determined as most
appropriate by the recovery plans developed for each Salmon Recovery Region (SRR).
Juvenile migrant salmon abundance.
The WDFW and the tribes are working on improving the juvenile migrant index to include tribal and other
sites not administered by WDFW. They wish to track and display productivity measurements as well. They
are determining how and where to better sample juvenile migrant abundance to reflect major population
groups and principle populations.
Habitat and water quality status/trends.
The Department of Ecology is creating a statistically valid statewide sampling framework of randomly
distributed sampling locations per US Environmental Protection Agency EMAP protocol. It will include
wadeable and non-wadeable streams and rivers and the sampling design will be developed to answer status
and trends at the statewide, SRR, and Watershed scale if fully implemented. The Department of Ecology is
to complete the framework and bring the results back to the Board at its May 2006 meeting for decisions
regarding funding in the 2007-09 Biennial Budget.
112
Water quantity index.
The FORUM and the Department of Ecology are working on a better way of displaying water quantity
problems for use in future reports. More flow gauging stations would help create a better picture of flow
problems in the state.
Barriers
WDFW is working on incorporating federal barrier information into the existing SSHIAP database and will
display barrier information and targets by SRR for the next SOS report.
Project Implementation. Estuary/Nearshore. The FROUM would like one or more indicators for
nearshore/marine habitat as part of the State of Salmon Report. A subcommittee was charged to bring
recommendations to the FORUM on a specific indicator or set of indicators to be included in the State of
Salmon Report that would characterize marine conditions in Puget Sound, the coast, and the lower
Columbia River estuary.
Action Effectiveness Monitoring
The state continues to support monitoring funded through the SRFB and the Forest and Fish Agreement as
crucial pieces that demonstrate that management actions taken by the state to restore watershed health and
recover salmon and steelhead populations are effective. The SRFB has funded a statistically robust
examination of 8 categories of restoration projects and also protection projects to determine if they have
been effective at the project scale in meeting design, habitat, and fish objectives. These projects are
sprinkled across Washington and Tetra Tech EC has been contracted to complete the field sampling and
evaluation. The sampling program uses a before and after control impact design (BACI).
In addition the SRFB has funded Intensively Monitored Watersheds (IMWs) to validate that restoration
actions taken on a cumulative basis do produce more juvenile and adult fish as a result. Clusters of
watersheds have been established in the Strait of Juan de Fuca, Hood canal, and Lower Columbia River to
test this assumption. Each uses a “control” watershed and two or three “treatment” watersheds. A fourth
IMW has been established in the Skagit River estuary to test the responses of juvenile chinook salmon
toward restoration of tidal estuaries.
The Department of Natural Resources receives funds from Congress to monitor and implement the Forest
and Fish agreement and associated HCP involving Washington Forest Practice rules. Monitoring has been
restricted to tracking the effectiveness of various forest practices required in the agreement. Future actions
will monitor forest conditions through status-trend monitoring of treated state and private forest lands.
The Department of Ecology continues to work to evaluate the effectiveness of TMDLs established under
the Clean Water Act.
Implementation Monitoring
The Implementation Subcommittee has been working to include information from conservation districts,
Bonneville Power Administration, and other sources in future State of Salmon Reports in order to better
reflect the diversity of effort being expended to recover salmon and to keep our watersheds healthy.
There is a need within the Washington natural resource agencies to have common requirements and
possibly a common database for tracking grants and projects associated with restoring watershed health and
salmon recovery so that there can be summarized reporting of restoration efforts. There have been previous
requests for funding of the next phase of the Natural Resources Data Portal that will enable composite
reporting of key information about salmon recovery and watershed health from multiple state agencies and
local government efforts as well.
113
ESTABLISHING A CALIFORNIA COASTAL MONITORING PROGRAM
Kevin Shaffer, California Department of Fish and Game
For questions or interest, contact (916) 327-0713; [email protected]
Monitoring Program Goals
The primary goal of the California Coastal Monitoring Plan (CalPlan) is to gather information relevant and
useful for successful management, conservation, and protection of coastal anadromous fishes. To do this,
the California Department of Fish and Game (CDFG) is working with the National Marine Fisheries
Service (NMFS) to create a program to evaluate the status and trends of:
• ESA and CESA listed anadromous salmonid species;
• Related environmental processes and conditions;
• Progress toward achieving recovery criteria;
• Threats or impacts of various threats to the resources.
The primary goal of this document is to identify needed elements to enable the Calplan to achieve these
objectives. Thus, the CalPlan will address scientific and fisheries management questions vital to conserve
the species. Early efforts are focusing on identifying the entities to conduct the monitoring, the audience
for the ensuing results and information, and the spatial and temporal commitment of the program.
Therefore, to first design, and then implement a monitoring plan requires the definition of its strategic
context.
Responsible Agencies
CDFG and NMFS will be the cooperating agencies that will be responsible for all initial aspects of the
monitoring plan, including 1) funding, 2) staged implementation, 3) field assessment and monitoring, 4)
data compilation, storage, and management, 5) analysis, and 6) reporting, and 7) the pursuit of additional,
crucial program resources. Because of the enormity of both initiating and maintaining coastal monitoring,
each agency likely will be solely responsible for some functions. For other responsibilities, it will be both
necessary and advantageous for the two agencies to work collaboratively.
Interested Audience
There are very high stakes in resource conservation, species protection, recreational interest, tribal
relations, and pressures on land and water development and other economic uses related to salmon wellbeing. To address these interests, both CDFG and NMFS, including its science laboratory and protected
resource divisions, know it is essential that a coastal monitoring plan be developed. Thus at the
management and leadership levels of both the State of California and federal government, there is interest
and concern over the successful development of coastal monitoring.
Additionally, increased interest in the health of anadromous salmonids has also resulted in increased
scrutiny. The institutions that allocate funding to the resource agencies want a better reporting on the
progress of habitat restoration and population recovery. Currently, neither fisheries agency is able to
provide the level of information needed to adequately address resource, management, and statutory
questions being asked concerning progress in recovery efforts. The Calplan will provide information to
both the State and federal legislatures, and in that way demonstrate progress of restoration and recovery
efforts and substantiate agency needs to continue pursuing successful conservation of coastal salmonids.
Although CDFG and NMFS are the primary organizations responsible for conservation of anadromous
salmonids, they are not the sole agencies interested in or already involved in monitoring and assessment
activities in California. Other federal agencies (e.g., National Parks Service and U.S. Forest Service), state
agencies (e.g., Department of Forestry and Fire Protection), Tribal Nations in northern California, water
resource agencies (e.g., Sonoma County Water Agency, State Water Quality Board), private industries
(e.g., forestry companies, fisheries organizations), and academic institutions (e.g., California State
114
universities, University of California campuses) have experience, expertise, and inherent interest in the
current and projected health of coastal anadromous salmonids.
Hence, there are several levels of interest and awareness of coastal salmonid and watershed monitoring.
The CalPlan needs to know and serve its several audiences, their priority, and their interest and potential
participation in coastal monitoring.
The CalPlan’s primary audience includes the leaders CDFG and NMFS and their fisheries programs, the
California Legislature, and Congress. The second primary audience of the monitoring plan is made up by
the managers and field specialists in both agencies. These people constitute the fisheries programs that will
have the main responsibility to implement a coastal assessment, monitoring, and adaptive recovery
program. Fisheries biologists in the field, scientists at the NOAA Science Lab, Geographic Information
System analysts, database specialists, and statisticians in CDFG and NMFS must be part of a coherent,
unified effort. Their understanding and concurrence with both the merit and feasibility of the coastal
monitoring plan are indispensable.
Additional audiences for the Calplan include the leaders and field staff of other involved agencies and
organizations and the citizens of California and across the nation, since it is through their support of natural
resource agencies that conservation of salmon and steelhead is made possible. This audience includes the
program managers and staff of assessment and monitoring programs in the other Pacific states and British
Columbia. Feedback and advice from and coordination with those across the Pacific Northwest who are
experienced and knowledgeable in salmonid monitoring will be invaluable for both implementation and
advancement of coastal monitoring in California.
Geographic Scope
The Coastal Monitoring Plan covers coastally draining watersheds from Mexico to Oregon. These
drainages range from small to moderate coastal drainage systems that are entirely or predominantly
influenced by coastal climate (e.g., moderate ambient temperatures regimes, fog belt, and onshore winds
with greater precipitation north of Monterey Bay) to large watersheds that extend well inland, where the
upper basins are characterized by higher elevation and interior climatic conditions (e.g., hotter summer and
colder winter regimes, snow-melt hydrology, drier summers). The Central Valley system is not included in
CalPlan.
Duration
The duration of the Coastal Monitoring Plan is founded on two factors. The first factor is the 100-year
time-frame for the evaluation of extinction risk conducted by federal scientists analyzing the viability of
salmonid populations (McElhany et al. 2000). The second factor is the variation and age composition of
spawning adult salmon and steelhead. Coho salmon demonstrate a relatively fixed three-year life cycle in
California (Weitkamp et al. 1995, Waples et al. 2001), and brood years do not overlap and a complete
generation is represented by three consecutive years of returning spawners. The maturation of Chinook
salmon inhabiting coastal watersheds from the Russian River to the Smith River is characterized by a
predominance of three- and four-year old fish. However, two-year olds are common, and cohorts may
include five- and six-year olds (Myers et al. 1998).
Species Covered
The intent of the CalPlan is to monitor anadromous species of coastal California. A phased approach will
utilized to achieve this intent. This document addresses the monitoring of Chinook salmon (Oncorhynchus
tshawyscha), coho salmon (O. kisutch), and steelhead (O. mykiss). It is anticipated that the next species to
be integrated into the CALPLAN will be coastal cutthroat trout (O. clarki). After this is accomplished,
green sturgeon (Acipenser medirostris), white sturgeon (A. transmontanus), and Pacific lamprey (Lampetra
tridentata) will be included in the CalPlan, in that order.
115
Population Status and Trend
For NMFS, the single-most important assessment is evaluating the viability of populations, more
commonly referred to as Viable Salmonid Population (VSP) analysis (McElhany et al. 2000). Several
factors inter-relate, providing an evaluation of what constitutes viable populations of anadromous
salmonids. The four, primary parameters include population size, growth rate, spatial structure, and
diversity (McElhany et al. 2000). To accomplish population viability analysis (PVA), factors such as
environmental variation, fish abundance, genetic variation and diversity, habitat structure and dynamics,
age- and sex-composition, marine, estuarine, and freshwater survival, fish size at various life-stages, fish
density, recruitment and straying, and fragmentation must be measured.
The first, primary order of population monitoring is the ESU, since both NMFS and CDFG have used this
population/evolutionary line description to evaluate the status of anadromous fish species (Weitkamp et al.
1995; Busby et al. 1997; Myers et al. 1998; CDFG 2002). NMFS recently proposed designating the
anadromous form of steelhead as Distinct Population Segments (DPS) rather than ESUs (FR 71(3):833862: see FR 61(26):4721-4725 for description of DPS). However, this proposed re-designation does not
alter how steelhead will be assessed and monitored.
The second, primary scale of monitoring is the population-level. At present, it is anticipated that all
populations identified as functionally independent (FI) or potentially independent (PI) by federal technical
research teams (TRT) will be monitored. At this time, it is not anticipated that dependent populations (DP)
will be monitored in central and northern California. Viability analysis of Oregon coastal coho salmon
establishes a president for this strategy. However, it is anticipated that DPs and ephemeral populations
(EP) likely will be part of the analysis for south-central and southern California steelhead. The final
decision for inclusion will be decided once assessments are conducted in southern California (see below)
and once the South-Central California Coast TRT publishes in population structure report. It is also
possible that DPs will be monitored within the Central California Coast coho salmon ESU. This decision
will be made in consultation with the North-Central California TRT.
An additional, third scale of monitoring will be conducted under the CALPLAN. This level is necessary
both for PVA and to implement the California strategy for recovery coho salmon. This level of monitoring
is referred to in several ways. However, all of these concepts have a common foundation: the delineations
are based in geography and similar environmental and ecological conditions. For central and northern
California ESUs, the stratification is referred to as diversity strata and for southern and south-central
steelhead ESUs, as populations groups.
The recovery strategy for coho salmon established a similar delineation, referred to as recovery units. Until
CDFG and NMFS integrate population parameters, population boundaries, and recovery metrics, it will be
necessary for evaluation of coho salmon population health and the progress of recovery using both the
federal and state stratifications. Until statistical and modeling decisions are finalized, the absolute number
of grouping that must be directly monitored is unknown. Initial decisions for southern California steelhead
and the obligation for monitoring under the recovery strategy for coho salmon already have been made.
Environmental Condition and Processes Status and Trends
The CalPlan will also assess and monitor fish habitat and watershed condition. Information collected will
be used in several analyses, including PVA, evaluation of recovery criteria, threats assessment, and
environmental trend analysis. Like fish population monitoring, environmental monitoring will occur at
several scales and address several parameters:
• Habitat Condition and Trend- status and trends in channel characteristics and key, priority
instream attributes related to fish habitat suitability and use;
• Watershed Condition- status and trends in watershed processes, riparian condition, thermal and
flow regimes, and geology;
• Estuarine Condition- attributes associated with fish survival, use, and occupation;
• Ocean Condition- attributes associated with fish survival and growth.
116
Future Iterations
The initial phase of the CALPLAN focuses on fish population dynamics and environmental conditions for
three species of anadromous salmonids. Future, phased implementation of the CALPLAN will address the
assessment and monitoring of additional fisheries priorities:
1. Remaining coastal, anadromous fishes;
2. Fish habitat restoration efficacy;
3. Fisheries management;
4. Hatcheries management.
117
Integrated Status and Effectiveness Monitoring Program (ISEMP) John Day Pilot
Project: The design and evaluation of monitoring tools for salmon populations and
habitat in the Interior Columbia River Basin.
Nick Bouwes
The John Day was chosen for a pilot project because (1) the John Day basin is
represented by several populations of the Mid-Columbia Steelhead ESU, listed as
threatened by NMFS in 1999 (64 FR 14517); (2) there is an opportunity to build quickly
on current research in the basin to develop answers to key management questions; (3)
past information suggests that focused research should address water supply, temperature,
and sediments as potential key limiting factors; (4) there is little influence of hatchery
fish in the basin; (5) the basin has key sites where fish traps can be effectively located;
and (6) the information from the research is likely to be transferable to other ESUs. The
John Day RME program will focus on two anadromous fish species: spring/summer
chinook (Oncorhynchus tshawytscha) and rainbow trout/Steelhead (Oncorhynchus
mykiss).
The John Day pilot project has outlined a plan to investigate modifying current RME
efforts in the basin to address local agency goals as well as the three programmatic goals
of Federal Tributary RME program: 1) to assess the ‘health’ or status of populations and
their environment; 2) to identify and prioritize restoration actions to increase freshwater
production as a means of increasing overall life-cycle survival; and 3) to quantify the
degree to which restoration actions are achieving its goals (Jordan et al. 2003). The
project relies on the collaboration of researchers and managers to: consolidate current and
historic information; define information needs; design research and monitoring programs
to establish status and trends, and causal relationships; and synthesize RME information
to determine limiting factors to prioritize and design restoration actions. Another major
component of the pilot project is to design methods to evaluate the effectiveness of
monitoring programs. The John Day pilot project is planning on evaluating project
effectiveness in an experimental management framework. Factors limiting fish
production in watersheds will be assessed, restoration actions will be designed to address
these problems, actions will be implemented as a large scale experiment (e.g. BACI
design), and intensive monitoring and research will evaluate the success of the action.
This study design is referred to as the Intensively Monitored Watershed (IMW) study,
and is not only a powerful approach to evaluate action effectiveness, but a small-scale
example of the process required to develop a basinwide RME program.
The development of the John Day RME pilot project is expected to be a long term
process whereby lessons learned through research and the Intensively Monitored
Watershed studies will influence current and future monitoring efforts. We are working
closely with personnel implementing current monitoring programs in the John Day as
changes to current programs will not likely be accepted unless we provide strong
justification of the benefit of action revisions to salmonid management in the basin.
Therefore, the RME pilot approach in the John Day is to develop a monitoring
framework developed as a cohesive, basin-wide RME program that will be agreed upon
118
by the multiple monitoring entities in the basin. Outlined below are the steps we have
been working on and plan to take to achieve a comprehensive RME program in the John
Day Basin.
Establish contacts and coordinate current activities. The John Day Basin is large and
crosses multiple jurisdictional boundaries. There are over 150 people from over 30
agencies and institutions that are connected with monitoring, restoration, or management
associated with fisheries in the basin. Historic and current monitoring programs have
collected over 25 years of fish, habitat, and water quality data within the basin. One
objective of the pilot project is to become well aware of current and historic efforts,
coordinating current efforts to reduce duplication, compile associated data and make it
available to all researchers, and to use this process to further working relationships with
local RME personnel.
In 2003, the Analytical Framework Group (AFG), made up of approximately 30 people
representing several agencies, was formed to identify information needs and coordinate
the development of research and monitoring projects. The AFG and the Habitat, Fish, and
GIS AFG subgroups continued to meet in 2005. In addition, pilot project coordinators
have met with approximately 100 local RME-related personnel, exposing them to pilot
project objectives, gathering existing project information and data, and assessing
monitoring needs within the basin.
We plan to further engage and broaden collaboration with personnel from USFS,
CTWSRO, CTUIR, SWCD, and other local agencies. Subbasin workgroups comprised
of agency biologists and restoration and management personnel will be formed to
increase communication and create a strong network among monitoring agencies. An
annual monitoring meeting is also planned to coordinate development of a basin-wide
monitoring study design, incorporating current and legacy monitoring activities of local
agencies.
Summarize current RME activities. Current status and trend fish and habitat monitoring
programs are implemented by state, tribal, and local agencies and these agencies are
represented in the AFG (Table 1). Activities include annual steelhead spawner surveys
(ODFW), water quality (USFS, BLM, ODEQ, OSU, CTWS and Monument SWCD), and
in-stream habitat surveys (EMAP and EPA). Their monitoring efforts have been
described in Bouwes (2004, 2005) as well as their initial proposals and annual reports.
Determine what information we need to inform goals. While there are a large number of
agencies and people conducting monitoring and research in the John Day, the size and
complexity of the basin ensures that large information gaps will exist within any single
monitoring program. An objective of the pilot projects is to assess multiple monitoring
programs to determine which information is essential, yet lacking, to reach status and
trend monitoring goals for the basin. Collaborations among agencies within the AFG and
review of the John Day Subbasin Plan (2005) have determined information gaps within
current monitoring actions.
119
Examples of information needs include, but are not limited to: population structure and
distribution information (e.g. genetic data) for salmon and steelhead; hatchery/wild fish
interactions, especially in areas with large percentages of hatchery origin salmon and
steelhead strays; fish passage monitoring; movement patterns of rearing salmon and
steelhead parr; high resolution remote sensing and riparian vegetation mapping; spatial
gaps in water quality monitoring; invertebrate monitoring; predation and food web
interactions; understanding spatial and temporal variability of collected samples;
compliance monitoring; implementation monitoring and project inventory; and
effectiveness monitoring. The pilot project and collaborators have initiated or are
planning studies to address these particular needs.
Establish causal relationships between ecological processes that control fish
production, and develop metrics and indices to better capture these mechanisms.
If we can understand the mechanistic relationships between fish and their environment,
we will more likely be successful in managing these populations. Our understanding of
factors limiting freshwater salmonid production will increase, and defining causal
relationships will assist our development of metrics and indices that capture relevant
ecological processes. The wealth of fisheries literature and past and current intensive
research in the John Day Basin, in part funded by other agencies (e.g. BOR), will help
define these causal relationships across multiple spatial and temporal scales. Several
monitoring projects in the John Day collect spatially and temporally overlapping fish and
habitat information that can be used to test causal relationships or generate new
hypotheses. We expect RME efforts and IMW studies will continue to discover and
refine causal relationships.
Several studies are currently underway in the John Day pilot project to determine causal
relationships and metrics that capture these relationships, such as fish movement and
water temperature relationships (OSU 2005). John Day Basin Sediment Assessment
(EPA) is evaluating sediment input sources, such as land use, road crossings, and channel
morphology, which are considered major limiting factors for fish production in the basin
(John Day Subbasin Plan). The EMAP- Western Regional Pilot Project is evaluating
relationships between physical and biotic factors and the ability to predict current ‘health’
of fish habitat. High stream temperature is considered another major limiting factor in the
John Day, and the John Day TMDL process is measuring several morphological and
riparian vegetation characteristics to parameterize the Heatsource model (Boyd, Matthew
and Kasper, Brian, 2002, www.heatsource.info), a mechanistic model relating these
inputs to stream temperature. This model will also evaluate how changes to factors such
as flow and riparian vegetation through restoration might impact temperature.
Invertebrate Productivity Monitoring. NOAA-Fisheries and Eco Logical Research/Utah
State University are currently evaluating macroinvertebrate metrics that potentially
predict fish growth when coupled with temperature metrics. The objective of this project
is to evaluate integrated metric approaches to estimate food availability for salmonids,
rather than exclusively using water quality and macroinvertebrate monitoring methods.
Project outcomes may include easy metric additions to habitat monitoring programs that
reflect food availability.
120
Landscape Changes and Patterns Project: Exploring Landscape Scale Influences on InStream Water Temperature. An existing time series of in-stream water temperature data
will be evaluated against the newly derived metrics of landscape complexity to assess
how landscape scale factors relate to measured in-stream variables.
Predation and food web interactions. Several species of native and non-native fish
species live in the John Day system. Very little is known about how these species
interact with anadromous fish. Some of these are warm water species that are potential
predators to juvenile salmonids, such as smallmouth bass, catfish, and pikeminnow,
especially in the migration corridor through which all anadromous species must pass.
Others, such as dace, shiners, and young pikeminnows, might compete with juveniles for
food resources. If so, parr density alone may not be the most relevant response variable,
if inter-specific competition is just as important as intra-specific competition. Many
warm water species seem to be expanding their range, which may further complicate the
success of some restoration actions. Conversely, restoration actions that decrease
temperature may decrease warm water species use of certain habitats and may provide an
added benefit to these types of actions. ODFW, the Subbasin Plan (2005), and the
ISAB/ISRP (2004a) have suggested that monitoring programs need to address these
interactions. We propose collecting diets and tissue or blood (non-lethal) stable isotopes
samples for all fish species and macroinvertebrates as a potential monitoring tool to
describe general food web patterns. Stable isotopes can describe general prey types
consumed for up to months. Collection of these samples during times of outmigration
through the lower mainstem, or times of high thermal stress above, at, and below the
interface of species overlap may elucidate these food web patterns.
Determine the accuracy and precision of information we need collected through
different protocols. The ability for a sample to provide accurate and precise information
depends heavily on the protocol used to collect that sample. Therefore, monitoring
programs need to adopt protocols that are the most informative yet are still affordable
(e.g. time, money, effort, etc.). This requires testing the accuracy and precision of several
protocols. Currently, there are multiple monitoring protocols to measure the same general
characteristic. Often these protocols differ enough that, although they are attempting to
describe the same general characteristic, they produce different information. These
differences produce inconsistencies that make detection of changes difficult or
information that is not even comparable across areas where multiple protocols are
implemented. Effort is being spent in the John Day to use habitat, invertebrate, and fish
sampling protocols that provide reliable information and are reasonable to implement.
Many protocols that are used in the current RME program were tested in 2005. Testing
of protocols will continue throughout the life of this project but is expected to diminish as
protocol differences are defined.
Comparison of satellite remote sensing information collected at different levels of
resolution. As part of the Landscape Changes and Patterns Project, remote sensing data
derived from the LANDSAT 5 Thematic Mapper sensor is being compared against
IKONOS high resolution multispectral imager. While this information has higher
121
resolution, the temporal coverage is limited, and it is much more expensive than
LANDSAT 5 images. The objectives of this analysis is to determine what information is
lost by using coarser resolution yet less expensive LANDSAT 5 data, and whether the
coarser resolution is adequate to describe changes in landscape and landuse patterns
evaluations used in the pilot projects.
Comparison of Aquatic Survey Protocols Used to Measurement of Physical Habitat
Attributes within the John Day Basin Oregon. This study was led by the USFS PIBO
program in collaboration with the Pacific Northwest Aquatic Monitoring Program
(PNAMP), and the John Day pilot project. Several of the most widely used stream
survey protocols within the Columbia River Basin were used to survey a common set of
sites. To ensure sufficient data to evaluate crew variability, at least three crews from
each program collected habitat and physical information at each site. A total of 12 sites
in the John Day drainage were surveyed with these protocols; four sites falling into each
of the following geomorphic reach types; 1) step-pool, 2) pool-riffle, and 3) plane bed. In
addition to these protocols a remote sensing approach (LiDAR) was compared and a
more precise method (engineering survey) was conducted to establish a benchmark or
‘truth’. Results from this study will describe the precision and accuracy of, and to help
RME programs decide on, the protocols used to collect habitat information.
As part of the South Fork IMW study conducted by OSU, the following protocols were
evaluated. Devising less stressful fish capturing techniques. Three alternative methods of
counting and capturing fishes, believed to be less stressful to the fish, were compared to
electrofishing: (1) visual counts by divers, (2) divers chasing fish into bag seines
(“snorkel-herding”) (3) chasing fish into bag seines using an electrical probe that emits an
irritating electrical field, but not one set to stun (“electro-herding”). All the new
techniques demonstrated to be less intrusive. Diver effects on fish behaviors. The
response of O.mykiss parr to snorkel surveyors was assessed by placing a remotely
controlled video camera in the stream and waiting 30 minutes before attempting to enter
the stream again. They found that fish will scatter into cover even when a diver enters
the stream two pools below the pool to be inventoried. Calibration of fish sampling
techniques. OSU conducted a calibration study in an attempt rectify the biases of
different gears to count fishes described above. Preliminary results suggest that
approximately 25%, 35%, 45% of the mark-recapture estimated population size were
detected by visual counts, snorkel-herding, and electro-herding, respectively. In addition,
rotary screw traps are being calibrated against, PIT tag antennas that have been installed
just above the screw trap as part of the South Fork John Day IMW. They are evaluating
impacts of release time and location of marked fish used estimate trap efficiency.
Difference in these treatments might suggest that attention should be given to capture
induced behavioral modification on trap efficiency calculations.
The Invertebrate Productivity Monitoring study is evaluating the spatial and temporal
variability of drift and benthic invertebrate metrics. Drift samples were taken with 2 to 4
replicate nets in each riffle, with 1-3 riffles sampled at each of seven sentinel sites to
describe the spatial variability within riffle, within site and between sites, respectively.
Samples were also collected during the morning, mid-day, and evening over 3 days
122
consecutive days for 3 sample dates to describe temporal variability within days, between
days, and throughout the season, respectively. This information will be used to determine
the level of effort (samples size and design) needed for a monitoring program to estimate
relevant stream invertebrate information.
Many of these studies have just been completed or are still in progress. We plan to
analyze this information and based on results, provide recommendations to monitoring
programs. In addition, several of these studies will be repeated to see if these sources of
variability are similar between years (e.g. calibration of fish and macroinvertebrate
sampling techniques).
We plan to also evaluate the level effort or resolution required to estimate a metric. For
example, in the South Fork IMW approximate 8,000 juvenile O. mykiss have been PIT
tagged. These tagged fish must migrate past the rotary screw trap (set daily) and a set of
PIT-tag antennas just upstream from the trap. If the ratio of tagged to untagged fish
remain relatively constant (an assumption of mark-recapture methods), then analysis of
this paired information may suggest that continuous reading of the antennas can be used
to estimate production of the subbasin, with much less effort spent collecting outmigrants
in the screw trap. We will also test this assumption in the Bridge Creek IMW. Further
review of screw trap data may also suggest that daily operation is not required, even in
the absence of antennas. Additionally, subsampling and comparison of high and low
resolution LiDAR data can lead insight into the level of resolution needed to provide
accurate estimates of stream characteristics relative to other habitat monitoring protocols.
We will also evaluate the ability to create cross walks between different sampling
protocols. For example, ODFW is willing to substitute LiDAR information in lieu of
habitat surveys, if LiDAR can provide the same information. This will allow ODFW to
increase snorkel survey reach lengths to estimate of juvenile salmonid abundance and
distribution for the same level of effort (the same crews conduct habitat and snorkel
surveys). Also, in the Invertebrate Productivity Monitoring study, we have collected
benthic and drift samples. PIBO collected drift samples with their standard benthic
invertebrate monitoring. If a relationship between these two methods can be developed
and we develop a metric that relates to fish growth, standard benthic invertebrate
monitoring samples may be available to predict fish growth.
Given what we learned about the accuracy and precision of different protocols,
develop a sampling design. While accuracy and precision of a sample is needed, the
ability to extrapolate a collection of samples to provide an accurate assessment at the
appropriate scale is dependent on the sampling design. The sampling design describes,
when, where, and how much to sample for a given sample area. The design is highly
dependent on the protocols used to collect the information and how the information will
be used. Protocol testing and variability comparisons will be used to improve monitoring
program design in the basin..
In conjunction with the RME monitoring program within the target basins, NOAAFisheries has initiated a Watershed Classification project to group similar 6th field
123
watersheds (Hydrologic Unit Code 6, HUC 6) in the Pacific Northwest according to
watershed characteristics. MCLUST software will be used to classify spatially
continuous GIS data such as precipitation, temperature, watershed topography, geology,
and hydrography metrics into discrete classes. This regionally continuous classification
of watersheds can be used to determine if the selected Intensively Monitored Watersheds
equally represent the landscape diversity. In addition, this classification scheme may be
provide a way to stratify monitoring sites for analyses, and as a means to extrapolate the
results beyond the boundaries of the study.
A similar classification approach to describe reaches was applied to the South Fork John
Day IMW. In 2004, a census of habitat and the distribution of fishes was conducted in
the SFJD, Black Canyon, Deer, and Murderers creeks. Data were gathered from FLIR
imagery, LiDAR images depicting vegetation structure, channel morphology and basin
topology, aerial photographs of push-up dams, multispectral satellite imagery and maps
identifying patterns of public ownership. This information was used to define discrete
reaches that formed the sampling strata that greatly reduced sample variance. Sentinel
sites (defined as a 5 pools sequence) or sites that would be sampled repeatedly, were
established based on this strata. These six sentinel sites represent the range of thermal
and habitat characteristics in the study streams and allow for contrasts to be made
between different habitat features to elucidate their influence on the different fish
responses. Bootstrapping procedures were used to sample the census information to
determine the number of samples required to produce an accurate picture of the census
area with a defined level of statistical power.
Based on the results of the testing of protocols and information from current monitoring
efforts, we will assess the level of effort (i.e. sample sizes) sufficient to detect defined
differences. Status and trend monitoring information has been collected by ODFW for 2
years and therefore the ability to address inter-annual variability and its implication on
sample designs is limited. We will use studies identifying causal relationships and
limiting factors to help describe where monitoring efforts need to be focused. We will
also conduct analyzes like the reach classification approach described above to develop
effectiveness monitoring designs. We will continue to refine sample designs over the
life of this project as new information becomes available.
Putting all the pieces together that result from implementing tasks under objectives 1 – 4
allows us to determine factors limiting populations in order to prioritize and develop
restoration programs and as hypotheses to test. There are few examples of effective
restoration actions in the Columbia River basin, and this is partially because actions often
do not address the limiting factor within the study area (Roper et al. 1997). Causal
relationships, as defined by current research and pertinent literature, between fish
production and habitat variables should be assembled into a framework (e.g. models) that
allows us to assess factors limiting fish production in the John Day Basin. Creation of
basin specific evaluation tools will be a synthesis of all lessons learned in specific John
Day pilot projects.
124
We are working with the John Day subbasin planning participants that have implemented
the Ecosystem, Diagnostic, and Treatment (EDT) model to assess limiting factors and
prioritize restoration efforts. EDT results have been summarized for each watershed in
2005. Some shortcomings to this general model’s application to the John Day have been
identified, and recommendations to produce more basin specific models have been made
(Subbasin Plan 2005). Sediment assessments and TMDL efforts are expected to
completed in 2007 and provide addition information on limiting factors in the John Day
Basin.
The John Day pilot project has initiated or collaborated on several studies that are
evaluating relationships between environmental factors and/or how these environmental
factors govern fish populations across multiple scales. At the landscape scale, satellite
remote sensing information, summarized in GIS, is being used to describe Landscape
Changes and Patterns and its potential impact on in-stream temperatures. Landscape
information is also being used to estimate in-stream Sediment Assessment, habitat quality,
and fish distribution. Airborne LiDAR and TIR Remote Sensing collects data at a much
higher resolution at a watershed/reach scale and is being used to inform these analyses as
well as the TMDL project. The TMDL project and the Sediment Assessment project are
relating riparian habitat information to in-stream temperature and sediment loading,
respectively. At the reach level, the South Fork IMW is relating temperature and habitat
characteristics to growth, density, survival, and production of juvenile O. mykiss across
different seasons. At this scale, the Invertebrate Productivity Monitoring study is relating
invertebrate abundance to fish growth. These projects have the potential to be
synthesized in a hierarchical fashion to estimate limiting factors across several scales.
We propose to use structural equation modeling (SEMs) and hierarchal models to
synthesize this information across several scales We plan to use this set of integrated
relationships to describe data collected through basin-wide monitoring to produce a
limiting factor analysis.
The above synthesis will require completion of a number of projects and thus will not be
available in the near term. Therefore, other models that focus on a reduced set of
potentially limiting factors (as judged by researchers in the John Day) might be
constructed for short term use. For example, stream temperature patterns are a
consequence of the interaction of important physical and ecological processes acting
within the landscape (Ward 1985). The HeatSource model used in the TMDL process,
describes many of this physical processes to define a heat budget for a reach. Watershed
Sciences is developing algorithms to process LiDAR information that can be used as
direct inputs into the HeatSource model. As is done in the TMDL process, impacts of
different scenarios, such as the increase of the riparian canopy through a riparian fencing
project or increased discharge by purchasing instream water rights, on stream temperature
can be estimated with the HeatSource model. The Invertebrate Productivity Monitoring
study is developing a relationship between prey density and temperature dependent
consumption rates (as estimated by the bioenergetics models) as has been observed with
other fishes (Budy 1996). This simple relationship could be used to estimate growth
potential of different stream reaches based on invertebrate and temperature data.
Incorporated with the HeatSource model, which describes temperature regimes under
125
restored and current conditions, this set of models could identify where temperature and
invertebrate production limits fish production. Restoration activities addressing these
factors can then be prescribed for these reaches.
Determine the effectiveness of restoration actions through effectiveness monitoring
or an experimental management framework, such as the Intensively Monitored
Watershed studies. Restoration actions are being implemented in the John Day on an
opportunistic basis. Restoration is often applied to degraded habitat under the
assumption that this will have a benefit to fish. However, this assumption is rarely based
on a limiting factors analysis and is rarely followed up on with monitoring to evaluate
restoration effectiveness. Many people working on RME projects in the John Day agree
that effectiveness monitoring is sorely lacking. The pilot project is developing
Intensively Monitored Watershed studies to design and test the impacts of restoration as
outlined in the Tributary Federal RME program. In addition to testing restoration
effectiveness, the comprehensive approach of IMWs will also address the above steps
and therefore inform whether this plan to develop a basinwide RME program needs
modifications. In 2005, we identified a set of potential IMWs.
The South Fork of the John Day River Intensively Monitored Watershed Study. We are
working collaboratively with the BOR (the funding agency) and OSU (contracted to
conduct the majority of the research) to develop the South Fork John Day (SFJD) as an
Intensively Monitored Watershed Study. The goal of this study is to examine the
influence of push-up dams on the production of redband steelhead trout (Oncorhynchus
mykiss - RBT). A related goal is to evaluate whether or not Lay Flat Stanchion Dams
(LFSD) will alleviate the putative problems caused by push-up dams. OSU conducted
extensive habitat and fish surveys collected in 2004 and 2005 in the SFJD and its
tributaries that are accessible by RBT (see Juvenile Production Project section above) to
form the baseline or pre-treatment information for this study. All push-up dams on the
lower SFJD will be replaced the end of the summer of 2006.
Bridge Creek Intensively Monitored Watershed Study. The purpose of the project is to
engage in watershed restoration at a scale sufficient to cause a detectable, populationlevel impact to the steelhead and that utilize this watershed. This project proposes to take
a process-based approach to restoration of Bridge Creek to cause aggradation of
approximately 4 km of incised mainstem stream bed and lower tributary reaches. Such
aggradation should raise floodplain water tables and lead to increased summer
streamflows, decreased stream temperatures, a narrower and more sinuous stream
channel, and a vastly expanded riparian forest. Thus we propose to restore one process
(aggradation), which in turn will trigger a series of positive feedback loops that restore
other biological and physical processes that maintain stream ecosystems. Comparison of
observed parr densities in Bridge Creek with densities in nearby healthy watersheds
suggests that the restored reaches will potentially increase steelhead rearing capacity
approximately 30 fold.
126
We propose 3 restoration structures that mimic strong, long-lasting beaver dams, in order
to accelerate aggradation processes. These include: (1) Beaver dam assisting devices or
post-pile fences: insertion of vertical posts or poles across the incised trench to provide
key members for beaver to build dams off of; (2) construction of small diameter log
bundles placed perpendicular to the stream in a series of steps; and (3) construction of a
series of rock steps. Each strategy should result in elevated water tables sufficient to
expand the riparian gallery forests and thus create a long-term supply of structural
material for the construction and maintenance of more beaver dams. All of these
structures should become “invisible” in a relatively short time period as aggradation and
vegetative colonization occurs. The three strategies range between the relatively low cost
option of enhancing habitat so that beaver will build stable dams versus the higher cost
option of actually constructing sediment aggrading structures. We have identified two
general areas along the mainstem of Bridge Creek totaling approximately 4 km in length
where criteria is appropriate and the likelihood of success is high. Criteria selection
included stream gradient, incision depth, floodplain width and presence of upstream
sediment supplies.
Learn and adapt. Given what we learn from implementing the above 5 steps, we will be
able to provide rationale and suggestions on how to change current monitoring designs to
best address goals in the John Day River basin. While we are working on completing the
above steps, we will be working with current monitoring efforts to determine near term
alterations that may need to be implemented to provide necessary information.
Ultimately, we would like to apply what we have learned through completing the above
steps to develop a well-coordinated and comprehensive monitoring program in the John
Day Basin as an example of how other basins should develop an RME program. The
success of the pilot project will be dependent on the ability to develop and maintain
working relationships with those who will ultimately be conducting the RME program.
Collaboration and demonstration of how the RME process can be improved will likely
result in an adoption of these suggestions. While this is not a fast track process, we
believe that this will lead to a longer term, well coordinated RME program than if we
immediately insisted on implementation of a theoretical RME plan.
127
Table 1. Salmonid and habitat Status and trend monitoring programs currently implemented in the John Day Basin. ‘*’ denotes
projects initiated by the John Day pilot project.
Agency &
Monitoring
Program Title
General characteristics
Specific indicators
Sampling frequency
Design
Protocol
Spatial scale
ODFW- Salmonid
Productivity,
Escapement,
Trend, and
Habitat
Monitoring in the
John Day Basin
Adults/redds
steelhead
Chinook
Escapement/Number,
Distribution, age structure, sex
ratio, size, genetics
adult, redd, and carcass
surveys
samplereach/watershed
estimate- subbasinbasin
Parr/Juvenilessteelhead/redband
trout,Chinook, cutthroat
Abundance, distribution,
relationships with habitat features
snorkel survey where
possible else electroshocking
sample-reach
estimate-reach to
subbasin
Smolts- steelhead, Chinook
Production, abundance, size,
genetics
smolt trap at base of
subbasins, mark-recapture to
estimate capture efficiency
subbasin-basin
SAR- steelhead, Chinook
smolt-to-adult survival rates
Steelhead- 50 random spatially
balance (17every yr, 16 every 4yrs,
17 new/yr) & long term index
sites
Chinook-census over known
extent, random beyond know
extent
50 random spatially balanced sites
(17every yr, 16 every 4yrs, 17
new/yr)
same sites as habitat survey
Sampled daily during
outmigration, estimated annually,
for SF, NF, MF and lower
mainstem
per cohort, tag a minimum of
5,500/annually, minimum of 500
per subbasin (SF, NF, MF, lower
JDR)
PIT tag fish captured in
smolt traps, detect adults
returning at Bonneville dam
subbasin-basin
Habitat status and trend
monitoring
LWD, pools, riparian cover,
gravel, fines
see
http://osu.orst.edu/Dept/O
DFW/freshwater/inventory/
pdffiles/habmethod.pdf
reach-basin
Macroinvertebrates-localized
assessments
Species/Community
50 EMAP random spatially
balanced sites
(17every yr, 16 every 4yrs, 17
new/yr)
same sites as habitat survey
annual
kick net samples
Reach/Site
Measure Clean Water Act
compliance
Temperature, alkalinity, bod,
chlorophyll a, cod, conductivity,
DO, e. coli, fecal coliform,N, P,
pH, turbidity
100+ habitat surveys, LiDAR,
TIR,7 days for 2 yrs at 5 sites all
other parameters, TMDL model
for NF, MF, SF, mainstem
LiDAR, FLIR, vegetation
morphological surveys, temp
loggers, hyrolabs, model
temperature load
sample-reach
estimate-reach to
subbasin
FS/ODEQ/OSU
CTUIR/
CTWSRO
ODEQ - TMDL
Project, Ambient
Monitoring
128
Water quality-localized
assessments of water quality,
3 hydrolabs by CTWSRO,
3 hydrolabs by SWCDs
habitat monitoring
Temperature, alkalinity,
conductivity, DO, pH, turbidity
Hydrolabs sample hourly,
site samples hourly to annually
hydrolabs, temperature
loggers
Site
Channel morphology, riparian
vegetation, substrate,
macroinvertebrates
GRTS sampling design through
Columbia Basin
5 site in JDB every year, 55 sites
every 5 yrs
protocols are described in
Henderson et al. (2005)
sample-reach
estimate-subbasinmultiple basins
implementation monitoring
to evaluate impacts of
grazing
riparian vegetation, stream bank,
and bankfull width measurements
Field Units (USDA 2003)
51 sites in JDB
protocols are described in
Henderson et al. (2005)
sample-reach
estimate-subbasinmultiple basins
ODEQ-EMAP
Western Pilot
Project
habitat and biological
monitoring
Channel morphology, riparian
vegetation, substrate,
macroinvertebrates, other
biological indicators
100+ EMAP random spatially
balanced sites
20 reference sites, sampled
between 2000-2004
EMAP protocols
sample-reach
estimate-reach to
subbasin
Subbasin Planning
Process
EDT compilation of expert
opinion
reach assessment of survival
impacts of 49 environmental
variables
nearly 1300 reaches set up in
EDT to describe current and
historical conditions, nearly
160,000 'data' entries
expert opinion
reach-basin
*UW/NOAAF/
BOR-Landscape
changes and
patterns
remote sensing of landuse
and landscape patterns
changes through time
agriculture, urban, logging,
vegetation type and health,
riparian vegetation, roads
1980-2000 every 4 yrs, 30m pixel
for Columbia River Basin
Remote sensing data derived
from the LANDSAT 5
Thematic Mapper sensor
watershed-basin
*EPA/NOAAFJohn Day
Sediment
Assessment
landscape analysis of
sediment loading, reach scale
sediment assessment
relative bed stability (RBS)
72 EMAP random spatially
balanced sites
21 sites, sampled between 20002003
use GIS information,
instream channel
characteristics to predict RBS
sample-reach
estimate-reach to
subbasin
*Watershed
Sciences- Airborne
LiDAR and TIR
Remote Sensing
LiDAR and thermal infrared
remote sensing of
watersheds
riparian vegetation height and
cover, channel and valley
morphology characteristics,
longitudinal thermal profiles
LiDAR- one time census for large
reach or watershed, approximately
15 cm pixel
TIR - one time census for large
reach or watershed, 0.5-1.5m pixel
LiDAR (measures laser
points of reflectance)
Thermal Infrared (TIR)
reach-watershed
CTWSRO,
CTUIR, *SWCDs
USFS-PACFISHINFISH Biological
Opinion (PIBO)
Effectiveness
Monitoring
Program
129
Salmon River Integrated Status and Effectiveness Monitoring Project
Introduction
The Integrated Status and Trend Monitoring Project (ISEMP) is an ongoing collaborative
effort to design, test, implement and evaluate Status and Trends Monitoring for salmon
and steelhead populations and their habitat and watershed-scale Effectiveness Monitoring
for management actions impacting salmon and steelhead populations and habitat in the
Interior Columbia River Basin. ISEMP takes a pilot-project approach to the research and
development of monitoring by implementing experimental programs in several major
subbasins of the Interior Columbia: the Wenatchee, Entiat, John Day, South Fork Salmon
and Lemhi River basins. The overall goal of the project is to provide regional salmon
management agencies with the data, information and tools necessary to design efficient
and effective monitoring programs. Specifically, ISEMP will generate quantitative
guidance on and examples of: the robustness and limitations of population and habitat
monitoring protocols, indicators and metrics; sampling design approaches for the
distribution of monitoring effort in time and space; analytical approaches to the
evaluation of monitoring data, information and programs; effective data management and
communication designs that support the use, standardization and compilation of
implementation, compliance, status, trends and effectiveness monitoring data by regional
data generators and decision makers; and finally the design and implementation of
watershed-scale restoration actions to maximize the quantifiable biological impact of the
actions and learning from the design and implementation strategy.
Within the Salmon River subbasin, a habitat and population status and trend monitoring
project is proposed for the South Fork Salmon River (SFSR) and a habitat action
effectiveness monitoring project is proposed for the Lemhi River. The following sections
summarize the primary elements of each of these projects and a concluding section
describes the status of the projects.
South Fork Salmon River Population and Habitat Status and Trend Monitoring
The South Fork Salmon River (SFSR) was selected to part of the ISEMP because: 1) it
represents a substantial contributor to Snake River spring/summer Chinook salmon
production and 2) ongoing monitoring activities are accompanied by a substantial time
series and present an opportunity to evaluate sampling alternatives in a “common
garden.”
The SFSR supports listed stocks of native summer-run Chinook salmon and B-run
steelhead (Matthews and Waples 1991). Historically, the South Fork Salmon River
(SFSR) was the single most important summer Chinook salmon spawning stream in the
Columbia River basin, accounting for approximately 50% of the summer Chinook
salmon redds enumerated in Idaho (Mallet 1974). Declines in natural escapement within
the SFSR have paralleled those of other Snake River stocks, resulting in their listing
under the Endangered Species Act (ESA). The Interior Columbia River Basin Technical
Recovery Team (ICBTRT) identified three populations of spring/summer Chinook
salmon in the SFSR, including the mainstem SFSR, the Secesh River, and the East Fork
130
SFSR (EFSFSR; ICBTRT 2003); together forming a single Major Population Group
(MPG).
A substantial number of Research, Monitoring, and Evaluation (RME) projects are currently
underway in the SFSR, primarily aimed at assessing the effectiveness of artificial propagation
at increasing the abundance of adult spring/summer Chinook salmon. These ongoing activities,
managed by the Nez Perce Tribe (NPT) and Idaho Department of Fish and Game (IDFG),
utilize substantial infrastructure (Figure 1). In addition, the NPT and IDFG conduct spawning
ground surveys that cover nearly all of the known range of spring/summer Chinook salmon
spawning habitat.
However, existing projects in the SFSR (with the exception of the Secesh River for
spring/summer Chinook salmon) primarily address subpopulations (e.g., the Johnson
Creek component of the EFSFSR spring/summer Chinook salmon population and the
upper mainstem SFSR component of the mainstem SFSR spring/summer Chinook
salmon population). These subpopulation level estimates may or may not adequately
represent the larger population, and data currently cannot be aggregated to yield an MPG
level population estimate due to gaps in existing data collection (e.g., in the lower
mainstem SFSR area). In addition, much of the available data for steelhead and resident
salmonids is collected opportunistically by projects targeting spring/summer Chinook
salmon, thus it is unclear how effectively these data portray the status and trends of these
non-target species/life-histories. Finally, the SFSR lacks an integrated and standardized
habitat monitoring component.
The proposed SFSR status and trends work will generate habitat and population status
and trend data for spring/summer Chinook salmon at the reach, population, and MPG
spatial scales; and steelhead and resident salmonids (e.g., rainbow trout) at the reach and
population scale. As summarized below, the precision and reliability of estimates
generated using proposed methods will be compared to existing methods following
standardization of data. Habitat sampling effort will be allocated to fixed (trend) and
probabilistically (status) selected sites, building when possible on existing fixed site
surveys. A response design and habitat attributes will be selected based on the results of
the habitat protocol comparison project (PNAMP 2005).
Existing infrastructure within the SFSR will be supplemented by a combination of
additional juvenile and adult sampling devices and extended length PIT tag arrays (Figure
1). Juvenile PIT tagging will occur in each major tributary (mainstem SFSR, the Secesh
River and the EFSFSR) via probabilistically distributed and fixed site seining effort
(which may also simultaneously serve a habitat monitoring function). Adults will be PIT
tagged at the fishwheel located near the confluence of the SFSR and mainstem Salmon
River. Proposed infrastructure will enable estimates of juvenile and adult Chinook
salmon and anadromous steelhead abundance at the scale of the SFSR MPG (for
spring/summer Chinook salmon), individual populations, and reaches (in the case of Lake
Creek spring/summer Chinook salmon). Abundance, productivity, and life-stage specific
survival estimates generated by the proposed infrastructure will be validated by
comparisons to estimates derived from existing and proposed infrastructure.
131
Salmon River
A
B
Lake Creek
AB
AB
Secesh River
A
B
Johnson Creek
Weir
DIDSON
Video
Fishwheel
A
B
Mainstem SFSR
Rotary Screw Trap
EFSFSR
PIT Tag Array
Figure 1. Location of existing (black) and proposed (red) juvenile and adult sampling
infrastructure in the SFSR.
Juvenile Abundance and Survival Estimates
Population specific and aggregate estimates of juvenile abundance and survival will be
estimated using information obtained from the recapture of juveniles PIT tagged during
seining surveys in each major tributary (mainstem SFSR, the Secesh River and the
EFSFSR). Recapture data for tagged individuals will be derived from repeat seining
surveys, interrogations at extended length PIT tag arrays, via recapture at the proposed
lower mainstem rotary screw trap, and at existing tributary rotary screw traps. Efficiency
estimates for rotary screw traps will be generated in two ways:
1. via marking a subsample of captured juveniles for release upstream of the trap
and performing a simple mark-recapture and
2. using the proportion of previously PIT tagged juveniles (e.g., tagged during
seining surveys) that are detected at the extended length PIT tag arrays
downstream of the screw traps that passed the traps without capture.
Efficiency estimates of extended length PIT tag arrays will be generated using two
methods:
1. via the proportion of juveniles PIT tagged at upstream rotary screw traps that are
not recorded at the PIT tag arrays during passage and
132
2. mathematically by calculating the probability that a PIT tag would not be detected
given the numbers of PIT tags detected at the upstream span of the tandem array
(A), but not the downstream span (B), the number detected B but not at A, and the
number detected at both A and B.
Expanding the previous example to the entire SFSR, fish tagged in each of the major
tributaries (Mainstem SFSR, EFSFSR, and the Secesh River) will be detected at the
lower mainstem rotary screw trap and via PIT tag interrogations at tributary specific PIT
tag arrays and the PIT tag array located on the lower mainstem SFSR. Population specific
juvenile abundance estimates obtained via proportional partitioning of the SFSR estimate
will serve two purposes:
1. to compare the precision of partitioning an SFSR scale juvenile abundance
estimate versus tributary specific estimates generated using rotary screw traps and
2. to evaluate the proportion of juvenile production not sampled by existing
infrastructure due to the placement of that infrastructure above varying fractions
of the spawning and rearing habitat.
Adult Escapement Estimates
Adult escapement estimates are currently generated by a Dual-frequency IDentification
SONar (DIDSON) array in the Secesh River, a video weir in Lake Creek, a temporary
picket weir in Johnson Creek, and a reinforced weir located in the upper mainstem SFSR.
The enumeration sites on the Secesh River have been extensively validated and
escapement estimates are accompanied by variance estimates. The weirs operated on
Johnson Creek and the upper mainstem SFSR utilize mark-recapture (via marking adults
at the weir and recovering marks during carcass surveys) to provide efficiency estimates,
thus enabling variance estimation for total escapement past the weirs. None of these sites
is currently operated to enumerate steelhead escapement, and the estimates provided by
enumeration sites on the mainstem SFSR, Johnson Creek, and the Secesh River are
located above varying fractions of the spawning habitat utilized by spring/summer
Chinook salmon. Because existing infrastructure does not enumerate total escapement for
any of the SFSR populations (with the possible exception of the Secesh River),
population level escapement estimates and estimates of escapement for the SFSR
spring/summer Chinook salmon MPG cannot be calculated. In addition, escapement of
steelhead is not currently estimated for either of the SFSR populations. We propose the
operation of a fishwheel on the lower mainstem SFSR to collect, PIT tag, and visually
mark returning adult spring/summer Chinook salmon and steelhead. Total escapement to
the SFSR for each species will be generated using the efficiency of the fishwheel to
expand captures. Efficiency estimates for the fishwheel will be generated using two
methods:
1. calculating the proportion of PIT tagged adults (originally tagged as juveniles via
proposed seining surveys and at existing rotary screw traps) detected at the
extended length PIT tag array located below the fishwheel that escape the
fishwheel and
2. by releasing a subsample of the adults captured and tagged at the fishwheel
downstream of the fishwheel and calculating the proportion that are detected at
the PIT tag array, but are not recaptured at the fishwheel.
133
Method one is preferred, as it limits handling stress, however the number of returning
adults with a PIT tag may be insufficient to generate efficiency estimates across strata
that are expected to effect efficiency (e.g., changes in flow).
Tributary and reach specific Chinook salmon and steelhead escapement will be estimated
via interrogation of PIT tagged adults at extended length PIT tag arrays located in each
tributary. Escapement estimates for spring/summer Chinook salmon generated via
parsing of the total SFSR estimate will be validated using the proposed lower Secesh
River DIDSON site, which will enumerate total escapement to that tributary for both
spring/summer Chinook salmon and steelhead.
The monitoring program development work piloted in the SFSR will also provide
retrospective statistical analyses to provide variance estimators for historical time series
data, and the development of statistical relationships between estimators derived from
proposed versus existing infrastructure and analytical methods. Thus the SFSR Pilot
Project is moving beyond the side-by-side protocol/indicator/metric tests underway in the
Wenatchee and John Day Pilots to experiment with an entirely new status and trends
monitoring approach designed top-down from the resource data management needs. For
example, IDFG has completed redd surveys for spring/summer Chinook salmon in index
areas of the mainstem SFSR, Johnson Creek, the Secesh River, and Lake Creek since the
late 1950’s. Surveys cover non-randomly selected reaches located in “important”
spawning areas for spring/summer Chinook salmon that are surveyed in a single pass
timed to correspond with the peak of spawning (Hassemer1993). These data provide an
important time series of redd abundance trend data, that have been widely utilized in
listing decisions and risk analyses (Holmes 2001, 2004a, 2004b, Hyun and Talbot 2004,
McClure et al. 2003). Nonetheless, these data are not accompanied by variance estimates,
and the degree to which these redd counts represent the “population” of redds within a
given year is largely unknown. Finally, the expansion of redds to estimate adult
escapement is accompanied by numerous assumptions (Roger and Schwartzberg 1986).
We propose to use the validated adult escapement estimates generated by this proposal to
evaluate the utility of the IDFG redd count time series as a measure of relative
abundance. Observed variance in the relationship between escapement and IDFG redd
counts will be used, if possible, to retrospectively evaluate the precision of the time
series. If a significant relationship can be developed, managers could elect to implement
escapement methods developed by this proposal as a possible replacement for single pass
or extensive redd surveys.
As discussed previously, the infrastructure currently operated to provide juvenile and
adult abundance estimates within the SFSR watershed generally provide suboptimal
information for steelhead, and Chinook salmon estimates are generally applicable to
some fraction of targeted populations, and these data cannot be aggregated for MPG level
estimates. While proposed infrastructure enables reach, population, and MPG level
estimates, managers may be justifiably hesitant to replace existing infrastructure without
developing a relationship between the proposed estimators and existing estimators. Thus
the following relationships will be developed:
134
1. escapement estimates generated for the EFSFSR (generated using the proposed
methods) versus estimates currently generated for Johnson Creek (the primary
EFSFSR tributary);
2. escapement estimates for the entire mainstem SFSR (generated using the
proposed methods) versus those currently generated for the upper SFSR;
3. escapement estimates for the Secesh River (generated using the proposed
methods) versus those generated from the existing DIDSON site; and
4. juvenile emigrant abundance estimates currently generated by infrastructure in the
Secesh River, upper mainstem SFSR, and Johnson Creek versus those generated
by the proposed design.
Finally, the SFSR Pilot project seeks to evaluate the efficacy of the proposed
infrastructure for providing juvenile and adult abundance, survival, and productivity
estimates independently of all existing SFSR infrastructure. If implementation and
validation of proposed infrastructure and estimates are acceptable for management
purposes (i.e., are reliable and provide adequate precision), the deployment of a similar
study design and associated infrastructure may enable reach, population, and MPG level
status and trend monitoring, for relatively low cost, in locations that currently lack such
estimators (e.g., the Middle Fork Salmon River). Thus the proposed design and
associated infrastructure will be evaluated for its potential as a template that can be
applied elsewhere.
Lemhi River Habitat Action Effectiveness Monitoring
Habitat restoration activities in the Lemhi are guided by the Lemhi Conservation Plan
(LCP). Ongoing and proposed habitat actions are aggressive and anticipated to result in
measurable biological responses, both in terms of physical habitat attributes (e.g., quality
and quantity of accessible habitat) and fish vital rates (survival/productivity, distribution,
and abundance) both at the scale of individual reaches and at the scale of the watershed.
These habitat actions are the primary mechanism intended to stimulate the delisting of the
ESA listed TRT identified population of spring/summer Chinook salmon (Oncorhynchus
tshawytscha) and steelhead trout (O. mykiss) as well as bull trout (Salvelinus confluentus)
populations that reside in the Lemhi River subbasin.
Actions proposed in the LCP are intended to:
1. Remove or reduce upstream and downstream migration barriers to fish and
provide access to available spawning and rearing habitat by:
2. Maintain or enhance riparian conditions characteristic of good habitat to ensure
that adequate vegetation persists to provide shade, increase bank stability and
protection, decrease sediment input, and promote the recruitment of large woody
debris.
3. Decrease sediment, temperature, provide quality substrate, increase the abundance
and quality of off-channel habitat, and increase pool frequency and quality to
improve productivity and survival.
Because ongoing and proposed habitat actions in the Lemhi watershed are targeted to
efficiently address limiting factors that have a predefined spatial structure (e.g.,
135
reconnection of tributaries) and since these actions require significant landowner
cooperation, there is a limited ability to randomly allocate habitat actions. Thus,
evaluation of the effectiveness of many of the habitat actions will take the form of
design-based experiments (ISRP 2005). The primary goal of the Lemhi Habitat Action
Effectiveness Pilot Project is to identify and quantify the effects of classes of habitat
actions (e.g., channel reconnection, flow augmentation, physical changes in channel
morphology etc.) on the abundance productivity/survival, condition, and distribution of
anadromous and resident salmonids within the Lemhi watershed. Because multiple
habitat actions may be collocated, assessing the individual effects of classes of habitat
actions will require a modeling approach. In addition, because the LCP uses a phased
implementation approach, wherein the effectiveness of Phase I habitat actions are used to
identify and prioritize Phase II actions, the model-based approach will be useful for
conducting a priori simulations to estimate the potential effects of different suites of
proposed Phase II habitat actions. A statistical framework developed for the Lemhi Pilot
Project produced the following tools:
1. Development and application of a watershed model to evaluate survival,
productivity, and carrying capacity by life-stage as a function of habitat
availability and quality, with a simulation component to estimate life-stage
specific benefits from increased habitat availability and/or quality.
2. Development and implementation of a study design that utilizes reach-specific
empirical measures of productivity, juvenile survival, condition, and distribution
across habitat classes, in treated and untreated areas, to determine whether
tributary reconnection and other habitat actions have provided high quality habitat
that benefits fish vital rates (survival, growth etc.).
3. Development and implementation of a design that evaluates the movement and
distribution of anadromous and resident salmonids to address the following
questions:
a) Are anadromous fish utilizing reconnected habitat?
b) Have the reconnections changed the distribution and connectivity of
resident fish?
The following sections summarize functions of the model and design-based features of
the Lemhi Pilot Project Study Design (QCI 2005), and describe a sampling design to
meet the information needs of the approach.
Tool one takes the form of a watershed model (based on Sharma et al. 2005) that views
fish vital rates (survival/productivity, abundance, and condition) as a function of the
quantity and quality of available habitat. These functions are constructed using both
coarse (e.g., Geographic Information Systems (GIS)) and fine (e.g., reach) scale habitat
measures (Figure 2). Once validated via the collection of empirical data within habitat
classes, the model provides a statistical framework to assess the effects of different
classes of habitat actions on life-stage specific vital rates (productivity/survival and
condition) of anadromous and resident salmonids.
The information needs of the watershed model are addressed by a sampling design
(Figure 3) that yields reach specific and aggregated estimates of life-stage specific
136
juvenile abundance, productivity/survival, condition, and distribution as well as adult
escapement across habitat classes and within treated and untreated stream reaches. The
use of tandem extended length PIT tag arrays within the study design provide information
necessary to determine whether habitat actions (e.g., channel reconnection) have
increased population connectivity, and whether anadromous salmonids utilize newly
available habitat (Figure 3).
Adult Escapement
We propose to generate escapement estimates for anadromous salmonids (and possibly
adfluvial residents) by modifying an existing mainstem irrigation diversion (L6) to serve
as a trap or as a location for a video monitoring station (Figure 3). A trap is preferred, as
that would enable adults to be PIT tagged. Aggregate escapement will be partitioned into
reach specific estimates via the interrogation of PIT tagged adults at mainstem and
tributary tandem extended length PIT tag arrays (Figure 3). Efficiency estimates for the
mainstem fish trap will be generated using two methods:
1. calculating the proportion of PIT tagged adults (originally tagged as juveniles via
proposed seining surveys and at existing rotary screw traps) detected at the
extended length PIT tag array located below the trap that escape the trap and
2. by releasing a subsample of the adults captured and tagged at the trap downstream
of the trap and calculating the proportion that are detected at the PIT tag array, but
are not recaptured at the trap.
If adults cannot be captured at the L6 diversion, tributary extended length PIT tag arrays
would be accompanied by video weirs or DIDSON arrays to enable partitioning of adult
escapement.
Juvenile Abundance, Survival, and Distribution
Some of the habitat actions proposed in the LCP are anticipated to primarily influence
survival at specific life stages. For example, augmentation of pool habitat might be
expected to exert the greatest influence on juvenile overwinter survival, thus measures of
survival from the presmolt to smolt life-stage may be most sensitive to the effects of that
habitat action. The collection of life-stage specific vital rates provides a means to
potentially separate the effects of collocated habitat actions. Thus, methods to generate
life-stage specific juvenile productivity/survival, abundance, and condition are required.
Juvenile abundance estimates (primarily targeting emigrating spring Chinook salmon) are
currently generated via rotary screw traps at two sites in the Lemhi watershed. We
propose to operate an additional screw trap in Hayden Creek (Figure 3) and extend the
operating period of the existing screw trap in the upper Lemhi mainstem to better
estimate steelhead emigration. These efforts will be useful in generating population level
juvenile emigrant abundance estimates, estimating survival from earlier life-stages, and in
determining the proportion of juvenile production that occurs in Hayden Creek, a
potential reference stream.
137
Habitat Quantity
Habitat Quality
Channel Characteristics by Land Use Type:
A. Relating habitat availability to capacity,
(ci) 13 and 14;
B. Calibration using empirical and GIS data,
19-23;
C. Hypothesis testing, 29 and 30 (crosssectional), 34-38 (pre/post).
Egg
1-3, (N 2,t)
Spawner
1-3, (N 1,t)
Fry
1-3, (N 3,t+1)
Harvest (T)
11, (o t+x)
Survival/Productivity by Life History Stage:
A. Relating habitat quality to
survival/productivity, (pi) 15 and 16;
B. Calibration using empirical estimates of
survival/productivity, 24-28;
C. Hypothesis testing, 31 and 32 (crosssectional), 34-38 (pre/post).
Parr
1-3, (N 4,t+1)
Presmolt
1-3, (N 5,t+1)
Smolt
1-3, (N 6,t+2)
Adult 8-10,
(o t+x)
Mature (Yes)
Ocean
Immature
8-10, (o t+x)
Mature
Survival (5-7),
(O t+x)
Figure 2. Schematic illustrating how the model develops relationships between habitat
quantity (capacity) and quality (survival/productivity) and stage-based abundance,
productivity, and survival. Grey boxes indicate those life stages for which metrics will be
inferred, notation in parentheses refers to model parameters, and numbers within the
boxes refer to equations in the Lemhi Study Design (QCI 2005).
Estimates of pre-migratory juvenile abundance and survival in reconnected versus
currently available habitat will be generated using PIT tags and visual marks, deployed
via seining at fixed and probabilistically selected sites throughout the Lemhi watershed
(e.g., in currently available, reconnected, and isolated habitat). Recaptures of PIT tagged
juveniles will occur during repeat seining surveys, at extended length PIT tag arrays, and
at rotary screw traps. Efficiency estimates for screw traps will be derived by marking and
releasing a fraction of each days catch above the trap and calculating the fraction that are
recaptured at the screw trap. Efficiency estimates for extended length PIT tag arrays will
be derived in two ways:
1. by calculating the number of juveniles tagged in a given tributary that are detected
at downstream sites (e.g., the mainstem PIT tag arrays and screw traps) that were
not detected at the tributary PIT tag array and
2. mathematically by calculating the probability that a PIT tag would not be detected
given the numbers of PIT tags detected at the upstream span of the tandem array
(A), but not the downstream span (B), the number detected B but not at A, and the
number detected at both A and B.
It is anticipated that PIT tags can be deployed in both anadromous and resident salmonids
(e.g., rainbow trout). Juvenile survival will be estimated by evaluating overall
systemwide recapture efficiency, generated by common PIT tag detections at upper and
lower screw traps, determining the fraction of fish that passed PIT tag arrays (using the
efficiency estimates described above), and via repeat seining surveys. Juvenile
138
distribution and movement will be evaluated by scrutinizing recaptures of tagged fish
during repeat seining surveys, by interrogations at extended length PIT tag arrays, and via
capture at rotary screw traps.
Habitat Sampling
Habitat sampling effort will be allocated to fixed (trend) and probabilistically (status)
selected sites. A response design and habitat attributes will be selected based on the
results of the habitat protocol comparison project (PNAMP 2005). Because the design
requires estimates of habitat availability for specific life-stages of juvenile resident and
anadromous salmonids, habitat surveys will incorporate temporal replication to achieve
seasonal estimates of habitat availability.
Upper Salmon River
Lemhi River
L6 Diversion
A
B
Hayden Creek
A
B
A B
Tributary Reconnect
A B
Screw Trap
Tributary Reconnect
A B
A B
PIT Tag Array
Adult Capture Facility
Tributary Reconnect
A B
Tributary Reconnect
Figure 3. Location of existing (black) and proposed (red) infrastructure for the Lemhi
River watershed.
Status of the Salmon River ISEMP
Study designs for the SFSR and Lemhi River were completed in November 2005 (QCI
2005) and reviewed by the Independent Scientific Review Panel (ISRP) in January 2006
(ISRP 2006-1). In 2006 we have proposed to implement a subset of proposed activities
on a trial basis, and we have completed a proposal for funding that covers the 2007-2009
fiscal year.
139
Federal Caucus RM&E Program Overview - 3/10/06
Jim Geiselman
Federal Caucus RM&E Team Lead
What are we doing and why?

Obtaining information needed to answer key management questions for
performance tracking (status and trend monitoring), effective planning and
implementation of mitigation actions (action effectiveness and uncertainty
research), and adaptive management.

This information is required for ESA BiOps, ESA Recovery Planning, and the
Northwest Power Act (Columbia Basin Fish & Wildlife Program).
What steps have been taken?

Identified general need for this RM&E program in the Federal All-H Salmon
Recovery Strategy and the 2000 FCRPS BiOp.

Developed Annual Implementation Plans for FCRPS BiOp required RM&E.

Formed a Federal Caucus RM&E Team in 2002 that includes a Planning Group
that oversees workgroups for Hydro, Tributary Habitat, Harvest/Hatchery, and
Estuary/Ocean.

Provided F&W Program project proposal reviews and ISRP briefings relative to
BiOp needs.

Developed a Draft Federal RM&E Plan and an Upper Columbia Monitoring
Strategy in 2003. http://www.salmonrecovery.gov/reports_and_papers/research/

Developed a more detailed RME Plan for the Columbia River Estuary in 2004.

Obtained ISRP/ISAB and Regional Agency Reviews of the Federal RM&E Plan
in 2003 and 2004.

Started Pilot Projects in the Upper Columbia, John Day and Salmon Rivers to
further test and develop status and trend and action effectiveness RM&E
approaches.

Helped to charter, develop, and participate in the Pacific Northwest Aquatic
Monitoring Partnership (PNAMP) and the Northwest Environmental Datanetwork (NED) for advancement of regional RM&E coordination and data
sharing standards and strategies.
140

Provided funding, participation, and coordination for advancement of the
Collaborative Systemwide Monitoring and Evaluation Project for development of
RM&E designs and RM&E collaboration with State and Tribal Fish Agencies.

Based on a new 2004 FCRPS BiOp, provided updated RM&E strategies through
the 2005 Updated Proposed Actions
(http://www.salmonrecovery.gov/reports_and_papers/biological_opinions/) and
the 2005-2007 BiOp Implementation Plan
(http://www.salmonrecovery.gov/reports_and_papers/biop_implementation/docs/
2005_2007_IP_Final.pdf)

Worked with the Northwest Power and Conservation Council to integrate the
Federal RM&E Plan Framework with the F&W Program Draft Research Plan and
Draft Monitoring Plan.

Provided supplemental guidance to the Columbia Basin F&W Program 2007-09
solicitation process through BPA that included an RM&E framework consistent
with the Federal RM&E Plan framework and PNAMP strategies for monitoring
standards.
http://www.efw.bpa.gov/Integrated_Fish_and_Wildlife_Program/05RME_BPA_s
olicitation guidance1122b.pdf
What are some next steps?

Work within the BiOp Remand Collaboration Process RM&E Workgroup to
further identify FCRPS BiOp RM&E needs for performance tracking and
adaptive management.

Update RM&E strategies through the 2006 FCRPS Updated Proposed Action and
BiOp.

Develop additional details for Recovery Planning RM&E Framework elements.

Continue to work with the NPCC staff and the ISRP to integrate ESA RM&E
needs within the F&W Program Research and Monitoring Programs.

Form a new Federal Caucus Monitoring Responsibilities Subcommittee
(FCMRS). This subcommittee will attempt to develop long-term agreements on
federal agency level commitments to monitoring needs.

Help to develop and implement a regional survey of monitoring projects and
programs through PNAMP.

Use the new FCRPS BiOp RM&E strategies, the Recovery Plan RM&E
Framework, the FCMRS agreements, and the PNAMP monitoring survey to
141
inform the funding and sequencing of RM&E projects under the F&W Program
and other federally funded programs in the NW.

Continue to advance on PNAMP, CSMEP, NED and RM&E Pilot Projects to
develop RM&E protocols, designs, and tools. Use these products as they become
available as standard requirements of RM&E projects under the Columbia Basin
F&W Program and other federally funded programs in the NW.

Update the 2003 Federal RME Plan in late 2006 or early 2007 to incorporate new
RM&E strategies, agency level responsibilities, and monitoring standards that
reflect a Coordinated Regional Network of RM&E Programs based on the
integration of new BiOps, Recovery Plans, F&W Program, updated federal, state
and tribal agency programs, and the latest products from PNAMP, CSMEP and
Pilot Projects.
What are the key framework components of the Federal RME Program?

Management questions and information needs for strategic focus.

Inventory of regional monitoring programs.

Standard definitions or terminology for different RM&E projects.

Standard and/or compatible design, sampling, and data management protocols.

Responsibilities, commitments, and cost sharing agreements.

An organizing structure for the different components of the RM&E framework.
What are the high-level management questions that define the federal RM&E needs for ESA
salmon recovery?

Are we meeting biological and programmatic level performance objectives and
addressing population level threats identified by the ESA BiOps and Recovery
Plans?

Where abundance, productivity, survival, spatial distribution, or genetic diversity
objectives are not being met, what are the limiting factors or threats that need to
be addressed?

What mitigation actions are most effective at addressing these limiting factors or
threats?
What are the definitions for the different types of RM&E within the Federal RM&E Program?
142
1. Fish/Wildlife Population and/or Environmental Status and Trend Monitoring –
census or statistically designed monitoring of fish or wildlife population and/or
environmental conditions (i.e. watershed conditions) to assess the current status or
change (trend) over time. This is sometimes referred to as an observational study
(ISRP, 2005). These monitoring data may also be used to correlate fish
performance with environmental conditions.

Ecosystem/Landscape level, broad-scale, periodic monitoring (referred to as
Tier 1 Monitoring)
Geographically localized, frequent monitoring (referred to as Tier 2
Monitoring)
2. Action Effectiveness Research – research to determine the effects of an action or
suite of actions on fish survival, productivity and/or habitat conditions (referred to
as Tier 3 monitoring). This is a manipulative experiment that statistically assesses
the effect of a treatment (action) condition relative to a control or reference
condition. Action effectiveness research can be performed for a localized effect
(project or stream reach level effect) or for a watershed level effect (intensively
monitored effect). Localized (project level) effects most commonly identify
changes in habitat conditions associated with the action, while fish or biological
responses may require a watershed level (intensively monitored approach) to
capture a broader area in which a biological response is expressed.
3. Uncertainties Research – research to resolve scientific uncertainties regarding the
relationships between fish or wildlife health, population performance (abundance,
survival, productivity, distribution, diversity), habitat conditions, life history
and/or genetic conditions (e.g., the existence and causes of delayed mortality,
hatchery spawner reproductive success relative to wild populations, etc.). This is
a manipulative experiment where variables are manipulated to infer or
demonstrate cause and effect relationships using statistical-designed hypothesis
testing. Uncertainties research does not include experimental research and
monitoring specifically targeting the effect of a mitigation or restoration action
(this is Action Effectiveness Research). It also does not include monitoring
(observational studies) of fish or habitat conditions with inferences from statistical
correlation assessments (this is Status and Trend Monitoring).
4. Project Implementation and Compliance Monitoring – monitoring the execution
and outcomes of projects. This type of monitoring does not require environmental
response data directly linking restoration actions to physical, chemical, or
biological responses.

Project Implementation monitoring determines whether projects were
carried out as planned, through documentation of the type and location of
management action, and whether the action was implemented properly or
complies with established standards. This is generally carried out as an
administrative review and does not require any parameter measurements
beyond those specified by the project design requirements. It is usually a
143
low-cost monitoring activity that should be included for all mitigation
activities.

Project Compliance monitoring determines whether specified project criteria
are being met, through a post-project auditing of project performance. This
type of monitoring would typically not be carried out by the project sponsor,
and may require the development of independent, compliance monitoring
projects. A limited, statistical-designed sample of projects could be monitored
annually for compliance.
How do some of these framework components fit together?
Tributary
Habitat
Hydro
System
Estuary and
Ocean
Harvest
Hatchery
Predators
Invasives
Management Questions
Agencies with Common
Questions and
Information Needs
Standard Designs
and Protocols, Cost
Sharing
Agreements
Information Needs
Indicators, RM&E
Designs, Spatial and
Temporal Scales
Status and Trend Monitoring
Projects
Action Effectiveness
Research Projects
Critical Uncertainties
Research Projects
Planning, Evaluation, Adaptive Management
Coordinated with Regional Monitoring
Programs
144
Monitoring Strategy
The Oregon Plan for
Salmon and Watersheds
North
Coast
Lower Columbia,
Sandy, Hood
Umatilla / Walla Walla
Grande Ronde /
Imnaha
Willamette Lower
Deschutes /
Westside
Willamette Mid
Coast
Cascade
Trout
Creek
Johnday
(North, Middle, Upper)
Power River /
Burnt River
Upper Deschutes /
Crooked River
Malheur -
Mid-South
Coast
Umpqua
Lakes Basins /
Owyhee
Oregon Closed Basins
Rogue River - South Coast
Klamath
The Oregon Watershed Enhancement Board - 2003
145
Oregon Plan for Salmon and Watersheds
Monitoring Strategy Overview
Desired Outcomes
OUTCOME ONE
Assessment of watershed
conditions and salmon
populations
Framework Questions
Implementation Strategies
What is the condition and capacity 1. Assess status and trends of
of aquatic habitat and watershed
watershed conditions and salmon
systems?
populations regionally.
2. Monitor habitat, water quality,
biotic health, and salmon, in select
watersheds.
3. Analyze habitat, water quality
and population trends at the
landscape scale.
OUTCOME TWO
Evaluation of Oregon Plan
restoration actions,
conservation measures, and
management practices
What is the benefit of Oregon
Plan restoration projects,
management practices, and
conservation programs relative to
adverse impacts and to natural
ecosystem variability?
4. Document conservation and
restoration projects, activities, and
programs.
5. Evaluate effectiveness of
restoration and management
efforts locally.
6. Evaluate the combined
effectiveness of restoration and
conservation efforts in select
watersheds.
OUTCOME THREE
Application of monitoring
results for use by
policymakers, agencies, and
the public
Does the Monitoring Program
provide information and analysis
for adaptive review of restoration
actions, management practices,
and Oregon Plan policies?
7. Standardize monitoring
collection, management, and
analysis efforts.
8. Coordinate and support
public-private monitoring and
partnerships.
9. Integrate information and
produce data products and
reports.
146
A Monitoring Strategy for the Oregon Plan for Salmon and
Watersheds
Overview
Throughout its development, the Oregon Plan emphasized the importance of monitoring the status of environmental
factors that affect watersheds and habitat quality as well as monitoring salmon population status and trends. Support
for monitoring and reporting represents the State’s commitment to evaluate the benefit of measures implemented to
improve watershed conditions and salmon populations and to make changes in policies or programs when necessary.
With Executive Order 99-01, the Governor expanded the original monitoring program developed for the 1997 Coastal
Salmon Restoration Initiative (CSRI) to include all watersheds and salmon species and to the habitats of native fishes
throughout the state. In 2001, the Oregon Legislature “institutionalized” the Oregon Plan by placing the state
authorities in statue, including those directing OWEB to develop and implement a statewide monitoring program in
coordination with Oregon Plan agencies and partners. This monitoring strategy seeks to meet these far-reaching
commitments to effectively assess long term trends in watershed health and salmon recovery.
The Monitoring Strategy delineates the expectations and geographic scope of monitoring and describes the process for
planning monitoring activities to meet program expectations. Rather than monitor “everything, everywhere”, an
integrated monitoring strategy needs to address fundamental information needs and match the level of inquiry to the
extent of population and habitat
response at appropriate spatial
and temporal scales. This can be
achieved by employing a
hierarchical structure of multiple
spatial and temporal scales for
data acquisition and analysis.
Oregon’s monitoring efforts are
supported through collaborative
work with NOAA Fisheries
Northwest Fisheries Science
Center, the Washington State
Salmon Recovery Office, and the
Columbia Basin Fish and
Wildlife Authority. Working with
these groups, Oregon is
contributing to a consensus that
supports monitoring programs
throughout the region based on a
consistent conceptual framework,
compatible methods, and efficient
implementation.
147
Page - 2
A Monitor
ing S
egon P
lan for S
almon and Watersheds
Monitoring
Stt r ateg
ategyy for the Or
Oregon
Plan
Salmon
Strategic Framework for Monitoring
The Monitoring Strategy provides a framework to evaluate existing monitoring efforts and to expand efforts to assess
the effectiveness of Oregon Plan and OWEB activities. The Framework outlines three Desired Outcomes, identifies
Framework Questions, and describes nine Implementation Strategies to achieve these objectives.
•
•
•
•
Desired Outcomes establish the overall scope and expectations for the monitoring program. The outcomes
are based on the need for credible information to inform policy and program decisions for the Oregon Plan
and to evaluate the effectiveness of program implementation. Outcomes apply to each spatial scale monitored
and to the reporting and analysis of results. Integration and realization of the outcomes will proceed at
different rates in different regions of the State.
Framework Questions clarify the intent and identify the main information needs for achieving an Outcome.
Questions help focus the purpose of each Strategy and help connect specific monitoring activities to
information needed for management and policy decisions. Framework Questions are further developed into a
series of Key Questions for each of the Outcomes.
Implementation Strategies are the general approaches necessary to accomplish the Outcomes. Strategies
establish strategic monitoring subject areas and provide guidance for developing work plans. Ultimately,
monitoring Strategies must address issues of the distribution of sampling effort, the protocols and methods
used, and the way information is managed.
Framework Questions clarify the intent and identify the main information needs for achieving an Outcome.
Questions help focus the purpose of each Strategy and help connect specific monitoring activities to
information needed for management and policy decisions. Framework Questions are further developed into a
series of Key Questions for each of the Outcomes.
148
A Monitor
ing S
egon P
lan for S
Monitoring
Stt r ateg
ategyy for the Or
Oregon
Plan
Salmon
Page - 3
OWEB and the Oregon Plan
OWEB programs support Oregon’s efforts to restore salmon runs, improve water quality, and strengthen
ecosystems that are critical to healthy watersheds and sustainable communities. OWEB is responsible for three
interrelated monitoring functions:
• strategic guidance for cooperative monitoring
• accountability for restoration investments
• reporting on the progress of the Oregon Plan.
Recent legislation, Senate Bill 945, directs OWEB to develop and implement a statewide Monitoring
Program in coordination with other state natural resource agencies for activities conducted under the Oregon Plan. OWEB is also obligated to ensure investment in local restoration activities results in positive
environmental, cultural, and economic benefits. Finally, OWEB is responsible for the creation of a Biennial
Report that provides an assessment of the implementation and effectiveness of the Oregon Plan. OWEB’s
multiple roles require the integration of science based natural resource information, analysis of restoration
activity efficacy, and accountability reported via the Biennial reporting process.
149
Page - 24
A Monitor
ing S
egon P
lan for S
Monitoring
Stt r ateg
ategyy for the Or
Oregon
Plan
Salmon
Oregon Plan Monitoring Strategy Details
Outcomes
Outcome One:
Provide a scientific assessment
of watershed conditions and
salmon populations.
Identify the appropriate
indicators of population and
watershed condition, the
appropriate scales of inquiry,
and the appropriate level of
precision needed.
Outcome Two:
Provide an evaluation of
Oregon Plan restoration
actions and conservation
measures
Evaluate the relative
importance of restoration
activities as a contribution to
watershed health. Develop
analytical models to evaluate
changes produced by the
Oregon Plan to target
conditions and recovery goals.
Outcome Three:
Provide useful information to
policymakers, agencies, and the
public through efficient and
coordinated monitoring
Oregon Plan partners
coordinate to implement
efficient monitoring, employ
scientific assessments, and
report results in ways that
promote adaptive responses
and informed participation.
Questions
Strategies
Example Data
What is the condition of aquatic
habitat and watershed systems?
1. What is the condition of salmon
populations at the ESU, Sub-Basin
and watershed scale?
2. What is the status and what are
the trends in aquatic habitats, water
quality, and stream flow?
3. What are the critical factors that
limit watershed function and salmon
productivity?
4. What constitutes detectable and
meaningful changes in habitat
condition and populations?
1. Assess general status and trends for
physical habitat, salmon populations, ,
and biotic conditions in Oregon subbasins and ESU regions at appropriate
scales.
2. Monitor habitat capacity, salmon
survival and productivity, and biotic
processes in selected watersheds within
each sub-basin or ESU region.
3. Analyze habitat trends and salmon
populations in the context of local or
regional effects, landscape influences,
and ocean productivity.
Landscape Characterization:
Riparian Condition: canopy
composition, site potential.
Habitat Condition: channel
morphology, fish passage.
Salmon: abundance, geographic
distribution, life history, diversity, and
productivity.
Biotic Condition: invertebrate
communities, , toxics.
Water quality: temperature, DO, pH,
sediment, bacteria.
Stream flow: duration, peak flow
events, minimum flows.
What is the benefit of Oregon Plan
restoration projects, management
practices, and conservation programs
relative to adverse impacts and natural
ecosystem variability?
5. What changes are occurring in
watersheds that improve stream
habitat quality?
6. What are the management practices
and programs that enhance or restore
watershed functions and salmon
populations?
7. What habitat changes and biotic
responses result from these projects,
practices, and programs?
8. What are the impacts of land use
and land management practices on
watersheds?
4. Document implementation of
restoration projects, conservation
activities, and agency programs.
5. Evaluate the local effectiveness of
restoration efforts by monitoring
representative samples of specific
project, activity, and program types.
6. Evaluate the combined effectiveness
of restoration efforts by monitoring
habitat and population response in a
structured sample of watersheds.
Broad Scale Indicators: land
use/land cover, road density, wetland
change, ocean productivity cycles.
Instream, riparian, road, and upland
project type, number and location.
Habitat and biotic indicators of
project effectiveness.
Compliance rates and effectiveness
measures of policy guidelines and
rules (i.e. Forest Practices Act
Monitoring).
Component and cumulative analysis
of restoration actions and
management program benefits.
Does the Monitoring Program provide 7. Standardize monitoring designs,
assessment protocols, and methods to
information and analysis for adaptive
manage and analyze data.
review of restoration actions,
8. Coordinate and support interagency
management practices, and Oregon
monitoring programs and publicPlan policies?
private monitoring partnerships.
9. Is there sufficient support and
9. Integrate information from multiple
guidance for local efforts so that
sources to produce data products and
monitoring evaluates restoration
reports that assess restoration efforts
effectiveness and contributes to
and evaluate progress toward recovery
broader scale assessments?
goals.
10. Does the Oregon Plan coordinate
effectively with state, federal, and
tribal assessment and monitoring
activities?
11. What is the level of public
understanding and acceptance of and
participation in the Oregon Plan?
12. Is monitoring information used
adaptively to guide actions and to
meet Oregon Plan reporting
requirements?
13. Does the monitoring help
evaluate progress toward
environmental benchmarks and salmon
recovery goals?
Comprehensive documentation of
who is monitoring what and where,
and what methods are used.
(agencies, Tribes, watershed councils,
SWCD's, landowners, other
organizations)
Assessment of natural resource data
management throughout the Pacific
Northwest.
Whole stream or watershed surveys,
synoptic assessments of salmon
populations and water quality, and
other OWEB funded and
cooperative monitoring.
Complimentary Program Data:
NW Forest Plan Acquatic and
Riparian Monitoring
Clean Water Act - DEQ TMDL
implementation.
Ag Water Quality 1010 Plans.
150
Overview of the Aquatic and Riparian
Effectiveness Monitoring Program
February 2004
What is AREMP?
The
Aquatic
and
R i p a r i a n
Effectiveness Monitoring Program (AREMP) is a multifederal agency program developed to assess the
effectiveness of the Aquatic Conservation Strategy in
maintaining or restoring the condition of watersheds in
the Northwest Forest Plan area (Figure 1). The goals of
the program include monitoring current condition of
watersheds and changes in condition through time. If the
strategy is effective, then the condition of watersheds
across the region should either remain the same as it was
when the strategy was implemented in 1994, or it should
improve--an increasing number of watersheds should be
in good condition if the strategy remains in place.
Program Objectives
Present Emphasis:
• Assess the condition of 250 watersheds within the
Northwest Forest Plan area by collecting information
on upslope, riparian, and in-channel attributes within
each watershed.
• Develop and validate decision support models that are
used to evaluate the data collected and assess the
condition of the watersheds that have been sampled
(see pg. 3 for more information).
Future Emphasis:
• Develop predictive models to improve use of
monitoring data, potentially reducing the number of
attributes measured and long-term monitoring costs;
• Provide a framework for adaptive monitoring at the
regional scale; and
• Provide information for adaptive management by
analyzing trends in watershed condition and
identifying causes of unsuitable or unacceptable
conditions.
Figure 1. AREMP will be sampling 250-6th
field watersheds in the Forest Plan area.
This sample size represents about 10% of
the total number of watersheds in the
Forest Plan area. Watersheds are sampled
each year over a 5-year rotation. Data from
the upslope, riparian, and stream channel areas
are integrated into an assessment of
watershed condition.
151
Monitoring Questions
The questions the monitoring program is charged with answering
are related to evaluating the effectiveness of the Aquatic
Conservation Strategy in achieving it’s goal of maintaining and improving the condition of watersheds in the
Forest Plan area. These questions include:
•
•
•
•
Are the key processes that create and maintain habitat conditions in watersheds intact?
Do key indicators suggest that habitat and biotic condition have improved or degraded ?
How does the proportion of watersheds in good condition change through time?
Are current management practices attaining the Aquatic Conservation Strategy’s objectives?
The definition of watershed condition developed by
the monitoring plan was based on the goals of the
strategy and on guidance provided by Reeves et al. (2003)1. A watershed is defined as being in “good”
condition if the physical attributes are adequate to maintain or improve biological integrity, with a focus on
diversity and abundance of native or desired fish species. Specific physical attributes include intact upslope
and riparian habitats that are biologically and structurally diverse and function properly, i.e., banks that are
stable, large wood is present in the stream channel, and sediment and nutrient inputs are similar to natural
levels. Flows should be adequate to maintain or improve riparian and in-channel habitat. Chemical
characteristics and water temperature must be within a range that maintains biological integrity. Further,
the system should be capable of recovering to desired conditions when disturbed by large natural events or
by land-management activities.
Because this definition is fish-centric, it is
possible to have a watershed with intact processes
that is not in good condition from a fish perspective.
Watersheds naturally vary in their condition, and
they periodically experience natural disturbances.
For this reason, it is important to remember that
unmanaged watersheds are not necessarily in “good”
condition. Although condition may be improving in a
recently disturbed watershed, no watershed is
expected to be in good condition all the time
regardless of the management history. Further, we
do not expect all watersheds to be in good condition
at any one point in time.
Watershed Condition, Defined
S. Lanigan
Although this stream in an old-growth forest is
considered pristine, fish abundance is very low. The
stream channel is scoured down to bedrock, therefore
spawning gravel is not available. The channel also has a
high width to depth ratio and lacks habitat complexity
despite an abundance of large wood.
About the Attributes
The monitoring program uses physical and biological
attributes that act as surrogates or indicators of
watershed processes. Information on roads and
vegetation in upslope and riparian areas are obtained
using geographic information systems (GIS)
analyses. Coverages developed by CalVeg (in California) and the Interagency Vegetation Mapping Project (in
Washington and Oregon) are used to determine the percent of the upslope and riparian area with conifers
of early, mid, or late-seral stage. Information on roads and streams is used to determine road density in
1
Reeves, G.H., D.B. Hohler, D.P. Larsen, D.E. Busch, K. Kratz, K. Reynolds, K.F. Stein, T. Atzet, P. Hays, and M. Tehan. 2003. Aquatic and riparian
effectiveness monitoring plan for the Northwest Forest Plan. Gen. Tech. Rep. PNW-GTR-577. Portland, OR: U.S. Department of Agriculture,
Forest Service, Pacific Northwest Research Station.
Page 2
152
riparian,
upslope,
or hazard
areas and the number of road-stream
About the
Attributes
(con’t)
crossings in each of the watersheds. A landslide assessment will soon be
added to the evaluation.
Stream channel physical and biological attributes are measured at
multiple randomly selected locations in the watershed, covering a distance
20 x the bankfull width, with minimum and maximum reach lengths of 160
and 480 m respectively. Physical attributes that are measured include
bankfull width and depth, entrenchment ratio, gradient, sinuosity,
substrate (D50 and pool tail fines), pool frequency, and wood frequency.
Biological data are collected on fish, aquatic and terrestrial amphibians,
benthic invertebrates, and periphyton. Sampling protocols can be
downloaded from our website. A data quality assurance/ quality control
program has also been implemented, with the following: formalized field
training, remeasurement of a subset of sample sites by an independent
field crew, field audits, and crew exit surveys.
What’s a Decision Support Model?
S. Lanigan
AREMP
employees
use
electrofishers as part of the
biological sampling conducted in
each watershed.
Decision support models document decision processes and allows the
same process to be applied consistently across time and space. The
models developed by the monitoring program are used to evaluate whether a particular watershed is in good
condition. Decision support models work by evaluating individual attributes (such as road density) and
calculating an evaluation score for each attribute. The model then aggregates the evaluation scores of all
attributes into a single watershed condition score.
Using decision support models has numerous advantages because assessments are repeatable, they
can be conducted at any spatial or temporal scale, and they can handle multiple data formats. More
importantly, as our understanding of how watersheds function increases, models can be refined and rerun on
data from earlier time periods to correct deficiencies. Additional information on decision support models
and the software used to run them, can be found on our website.
Managers use our decision support models to estimate impacts of management activity on resources
and for prioritization. Assessments are transparent, explanations to stakeholders are easy and logical.
Model Construction and Refinement
The Northwest Forest Plan encompasses more than 24 million
acres of federal lands in the Pacific Northwest. Stream and
riparian habitat conditions vary greatly across the Forest Plan area
because of natural and management-related factors. For example,
precipitation ranges from several hundred inches per year near the
coast to less than 20 inches per year on the east side of the
Cascade Range. To account for the diversity within the Forest Plan
area, a decision support model is being constructed, refined, and
peer-reviewed for each of seven different physiographic provinces
(Figure 1) during workshops attended by local agency professionals.
The workshops consisted of an informal group process through
which participants came to consensus on how the model evaluated
individual attributes and aggregated the scores of individual
Page 3
S. Lanigan
(left to right) Tom Robison, Okanagan/
Wenatchee NF, Gary Ketcheson and Jim Doyle
from the Mt. Baker-Snoqualmie NF were three
of nearly 50 participants who provided
technical expertise and local knowledge for
decision support model construction and
refinement.
153
attributes. Participants examined each
attribute class (e.g., roads). For each
class, groups identified the processes
that should be accounted for in the
model, then selected the attributes that
were the best surrogates for each
process. For example, in the roads
evaluation, attributes were selected to
describe
the
hydrologic
connectivity of the road with the stream,
the potential for mass failure, and
floodplain constriction. As we discussed E. Moberly
each attribute, we developed evaluation
The decision support model integrates information collected in the
criteria and determined how the
stream, riparian, and upslope areas. With this approach, we can
attribute evaluation scores should be
take a comprehensive view of watersheds.
aggregated.
Following the workshops, models were
constructed and run, and the results are being returned to the workshop participants. Participants compare
the model results with their knowledge of the condition of watersheds and suggest refinements to the
model as necessary. The suggested changes will be made to the model and the new results will again
evaluated. Data from the 2001 and 2002 field seasons are being used during the refinement process, and
data collected during the 2003 field season will be used to validate the models.
While constructing each model, numerous decisions were made (for example, evaluation criteria values)
that were based on data, published literature, and professional judgment. As part of the quality assessment
of the model and its results, we documented the basis for each decision as well as each workshop
participant’s confidence in the decision.
AREMP Products
Everything produced by the monitoring program, including data,
models, and reports is available. A database that contains all
field and GIS data is currently maintained by the
Want to know more?
monitoring program. Data requests should be directed to
Chris Moyer.
For more information, visit our website
The decision support model and evaluation criteria will
www.reo.gov/monitoring/watershed be distributed to local units for their use when the
refinement process is complete (summer 2004). Questions
or contact one of us directly.
regarding the decision support model should be directed
to Kirsten Gallo.
• Steve Lanigan, Team Leader,
Annual progress reports and other Northwest Forest
503.808.2261, [email protected]
Plan monitoring information are located on our website.
• Peter Eldred, GIS Coordinator,
Data summary reports for each watershed sampled are
541.750.7078 [email protected]
distributed to local units each year to provide data and
other products. Forthcoming reports in 2004 include:
• Kirsten Gallo, Aquatic Ecologist,
• 10-year Interpretive Report for the Forest Plan,
• the decision support model construction and peer
review process,
• Chris Moyer, Fish Biologist,
• the monitoring program evaluation, and
• the data quality assurance/ quality control (QA) plan,
This overview was written by Kirsten Gallo.
with the results from this program.
154
Introduction to EMDS
What is EMDS?
Ecosystem Management Decision Support (EMDS) is software used to develop and run decision support models.
These models can be used to conduct objective ecological assessments by integrating diverse kinds of data, such as
in-channel habitat indicators and upslope vegetation. Key features of EMDS include transparent analysis process,
consistent interpretation of data, identifying gaps in data, and research needs. EMDS is an extension of ArcGIS (or
ArcView) and can be used to conduct assessments at any geographic scale.
How can I use EMDS?
•
Uses of EMDS include:
Assessments of current ecological condition and changes in condition
over time at any spatial or temporal scale.
•
Estimations of how a proposed management activity will impact
ecological condition, including the magnitude of impact.
•
Prioritization of restoration efforts and data collection. Assessments are
transparent, therefore explanations to stakeholders are very easy and
logical.
EMDS is available free of
charge on the web.
http://www.fsl.orst.edu/emds
How does EMDS work?
EMDS evaluates individual data then aggregates this information to make
an over all assessment of condition. Evaluation criteria are developed by
the user (or expert panels of users) to evaluate individual data parameters. Data are compared to the criteria and
given an evaluation score that ranges between (+)1 and (–)1 (Figure 1), where (+)1 indicates that the resource is in
“good” condition (technically, the data fully support the premise that the resource is in acceptable condition). A
value of (–)1 indicates that the resource is “poor” condition (i.e., the
data do not support the premise that the resource is in acceptable
condition). Evaluation scores between (-) 1 and (+) 1 reflect the
gradient that lies between “good” and “poor” condition. The EMDS
evaluation process is much like that used by the National Marine
Fisheries Service (NMFS) Matrix of Pathways and Indicators. The
difference between the two processes is that the NMFS procedure places
data into the categories of “properly functioning,” “at risk,” and “notproperly functioning,” whereas EMDS uses a continuum between
“good” and “poor” condition.
Criteria curves may be constructed in several ways, using a
maximum of four criteria. The top panel in Figure 1 shows an instance
where two criteria were used, including one criterion for “good”
condition (point b), and one for “poor” condition (point a). The lower
panel in Figure 1 shows the use of four criteria. In this instance, a range
of “good” condition values exist (between points b and c). This
type of criteria curve could be used to evaluate water temperature, for
example, where a range of values is acceptable for good fish
production.
Temperatures outside the acceptable range become
Figure 1. Evaluation curve possibilities,
increasingly detrimental to fish production.
using two or four criteria.
To assess condition, individual evaluation scores are aggregated
into a single score (range = (-)1 to (+)1) using user-defined rules. Rules
that produce a score weighted toward the resource in either the “best” or “worst” condition may be used.
Alternatively, a score can be based on the unweighted average of the indicator evaluation scores. Selection of the
rules should be based on knowledge of the system and ecological processes. The “worst-case” scenario would be
155
How does EMDS work? (continued)
used to allow a single variable to override other
variables. For example if the user is evaluating the
ability of a stream reach to support fish, water
temperature may be allowed to override other habitat
attributes because if the water is too warm, it doesn’t
matter how many pools are present. Conversely, an
average could be used if habitat variables in “good”
condition can compensate for those in “poor”
Figure 2. Example model structure. The rule (AND in this
condition. In addition to controlling how scores are
example) determines how the evaluation scores of the biological,
aggregated, the user may also weight specific indicators
physical, and chemical condition evaluation scores are
relative to their importance in the analysis.
aggregated.
In the simplified model structure shown in
Figure 2, the condition of a stream reach is determined using the average of biological, physical, and chemical conditions
in the reach that has been weighted toward the lowest score (which reflects the “poorest” condition). The factors
contributing to reach condition (biological, physical, and chemical
condition) are each determined by aggregating the evaluation scores of
attributes. For example, chemical condition may be the aggregated score
of dissolved oxygen, pH, and heavy metal concentrations in the reach.
What do the model results look like?
The model output includes a map of the analysis area that
indicates the current status of resource condition. Shown in Figure 3 is a
5th-field watershed and its associated 6th-fields. In this example, analysis
was performed at the subwatershed scale and individual 6th-fields are
colored according to condition. By running the model using data from
time 0 and time 1, we can examine changes in resource condition. For
example, one could evaluate stream survey data from 1996 and 1998 to
examine the impacts of the 1997 flood on watershed condition. Although
decision support models can not be used for predicting future trends,
EMDS does
contain tools
Figure 3. Example model output for a
f
o
r
watershed-level analysis. Shown are 6th-field
conducting
subwatersheds color coded according to
“what
if”
condition.
scenarios. For
example, one
can estimate how watershed condition will improve if 500
pieces of large wood were added to the stream.
In addition to map products, EMDS analyzes the
contribution of individual data parameters to the overall
condition score. This information can be used to identify
types of restoration projects needed within watersheds. In
Figure 4, the biological and physical components of the
Figure 4. Contribution of biological, physical, and chemical
analysis are contributing positively to the overall score, factors to reach condition scaled from (-1) to (+) 1.
whereas the chemical component is not. Restoration efforts
should therefore be directed toward the causes of chemical
impairment. A look at the attributes included in the chemical condition analysis
should help the user focus on the those attributes contributing the most to the negative
score. Because analyses are conducted in ArcGIS, we can use queries to look at
impairment across large spatial scales. We can ask questions such as “of the
watersheds in poor condition, which have road density lower than X?” This type of
analysis can assist in determining which watersheds need restoration or to examine the
Prepared by
extent of a disturbance. When it comes time to prioritize restoration, the EMDS
Kirsten Gallo, [email protected]
156
package includes an application that makes these prioritizations easy.
United States
Department
of Agriculture
Forest Service
Rocky Mountain
Research Station
General Technical
Report RMRS-GTR-162
September 2005
PACFISH/INFISH Biological
Opinion (PIBO): Effectiveness
Monitoring Program Seven-Year
Status Report 1998 Through 2004
157
Abstract ______________________________________
Henderson, Richard C.; Archer, Eric K.; Bouwes, Boyd A; Coles-Ritchie, Marc S.;
Kershner, Jeffrey L. 2005. PACFISH/INFISH Biological Opinion (PIBO): Effectiveness Monitoring Program seven-year status report 1998 through 2004. Gen. Tech.
Rep. RMRS-GTR-162. Fort Collins, CO: U.S. Department of Agriculture, Forest Service,
Rocky Mountain Research Station. 16 p.
The PACFISH/INFISH Biological Opinion (PIBO) Effectiveness Monitoring Program was
initiated in 1998 to provide a consistent framework for monitoring aquatic and riparian resources on most Forest Service and Bureau of Land Management lands within the Upper
Columbia River Basin. This 7-year status report gives our funding sources, partners, and
the public an overview of past activities, current business practices, products and publications, and future program directions. It is designed to increase accountability and summarize our accomplishments during the initial phase of the program.
Key words: PIBO, Effectiveness Monitoring, change detection, current condition, sampling protocol, budget
The Author
Richard C. Henderson is a Fisheries Biologist for the U.S. Department of Agriculture, Forest Service,
Fish and Aquatic Ecology Unit in Logan, UT. He earned his B.S. degree in fisheries biology from Colorado State University in Fort Collins in 1992. In 1998 he received his M.S. degree in fisheries biology
from Utah State University, Logan. Rick joined the Forest Service in 1998 as the Project Manager for the
PACFISH/INFISH Effectiveness Monitoring Program.
Eric K. Archer is a Fishery Biologist with the U.S. Department of Agriculture, Forest Service, in Logan,
UT. Eric received a B.S. degree in fisheries management in 1996 and an M.S. degree in aquatic ecology
in 2002 from Utah State University, Logan. Eric joined the Forest Service as a Fisheries Biologist in 1999
and has worked as a Supervisor of Field Operations and Data Analyst for the PACFISH/INFISH Effectiveness Monitoring Program.
Boyd A. Bouwes is a Biogeographer for the U.S. Department of Agriculture, Forest Service, in Logan,
UT. He received a B.A. degree in geography from the University of Wisconsin Madison in 1995 and an
M.S. degree in biogeography from Portland State University in 2000. Boyd joined the Forest Service in
1999 and has worked as the Assistant Supervisor of Field Operations and GIS Specialist for the PACFISH/INFISH Effectiveness Monitoring Program.
Marc C. Coles-Ritchie is a Riparian Vegetation Ecologist for the U.S. Department of Agriculture, Forest
Service, in Logan, UT. He received a B.A. degree in comparative literature from the University of Massachusetts and an M.S. degree in environmental studies from Bard College. Marc completed a Ph.D.
degree in ecology from Utah State University in 2005. Marc became the Riparian Ecologist for the PACFISH/INFISH Effectiveness Monitoring Program in 2004.
Jeffrey L. Kershner is a Washington Office Aquatic Ecologist with the U.S. Department of Agriculture,
Forest Service, Fish and Aquatic Ecology Unit, in Logan UT. He completed a B.S. degree in 1976 at
Humbolt State University, Humboldt, CA, in fisheries. He earned an M.S. degree in natural resources
from Humboldt State University in 1982 and a Ph.D. degree in ecology from the University of California,
Davis, in 1991.
The use of trade or firm names in this publication is for reader information and does not
imply endorsement by the U.S. Department of Agriculture of any product or service
158
Contents _____________________________
Page
Commonly Asked Questions ....................................................... 1
Introduction .................................................................................. 3
Sampling Summary ..................................................................... 4
PIBO Program Resources ........................................................... 5
Analysis and Results ................................................................... 6
Data Management and Distribution ........................................... 11
Interaction with Forest Service and BLM Staff .......................... 12
Interaction with Other Monitoring Programs .............................. 13
Future Direction and Study Questions ...................................... 14
Point of Contact ......................................................................... 15
Acknowledgments ..................................................................... 15
PIBO Publications and References Cited .................................. 16
Other References ...................................................................... 16
You may order additional copies of this publication by sending your
mailing information in label form through one of the following media.
Please specify the publication title and number.
Telephone
(970) 498-1392
FAX
(970) 498-1396
E-mail
Web site
Mailing Address
[email protected]
http://www.fs.fed.us/rm
Publications Distribution
Rocky Mountain Research Station
240 West Prospect Road
159
PACFISH/INFISH Biological
Opinion (PIBO): Effectiveness
Monitoring Program Seven-Year
Status Report 1998 Through 2004
Richard C. Henderson
Eric K. Archer
Boyd A. Bouwes
Marc C. Coles-Ritchie
Jeffrey L. Kershner
Commonly Asked Questions _______________________
1. What is PIBO? PIBO stands for the PACFISH/INFISH Biological Opinion
Effectiveness Monitoring Program. The program was initiated to evaluate the
effect of land management activities on aquatic and riparian communities at
multiple scales and to determine whether PACFISH/INFISH management practices are effective in maintaining or improving the structure and function of
riparian and aquatic conditions. Our study area includes 20 USDA National
Forests and nine USDI Bureau of Land Management (BLM) Field Units within
the Interior Columbia River Basin.
2. Who is PIBO? We have seven full time Forest Service employees with
backgrounds in fisheries, riparian ecology, geography, data management,
and support services. We also employ up to 45 seasonal technicians. We
are based in Logan, UT.
3. How is the PIBO program funded? The program is funded by Forest Service
Regions 1, 4, and 6 and the Oregon/Washington and Idaho State Offices of the
BLM. Funding for special projects has been provided by the Forest Service’s
Fish and Aquatic Ecology Unit and Stream Systems Technology Center, Forest
Service Region 1, Oregon/Washington BLM, and the Salmon Challis National
Forest.
4. When will we begin reporting changes and trends in resource conditions?
In 2006 we will begin our second sampling rotation, when we resample sites
initially sampled in 2001. Comparing these two sampling periods will allow us
to begin describing changes in each attribute we measure. By 2010, approximately 1,000 stream reaches will have been resampled, giving us the ability to
describe change between the periods of 2001 to 2005 and 2006 to 2010. Assessments of trends in resource conditions will begin in 2011, because a
minimum of three samples are required for trend analyses.
5. How is the PIBO-EMP working with other large-scale monitoring programs to develop compatible sampling methods and sampling design?
We have worked extensively with the Northwest Forest Plan – Aquatic and
Riparian Effectiveness Monitoring Program (AREMP) to standardize sampling methods between our programs. Beginning in 2004 we will be using
identical methods for a core set of physical habitat attributes and macroinvertebrate sampling. In 2005, we will be participating in a sampling
protocol comparison study with other Federal and State monitoring programs.
USDA Forest Service Gen. Tech. Rep. RMRS-GTR-162. 2005
160
1
6. Is the PIBO program consistent with recommendations in the Forest Service Aquatic and Ecological Unit Inventory (AEUI) proposal? Yes. The
2004 standardized sampling protocol developed with the AREMP program is
consistent with the draft AEUI protocol for most attributes.
7. Can the results be used to answer status and trend questions at smaller
spatial scales (for example, individual Forests or BLM units)? Results from
sample size analyses suggest that we will be able to detect changes in resource
condition at the scale of individual Forests and BLM Field Offices (35 to 90
sites) for many of the attributes we measure. Comparisons between reference
and managed sites also support these results. We are beginning to work with
Forest Planning teams to determine how our study design and data collection
can be used to address their monitoring questions.
8. What have we learned about the current condition of aquatic and riparian
resources? Analyses have focused on whether our sampling methods can detect differences in the resource condition between managed and reference sites.
The assumption is that if the methods can detect differences, then they will be
useful in detecting changes from current management practices. We found significant differences for eight of 12 physical habitat attributes and for
macroinvertebrates, but no difference in riparian vegetation attributes. Analysis
of PACFISH/INFISH Riparian Management Objectives suggest that several of
the standards are unrealistic, especially wetted width:depth ratio and percent
undercut banks. Stratification of the data by Regions or State Offices suggests
that standards need to reflect local environmental factors.
9. How can the data, summary results, reports, and publications be accessed?
Information about the project can be found on our Web page or by contacting us
directly. Our Web site www.fs.fed.us/biology/fishecology/emp.html includes an
overview of the program, sampling protocols, publications, and an employment
page. We have also developed a second site where original and summarized data
can be accessed for all stream reaches we sample. If you have further questions,
please contact us at [email protected].
10. Will PIBO become obsolete when Land Management Plans replace
PACFISH and INFISH with new aquatic conservation strategies? No.
We use a probabilistic sampling design to nearly randomly choose sample
sites. This approach is appropriate for assessing the effectiveness of aquatic
conservation strategies (ACS). As question 7 above describes, our sample
sizes should be adequate to assess the effectiveness of new ACS’s as Land
Management Plans are revised.
2
1612005
USDA Forest Service Gen. Tech. Rep. RMRS-GTR-162.
Introduction ____________________
The PACFISH/INFISH Biological Opinion (PIBO)
Effectiveness Monitoring Program was developed in
response to monitoring needs addressed in the Biological Opinions for bull trout (U.S. Department of the
Interior, Fish and Wildlife Service 1998) and steelhead
(U.S. Department of Commerce, National Marine
Fisheries Service 1998). It provides a consistent framework for monitoring aquatic and riparian resources
within the range of the Pacific Anadromous Fish
Strategy (PACFISH) and the Inland Fish Strategy
(INFISH), and will determine whether land management practices are maintaining or improving riparian
and aquatic conditions at both the landscape and
watershed scales on Federal lands throughout the
Upper Columbia River Basin.
The program began with a pilot study in 1998 on
Forest Service lands within the Salmon River Basin
of central Idaho. In 2000, the Interagency Implementation Team (IIT) expanded the pilot study to include
Federal lands within the Interior Columbia River Basin.
This includes Forest Service lands within PACFISH and
INFISH (20 National Forests) and BLM lands that are
within PACFISH or contain bull trout (10 Field Offices
and Resource Areas). (Throughout this document we
use the term “Field Units” to include Forest Service’s
Forests and Ranger Districts, and BLM Districts, Field
Offices, and Resource Areas.) During the pilot study
we focused on evaluating sample methods, addressing
study design questions, and developing a centralized
team to implement the program. The study design was
finalized in the winter of 2000 (Kershner and others
2004b), and the PIBO Effectiveness Monitoring Program officially began in 2001.
This 7-year status report will give our funding
sources, partners, and the public an overview of past
activities, current business practices, products and
publications, and future program directions. It is
designed to increase accountability and summarize
our accomplishments during the initial phase of the
program.
Objectives and Study Design
The program goal is to determine whether
PACFISH/INFISH management practices are effective in maintaining or restoring the structure and
function of riparian and aquatic systems. The specific
objectives are:
1. Determine whether a suite of biological and physical attributes, processes, and functions of upland,
riparian, and aquatic systems are being degraded,
maintained, or restored across the PIBO landscape.
2. Determine the direction and rate of change in
riparian and aquatic habitats over time as a function of management practices.
3. Determine if specific “Designated Monitoring
Area (DMA)” practices related to livestock
grazing are maintaining or restoring riparian
vegetation structure and function.
The study area contains 3,547 subwatersheds
(sample units) with at least some Forest Service or
BLM ownership. A generalized random tessellation
stratified design (GRTS) was used to select subwatersheds to achieve a random, nearly regular sample pattern
throughout the study area (Kershner and others 2004a).
Approximately1,300 subwatersheds were selected for
sampling during the first 5 years (2001 through 2005).
Each of these subwatersheds will then be resampled
on a 5-year rotation beginning in 2006 (table 1).
The subwatersheds were divided into two groups
based on management history. A subwatershed was
considered “reference” if it was not grazed by livestock in the last 30 years, road densities were less than
Table 1—The table displays the sampling design at full implementation where 250 subwatersheds would be
sampled each year from 2001 through 2007. These sites will be resampled every 5 years. An additional
50 watersheds were selected for annual sampling (sentinel sites). The actual number of subwatersheds
sampled at half implementation in 2001 and 2002, and full implementation in 2003 and 2004, are shown
in parentheses.
Sampling
design category
Sentinel
Group 1
Group 2
Group 3
Group 4
Group 5
2001
2002
2003
2004
2005
2006
2007
50(38)
250(152)
50(26)
50(48)
50(50)
50
50
250
50
250(106)
250
250(233)
250(241)
250
162
3
0.5 km per km2, riparian road densities were less than
2
0.25 km per km , and there was no historic dredge or
hardrock mining in riparian areas. All other subwatersheds were considered “managed.”
Sample Site Selection
To address our first objective, we sampled an “integrator” reach in each randomly chosen subwatershed.
These sites were chosen because they are the most
likely location to show integrated effects from upstream
management actions, and 83 percent of these sites are
located in the most downstream response reach (defined as having a stream gradient less than 3 percent)
on Federal land with the remaining sites (17 percent)
at the downstream most transport reach (stream gradient between 3 and 5 percent) on Federal lands. The
design also requires at least 50 percent Federal ownership upstream of the site. A suite of physical stream
habitat attributes, riparian vegetation characteristics,
and macroinvertebrate samples are collected at these
sites.
To address our second objective, we randomly
selected 25 reference and 25 managed integrator sites
for annual sampling (sentinel sites).
Our third objective required us to sample a DMA
within each selected subwatershed where cattle grazing
occurs within the riparian area. The location of the
DMA is determined by the Field Units and is used for
annual “Implementation” monitoring (USDA 2003).
This link between implementation and effectiveness
monitoring provides an adaptive management feedback
process. Only riparian vegetation, stream bank, and
bankfull width measurements are collected at these
sites.
Field Sampling Protocols
In 1997 an interagency team was convened to determine which physical and biological attributes should
be measured to answer the program objectives
(Kershner and others 2004b). The original sampling
methods we used to measure the attributes came from
a variety of sources. Since 1998 we have continued to
evaluate and refine each of the methods based on
feedback from field crews and results from quality
assurance tests.
The stream habitat protocol includes methods for
assessing channel cross-sections, gradient, habitat
units, large wood, sinuosity, streambed substrate,
streambank parameters, water temperature, and aquatic
macroinvertebrates (Dugaw and others 2004). The
riparian vegetation protocol describes methods for
sampling the species composition along the greenline
and across the riparian area (Coles-Ritchie and others
2004a,b). Both protocols for the 2004 field season can
be downloaded from our Web site at www.fs.fed.us/
biology/fishecology/emp.
Sampling Summary ______________
The PIBO program has sampled within all 20 National Forests, three Resources Areas within Oregon/
Washington BLM, and five of six Field Offices within
Idaho BLM. During the pilot years from 1998 to 2000,
196 subwatersheds were sampled. Since the start of
the first 5-year rotation in 2001, 783 subwatersheds
have been sampled (table 2, fig. 1). Additional sites
were sampled for protocol tests, annual quality control
assessments, and for a variety of special projects. This
information is summarized annually in reports tailored
for each Forest and BLM Field Unit.
Table 2—Summary of all reaches sampled from 2001 through 2004. The sum
of the number of subwatersheds is greater than the total due to multiple ownership in several subwatersheds. Similarly, the number of
reaches is greater than the number of subwatersheds due to both
integrator and DMA reaches within some subwatersheds.
Reaches
location
Region 1
Region 4
Region 6
BLM Idaho
BLM OR/WA
Total
4
Subwatersheds
288
232
184
50
39
783
Integrator reaches
Managed
Reference
211
169
162
26
9
577
77
59
14
150
DMA’s
17
53
80
34
28
212
1632005
USDA Forest Service Gen. Tech. Rep. RMRS-GTR-162.
Figure 1
Figure 1—The figures show the study area and the 783 subwatersheds sampled between 2001 and 2004.
Detecting Change in Resource Condition
PIBO Program Resources ________
We will begin analyzing the dataset for changes in
the conditions of aquatic and riparian resources (Objective 1) in 2006. This is the first year of the second
rotation, when we begin resampling stream reaches
initially sampled in 2001. We will summarize the
change for nearly 250 sites each year until 2010. By
this time about 1,000 stream reaches will have been
resampled, giving us the ability to describe changes
between the periods of 2001 to 2005 and 2006 to 2010.
Assessments of trends in resource conditions will begin in 2011, since a minimum of three samples are
required for trend analyses.
Funding for the program has steadily increased since
1998. In fiscal year 1998, we received $70,000 from
Region 4 to initiate the program. The annual budget
increased to $1,298,000 in fiscal year 2004, with the
majority of funding coming from the three Forest Service Regions and two BLM State Offices (table 3).
Additional funding has been provided through partnerships with the Forest Service Fish and Aquatic Ecology
Unit, Forest Service Stream Systems Technology Center, Utah State University, BLM Washington Office,
and individual Forests. Funding from partners has
ranged from $0 to $75,000 annually, accounting for
up to 6 percent of our budget.
164
5