CHANGED AND UNFORSEEN CIRCUMSTANCES 7.1 CHANGED …

Skookumchuck Wind Energy Project Habitat Conservation Plan Lewis and Thurston Counties, Washington

Chambers Group, Inc. / 20898 93 WEST, Inc.

CHANGED AND UNFORSEEN CIRCUMSTANCES

7.1 CHANGED CIRCUMSTANCES

The No Surprises Rule defines “changed circumstances” as “circumstances affecting a species or geographic area covered by a conservation plan that can reasonably be anticipated by plan developers and the USFWS and that can be planned for (e.g., the listing of new species or a fire or other natural catastrophic event in areas prone to such events)” (50 CFR §§ 17.3).

The HCP must identify provisions to compensate for negative impacts to covered species from changed circumstances in order to qualify for No Surprises Rule assurances. If circumstances change, the permittee must implement any provisions included in the HCP and/or ITP that address such circumstances.

The following sections describe the changed circumstances that the Applicant and the USFWS can reasonably anticipate and for which responses can be planned. The responses provided for each changed circumstance represent an opportunity for the Applicant and the USFWS to reevaluate the effectiveness of the conservation program and adjust priorities, reallocate resources, or otherwise modify how the HCP is implemented.

7.1.1 Changed Circumstances

If a changed circumstance occurs, the Applicant will notify the USFWS Lacey Ecological Services Field Office and the USFWS Region 1 Migratory Birds Division in writing via email or letter within two weeks of the event.

7.1.2 Monitoring Is Inadequate to Detect Murrelet Carcasses

If, after a three-year Evaluation Monitoring period, no murrelet carcasses have been found during standardized searches but the monitoring results produce a fatality estimate exceeding an adaptive management threshold (2.496 murrelets per year), the situation will indicate that the monitoring as designed is inadequate to inform adaptive management decisions and enable evaluation of compliance with the ITP. This situation would result from one or more of the parameters of murrelet detection probability (i.e., carcass distribution, searcher efficiency, carcass persistence) that is substantially different than considered in this HCP and would require the Applicant to consider alternative compliance monitoring strategies for murrelets. The monitoring program presented in Appendix G has been designed to avoid triggering this changed circumstance.

Trigger: After the three-year Evaluation Monitoring period, zero murrelet carcasses have been found during standardized searches and the cumulative murrelet fatality estimate exceeds the requested take of 7 murrelets (2.496 × 3 = 7.488 rounded down).

Response: The Applicant will report the monitoring results to the USFWS in the annual compliance monitoring report and notify the USFWS that a changed circumstance has occurred. The Applicant will then review the current state of the science on avian mortality monitoring at wind energy facilities and develop an alternative compliance monitoring plan for murrelets. The alternative monitoring plan will use the best available scientific information to improve murrelet detection probability to a level sufficient to evaluate compliance with the ITP, or to the extent practicable. The Applicant will provide the alternative compliance monitoring plan to the USFWS for review prior to implementation during the next ITP year.



7.1.3 Use of Technological Advances

Over the course of the ITP term, new information on murrelets, eagles, and wind power interactions may become available; new methods for monitoring mortality may be developed; or technological advances may be developed to minimize murrelet or eagle mortality at wind turbines. The Applicant may wish to apply one or more of these new developments into the operations and/or monitoring plan outlined in the HCP.

For example, new information, methods, or procedures for monitoring or operation may become available during the course of the ITP term. It is expected that over time, results of post-construction mortality monitoring, findings from research conducted as part of the conservation measures implemented under this HCP, and results from research and evaluation related to the wind industry made elsewhere will be used to inform changes to the operation and monitoring plans. Non-Project-related advancements in scientific understanding of murrelet biology and behavior may be used to inform changes to the avoidance, minimization, mitigation, and conservation measures.

The Applicant may choose to utilize any alternative monitoring methods should they be demonstrated, based on the best available science to be as or more effective than the methods described in this HCP, as approved by USFWS. Similarly, other technological advances or new techniques and information may become available during the course of the ITP term that the Applicant may want to use to more effectively implement other areas of the HCP, such as adaptive management. The Applicant will work with USFWS to ensure that any new information or techniques that are planned to be used are compatible with the biological goals and objectives of the HCP.

Any new method, information, or technology will be considered only if it has been demonstrated in an acceptable scientific study, has been approved by the USFWS as the best available science, and will not require an increase in the take authorization for the Project or require additional funding.

7.1.4 New Species Listing

Upon occasion, the USFWS adds species to the federal list of threatened and endangered species or creates or modifies designations of critical habitat. It is possible that the USFWS could list a species that occurs within the Plan Area or designate critical habitat within the Plan Area after ITP issuance. USFWS publication of a Final Rule in the Federal Register will signify the occurrence of this changed circumstance. The Applicant has the ability to discuss this circumstance with the USFWS at any time prior to the publication of a Final Rule, including upon publication of a proposed rule. The Applicant will evaluate if the Project, with consideration of the avoidance and minimization measures already included in the HCP’s conservation program, is sufficient to avoid take of the newly listed species or adverse modification of critical habitat. If the Applicant determines that incidental take of the newly listed species is reasonably certain to occur, the Applicant will coordinate with the USFWS to obtain authorization for incidental take of the newly listed species, either by amending this HCP or by preparing a new HCP, and will implement operational measures necessary to avoid take of the newly listed species unless and until authorization is obtained.

7.1.5 Climate Change

Global climate change is the observed increase in mean global temperature due to an increase in greenhouse gas emissions, primarily carbon dioxide, as a result of human industrialization (IPCC 2007). According to IPCC (2007), the earth’s climate has warmed between 0.61 and 0.89 °C (1.1 and 1.6 °F) over the past century. Over the past 30 years, temperatures have risen more rapidly in winter than in any other



season. Some of the changes have been faster than previous assessments had suggested (Melillo et al. 2014). Additionally, climate change has been studied for the impacts on wind speed variability. By the end of the twenty-first century, it is likely that more frequent and higher intensity wind events will occur (Pryor and Barthelmie 2010).

Global climate change is also predicted to include secondary global effects, such as sea-level rise and changing weather patterns; rainfall patterns, storm severity, snow and ice cover, and sea level appear to be changing already (USEPA 2016b). As a consequence of these changes, flood, drought, and fire regimes are likely to be altered in frequency and intensity in the future. Many climate change-related impacts to wildlife are likely to manifest through species’ life history changes. Species with highly specialized habitat needs, with narrow environmental tolerances, that are dependent on specific climatic triggers or cues, and with limited or no ability to disperse and/or colonize new or more suitable areas are likely to be the most susceptible to climate change-related impacts.

Murrelets and, to a lesser extent, eagles have many of these life-history attributes and are, therefore, likely to be affected by climate change. However, very limited information currently exists to assess potential impacts of climate change on murrelets or eagles. Although the manifestations of climate change are expected to be complex and widely varied, several potential negative impacts to murrelets and eagles may occur.

The Applicant has agreed to operational restrictions during times that coincide with nesting season of murrelets (i.e., June 1 to August 9; see Chapter 6). If the timing of these activities or flight patterns shift, minimization methods included in this HCP could be less effective at minimizing take of murrelets. Within six months of publication, the Applicant will review all newly released material by the USFWS on the status of murrelets, including five-year status reviews and revisions to recovery plans, to determine if an observed change to any murrelet life-history periods has occurred that may necessitate the need to alter the timing of operational restrictions if there is not an increase in duration of curtailment. Since shifts in timing of murrelet dispersal and migration would likely be apparent in the timing of murrelet fatalities at wind facilities, the Permittees will also use data from post-construction monitoring studies at available wind facilities to document potential shifts in the timing of murrelet fatalities from current conditions. Responses to shifts from climate change will be addressed through the murrelet adaptive management strategy (Section 6.4.1).

7.1.6 Repowering

Prior to the end of the ITP term, the Applicant may decide to repower the Project, in which case the Applicant will confer with USFWS regarding the potential impacts of the proposed repowering project. Repowering can take on varying forms, including increasing the size of the wind turbine. If Applicant decides to repower the Project in a way that may impact the take estimate for the Covered Species, the Applicant will work with USWFS to estimate the impact of the repowered project and compare that change to the remaining take authorized by the ITP. If the adjusted take estimates would put the Applicant at risk of exceeding its take limits, the Applicant will amend the HCP to account for potential incidental take, mitigation, and monitoring requirements associated with repowering of the Project.

7.1.7 Wind

As stated above, climate change is likely to result in more intense and more frequent high wind events. Brief but violent windstorms sometimes wrack western Washington, including the forests of southwest



Washington. The storms typically start as tropical typhoons and are carried to land by the jet stream. The historical record shows a small number of hurricane-force storms hit the Washington coast in the last 200 years. Some of the larger storms have blown down a substantial amount of standing timber. However, the most severe damage (storms in 1921, 1962) occurred before the advent of modern forest management practices, including maintenance of riparian buffers. The region does routinely experience locally strong winds, which can uproot trees. However, in recent decades severe disturbance and windthrow have been localized.

Windthrow can be expected to affect the acquired mitigation parcels to some degree over the life of the HCP. The reasonably foreseeable effect of such windthrow will be limited to individual trees or small clumps of trees. Windthrow is normal across the managed landscape and an expected part of the forest ecology. Small-scale windthrow is not expected to have a long-term significant adverse impact on suitable murrelet nesting habitat. Any potential impacts on suitable nesting habitat also will be minimized mature and maturing buffer areas, which are more likely to absorb any windthrow impacts.

It is reasonably anticipated that a rare, catastrophic wind event could knock down several acres of suitable nesting habitat, particularly if the timber is in an isolated stand or has hard edges on multiple sides. Some portions of suitable nesting habitat in the mitigation parcels are adjacent to recent harvest units and have hard edges on one side or, in small areas, more than one side. Due to the inclusion of maturing buffer areas in the mitigation parcels, the potential for windthrow damage to these areas of suitable nesting habitat will diminish during the term of the HCP as adjacent former harvest units mature and buffer the more mature suitable nesting habitat.

If a catastrophic windstorm results in the complete blow-down of more than 44 acres of suitable nesting habitat within a mitigation parcel, the Applicant will provide information regarding such windthrow to the USFWS within 30 days of discovery. The Applicant will confer with the USFWS to establish appropriate supplemental or changed prescriptions for management of the affected area. If the USFWS and WDFW conclude that the affected area is no longer suitable murrelet nesting habitat, then the USFWS and WDFW will work with the Applicant to identify the portions that may be salvage harvested in accordance Washington’s Forest Practices Act. The Applicant may sell the downed timber and sell or exchange the affected area and return it to active timber management. The revenue from the timber and land, together with funds as needed from the Changed Circumstances Fund, will be used to acquire a replacement mitigation parcel of similar characteristics to the portion impacted by the catastrophic windstorm.

7.1.8 Fire

Although it is possible to generalize that fire is an important element of forest ecology, it is not possible to specify the temporal or spatial effects of fire for any given area. Fires are by their nature unpredictable as to their timing, intensity, scale, and the area affected. A few periods of extensive forest burning – more than a million acres – have occurred in western Washington in the last 700 years. Many smaller fires have occurred in western Washington since western settlement; but none have been of the magnitude represented by these large fires, which appear to have been driven primarily by climatic variation. Of these smaller fires, those in excess of 10,000 acres also are rare and have occurred in the dryer portions of the region. Most fires have been less than 5,000 acres.

In light of the region’s fire history, it is not reasonably foreseeable that large-scale, stand-replacing fires will occur on mitigation lands during the life of this HCP. Thus, it is unnecessary to provide for new, different, or additional mitigation or conservation measures based on any speculation that such events



could occur, as these events qualify as unforeseen circumstances. Certain supplemental procedural prescriptions, however, will be applicable in the event of smaller fires.

Any forest fire that may occur during the term of the HCP on lands leased from Weyerhaeuser for the Project will be responded to and suppressed in accordance with strategies established by the Washington Department of Natural Resources and Weyerhaeuser. Fires on those lands, while they could affect operation of the Project, would not affect murrelet conservation, as the lands in the vicinity of the Project do not contain suitable murrelet habitat. For mitigation lands, fire response and suppression on mitigation lands will be the responsibility of the entity contracted to manage those lands. All measures reasonably necessary to extinguish a fire, including measures that deviate from the management prescriptions for the mitigation lands, will be implemented, again, consistent with the strategy for responding and suppressing forest fires established by the Washington Department of Natural Resources. The Applicant and the mitigation land managers may have little ability to influence fire suppression strategy. However, to the extent reasonably possible and where consistent with the primary goal of containing and extinguishing the fire, they will encourage the development of a fire-response strategy that is consistent with management prescriptions for the mitigation lands and that furthers rather than diminishes the functions that such prescriptions have been designed to provide.

If a fire involves more than 50 acres of mitigation lands, Applicant will provide the USFWS with information regarding the fire within 30 days. Once such a fire is extinguished and unless such fire is an “unforeseen circumstance” as described above, Applicant and the USFWS will confer to establish appropriate supplemental or changed prescriptions for management of the mitigation lands. These additional or changed prescriptions will be established consistent with preserving and developing habitat legacies created by the fire (e.g., upland snags) that are consistent with development of suitable murrelet nesting habitat. Reforestation consistent with these objectives also will be implemented. If the USFWS and WDFW conclude that the affected area is no longer suitable murrelet nesting habitat, then the USFWS and WDFW will work with the Applicant to identify the portions that may be salvage harvested in accordance with Washington’s Forest Practices Act. The Applicant may sell the downed timber and sell or exchange the affected area and return it to active timber management. The revenue from the timber and land, together with funds as needed from the Changed Circumstances Fund, will be used to acquire a replacement mitigation parcel of similar characteristics to the portion impacted by the catastrophic windstorm.

7.2 UNFORESEEN CIRCUMSTANCES

“Unforeseen circumstances” are changes in circumstances affecting a species or geographic area covered by an HCP that could not reasonably have been anticipated by plan developers and the USFWS at the time of the conservation plan’s negotiation and development and that result in a substantial and adverse change in the status of any covered species. The USFWS will have the burden of demonstrating that unforeseen circumstances exist and must base the determination on the best scientific and commercial data available. The USFWS shall notify the Applicant in writing of any unforeseen circumstances the USFWS believes to exist.

The No Surprises Rule states that the USFWS may require additional conservation measures of an incidental take permittee as a result of unforeseen circumstances “only if such measures are limited to modifications within conserved habitat areas, if any, or to the conservation plan’s operating conservation program for the affected species, and maintain the original terms of the conservation plan to the maximum extent possible.” The USFWS shall not require the commitment of additional land, water, or financial resources by the permittee without the consent of the permittee or impose additional



restrictions on the use of land, water, or other natural resource otherwise available for use by the permittee under the original terms of the ITP. No Surprises assurances apply only to the species adequately covered by the HCP (i.e., the murrelet) and only to those permittees who are in full compliance with the terms of their plan, permit, and other supporting documents, as applicable.



FUNDING ASSURANCES

The Applicant estimated the costs and guarantees the funding for the implementation of the HCP, which includes operation and maintenance costs, mitigation and monitoring costs, deterrent technology maintenance costs, inflation, adaptive management, and contingency costs. The Applicant will provide all funds necessary for the mitigation and continued monitoring for 30 years at a level consistent with the budget shown in Table 35.

The primary minimization method for the murrelet is the implementation of curtailment in the manner set forth in Section 6.1.2. Curtailment will result in revenues lost but does not require funding costs. The cost of the Identiflight® units is $150,000. Pursuant to the contract between the Applicant and Identiflight®, this price includes ongoing maintenance of the two units for the duration of the ITP term. These units will be purchased prior to commencement of Covered Activities ahead of the impacts; therefore, financial assurances will not be needed for the initial two Identiflight® units. The price of the murrelet mitigation lands is the result of an agreement between the Applicant and the [mitigation entity]. The dollar amount includes land acquisition, mitigation land management, effectiveness monitoring, and annual reporting. The mitigation lands will be acquired (including funding for management, monitoring, and reporting) prior to the commencement of Covered Activities. The price of the net removal program is an estimate provided by a third-party administrator of a net removal program and includes the cost to identify and remove newly lost nets. The net removal funding will be provided to the third-party administrator in advance of the Covered Activities. The Applicant will pay for the cost of power pole retrofits for the first five-years of the ITP term prior to the start of the Covered Activities. The amount below reflects the $4,000 per pole estimate included in this HCP, but the actual amount provided will be adjusted for the actual price of the power pole retrofit. Consistent with BGEPA regulations, the amount of mitigation needed for eagle offset will be reevaluated every five-years. The Applicant will pay for additional power pole retrofits for the next five-year increment, if needed, prior to the end of the fifth year. Because retrofit payments will be completed ahead of any impacts to golden eagles, no additional financial assurances are necessary. Monitoring costs have been determined based on estimates provided by qualified consultants. The cost of Evaluation, Implementation, and Re-Evaluation Monitoring (if needed) will be provided for in the annual operational and maintenance budget. The annual report will include a statement and signature from an authorized-corporate representative that the budget includes an amount sufficient to cover the cost of monitoring for the next monitoring year. The Applicant will maintain a letter of credit for the cost of one year of Evaluation Monitoring throughout the ITP term.

The Applicant intends that any costs incurred due to Changed Circumstances (Section 7.1) or the Adaptive Management responses (Sections 6.4.1 or 6.4.2) will be paid out of the Project O&M budget. However, to assure that funding is available should Changed Circumstances (Section 7.1) or the Adaptive Management responses (Sections 6.4.1 or 6.4.2) be triggered, the Applicant will provide a Letter of Credit from a Federal Deposit Insurance Corporation (FDIC)-insured lender or a Corporate Guarantee and renewed through the ITP term. The amount of the letter of credit or Corporate Guarantee will be equal to the cost of an additional year of Evaluation Monitoring (as described above), an additional Identiflight® unit, and the cost to replace 44 acres of murrelet mitigation land. Documentation of payments and confirmation of letter of credit renewals will be included in the annual reports submitted to USFWS.



Table 35: Funding Assurances

HCP Action Estimated Cost Cost Basis and Assumptions Financial Assurance

Minimization a. murrelet

curtailment No additional cost Section 6.1.2. – minimization

measures will not cost money to implement, but will result in lost revenues

N/A

b. eagle curtailment – Identiflight® units

$300,000 Section 6.2.2. – cost estimate based on two units, and purchase includes ongoing maintenance of units. Costs associated with curtailment implementation will not require financial assurances as that cost is through revenues lost from power production.

Will be paid prior to start of Covered Activities

Mitigation a. murrelet lands $3,000,000 Section 6.1.3 – cost estimate based

on mitigation proposal for __ acres, including acquisition, management, monitoring, and adaptive management

Will be paid prior to start of Covered Activities

b. net removal $450,000 Section 6.1.3.4. – cost estimate is based on an estimate to fund net removal program.

Will be paid prior to the start of the Covered Activities

c. eagle power pole retrofits

$468,000 Section 6.2.3 – cost estimate based on an estimate of $4,000 per pole applied to the number of poles determined by the USFWS’s REA for golden eagles (117 poles). It is anticipated that the actual cost per pole will be less than $4,000.

Will be paid prior to start of Covered Activities every five years (as needed)

Monitoring a. Evaluation

monitoring $248,507 per year $625,209

Permit Term Total

Section 6.3.2 - Will be included in each year’s O&M budget for the first three years of operation with 1.7% inflation.

Corporate Certification of O&M budget and Letter of Credit

b. Implementation monitoring

$226,708 per year $1,107,822 Permit Term Total

Section 6.3.2 – Includes training and will continue for life of the Project. Will be included in each year’s O&M budget following the Evaluation budget with 1.7% inflation for years 4 to 30 of the Project.

Corporate Certification of O&M budget and Letter of Credit (same LOC maintained for Evaluation will apply for Implementation)



HCP Action Estimated Cost Cost Basis and Assumptions Financial Assurance

c. Re-evaluation monitoring

$215,529 Section 6.3.2 – if Re-evaluation monitoring is triggered, this amount will be included in the following year’s O&M budget with 1.7% inflation based on year of Project.

Corporate Certification of O&M budget and Letter of Credit (same LOC maintained for Re- Evaluation will apply for Implementation)

d. Effectiveness monitoring

No Additional Cost Section 6.3.5 Costs of effectiveness monitoring are included in the price of murrelet lands and power pole retrofits.

Included in payments for mitigation

Changed Circumstances Fund

$______ Sections 6.4.1, 6.4.2, and 7.1 — Potential additional costs may be additional deterrent technology

Letter of Credit

Total HCP Costs7

The Applicant will incur most of these costs prior to the start of Covered Activities and any incidental take. For example, the costs associated with Sections 6.1.2 and 6.1.3 will be set aside in an endowment fund. Because the Applicant commits to funding these measures prior to conducting Covered Activities or incurring any incidental take, no additional financial assurances are required. Only the costs associated with additional power pole retrofits, future monitoring, and changed circumstances will occur after Covered Activities have begun and renewed through the ITP term.

7 Not including revenues lost to curtailment.



AMENDMENTS

It may be necessary at some time over the duration of the proposed permit for the USFWS and the Applicant to clarify provisions of the HCP or the requested ITP with respect to program implementation or the meaning and intent of language contained in these documents. Such clarifications should not change the substantive provisions of any of the documents in any way but merely clarify and make more precise the existing provisions.

In addition, it may be necessary to make administrative changes or minor modifications to the documents at some time over the duration of the proposed permit. Such changes should not result in substantive changes to any provisions of the documents but may be necessary or convenient to represent the overall intent of the Applicant and the USFWS. Examples of such administrative changes or minor modifications include correction of typographic errors in the documents, changes in the legal business name or mailing address of a permittee, or clarification of reporting procedures. Requests for administrative changes and minor modifications must be received in writing and may be reviewed and approved by the USFWS Regional Office or by the appropriate USFWS Ecological Services Field Office in accordance with applicable regulations and policies (50 CFR 13).

Except as provided for above, the HCP and the ITP may not be amended or modified in any way without the written approval of the Applicant and the USFWS. The USFWS will not need to publish an HCP or ITP amendment when levels of incidental take do not increase and the activity does not expand in ways not analyzed in the original NEPA or Section 7 documents. In limited instances, amendments to the HCP or the ITP may require publication in the Federal Register. These instances would include changes in Permit Area, Proposed Action, Covered Activity, type or amount of take, additions to Covered Species, significant changes to the minimization, or mitigation measures under the HCP, including funding, that may affect the type or amount of take, the effects of the Covered Activities, or the nature or scope of the minimization or mitigation measures in a manner or to an extent not previously considered (through adaptive management of changed circumstances) in the HCP. Such amendments will be processed by the USWFS in accordance with the provisions of the ESA and the applicable regulations (50 CFR 13 and 17) and will be subject to the appropriate level of environmental review under the provisions of NEPA.



REFERENCES

Akçakaya, H.R. 1997 Ecological risk analysis for the marbled murrelet. The sensitivity of viability to the

parameters of the zone-4 metapopulation model. Applied Biomathematics. Setauket, New York. 20 pp.

Arnett, E.B. and R.F. May 2016 Mitigating Wind Energy Impacts on Wildlife: Approaches for Multiple Taxa. Human–

Wildlife Interactions 10:28–41.

Arnett, E.B., G.D. Johnson, W.P. Erickson, and C.D. Hein 2013 A synthesis of operational mitigation studies to reduce bat fatalities at wind energy

facilities in North America. Bat Conservation International and Western EcoSystems Technology, Inc. Austin, Texas.

Avian Power Line Interaction Committee (APLIC) 2012 Reducing Avian Collisions with Power Lines: The State of the Art in 2012. Edison Electric

Institute and APLIC. Washington D.C.

Baglien, J.W 1975 Biology and habitat requirements of the nesting Golden Eagle in southwestern Montana.

M.S. Thesis. Montana State University, Bozeman. 63 pages.

Band, B. 2012 Using a collision risk model to assess bird collision risks for offshore windfarms. Report to

the Crown Estate Strategic Ornithological Support Services. Project SOSS-02. Thetford, British Trust for Ornithology.

Barbaree, B.A., S.K. Nelson, and B.D. Dugger 2014 Marine space use by marbled murrelets (Brachyramphus marmoratus) at a mainland fjord

system in southeast Alaska. Marine Ornithology 43:1–10.

Bates, J.W. and M.O. Moretti 1994 Golden eagle (Aquila chrysaetos) population ecology in eastern Utah. The Great Basin

Naturalist 54:248–255.

Bay, K., K. Nasman, W. Erickson, K. Taylor, and K. Kosciuch 2016 Predicting eagle fatalities at wind facilities. Journal of Wildlife Management 80:1000–

1010.

Beissinger, S.R. 1995 Population trends of the marbled murrelet projected from demographic analyses. Pages

385–393 in C.J. Ralph, G.L. Hunt, M.G. Raphael, and J.F. Piatt (Tech. eds.), Ecology and Conservation of the Marbled Murrelet. General Technical Report PSW-GTR-152. U.S. Forest Service, Pacific Southwest Research Station, Albany, California.



Beissinger, S.R., and M.Z. Peery 2007 Reconstructing the historic demography of an endangered seabird. Ecological

Applications 16:1516–1528.

Bloxton, T.D., and M.G. Raphael 2009 Breeding ecology of the marbled murrelet in Washington State: project update 2004-

2008. U.S. Department of Agriculture, Forest Service Pacific Northwest Research Station Portland, Oregon. 41 pp.

Bolker, E.D., J.J. Hatch, and C. Zara 2006 Modeling bird passage through a windfarm. Draft. Departments of Mathematics and

Computer Science. University of Massachusetts, Boston.

2014 Modeling how windfarm geometry affects bird mortality. arXiv:1408.1580 (preprint).

Boulanger, J.G. 2000 Quantitative tools for the monitoring of threatened species: a marbled murrelet case

study. Pages 225–230 in: Proceedings of a conference on the biology and management of species and habitats at risk, Kamloops, B.C., 15-19 Feb. 1999, Volume One (Darling, L.M., ed.). B.C. Ministry of Environment, Lands, and Parks, Victoria, B.C., and University College of the Cariboo, Kamloops, B.C.

Boulanger, J.G., K. Martin, G. Kaiser, and A.E. Derocher 1999 Evaluation uncertainty in estimating population trends of marbled murrelets. Pages 55–

63 in A. W. Diamond and A. W. Nettleship (eds.), Biology and Conservation of Forest Birds. Society of Canadian Ornithologists, Fredericton, NB, Canada.

Bradley, R.W. and F. Cooke 2001 Cliff and deciduous tree nests of marbled murrelets in Southwestern British Columbia.

Northwestern Naturalist 82: 52–57.

Buehler, D.A. 2000 Bald eagle (Haliaeetus leucephalus). In The Birds of North America, No. 506 (A. Poole and

F. Gill, eds.). The Birds of North America, Inc., Philadelphia, Pennsylvania.

Burger, A.E. 2002 Conservation assessment of marbled murrelets in British Columbia, a review of the

biology, populations, habitat associations and conservation. Canadian Wildlife Service, Environmental Conservation Branch, Pacific and Yukon Region. Technical Report Series 387.

Burke Museum of Natural History and Culture 2011 Washington Territorial University Herbarium Database. Available Online

at:http://www.burkemuseum.org/research-and-collections/botany-and-herbarium/collections/database/. Accessed May 2018.

Carter, H.R. and S.G. Sealy 1986 Year-round use of coastal lakes by marbled murrelets. Condor 88:473–477.



Caswell, H. 2001 Matrix population models: construction, analysis and interpretation. Sinauer Associates,

Sunderland, Massachusetts.

Cooper Beauchesne and Hemmera Envirochem, Inc. 2016 2015 marbled murrelet/migrant songbird radar monitoring. Cape Scott Wind Farm, Port

Hardy, British Columbia. Prepared for Cape Scott Wind LP. Markham, Ontario, Canada.

Craig, T.H., and F.H. Craig 1984 A large concentration of roosting golden eagles in southeastern Idaho. Auk 101:610–613.

Desimone, S.M. 2016 Periodic status review for the marbled murrelet in Washington. Washington Department

of Fish and Wildlife, Olympia, Washington.

Dias, C.T.D.S., A. Samaranayaka, and B. Manly 2008 On the use of correlated beta random variables with animal population modelling.

Ecological Modelling 215:293–300.

Eichhorn, M., K. Johst, R. Seppelt, and M. Drechsler 2012 Model-based estimation of collision risks of predatory birds with wind turbines. Ecology

Society 17:1–12.

Falxa, G.A., M.G. Raphael, C. Strong, J. Baldwin, M. Lance, D. Lynch, S.F. Pearson, and R.D. Young 2016 Status and trend of marbled murrelet populations in the Northwest Forest Plan Area.

Chapter 1 in Falxa, G.A. and M.G. Raphael, technical coordinators (2016): Northwest Forest Plan—the first 20 years (1994-2013): status and trend of marbled murrelet populations and nesting habitat. General Technical Report PNW-GTR-933. U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station, Portland, Oregon.

Flowers, J.M. 2015 Design and testing of an integrated wildlife-wind turbine interactions detection system.

Thesis, Oregon State University, Corvallis, Oregon.

Friesen, V.L., T.M. Burg, and K. McCoy 2007 Mechanisms of population differentiation in seabirds. Molecular Ecology 16:1765–1785.

Gehring, J., P. Kerlinger, and A.M. Manville 2009 Communication towers, lights, and birds: successful methods of reducing the frequency

of avian collisions. Ecological Applications 19: 505–514.

Hamer, T.E., and S.K. Nelson 1995 Characteristics of marbled murrelet nest trees and nesting stands. Pages 69–82 in C.J.

Ralph, G.L. Hunt, M.G. Raphael, and J.F. Piatt (Technical editors). Ecology and Conservation of the Marbled Murrelet. General Technical Report PSW-GTR-152, USDA Forest Service, Pacific Southwest Research Station, Albany, California.



Hebert, P.N. and R.T. Golightly 2008 At-sea distribution and movements of nesting and non-nesting marbled murrelets

(Brachyramphus marmoratus) in northern California. Marine Ornithology 36:99–105.

Hobson, K.A. 1990 Stable isotope analysis of marbled murrelets: evidence for freshwater feeding and

determination of trophic level. Condor 92:897–903.

Huff, M.H., M.G. Raphael, S.L. Miller, S.K. Nelson, and J. Baldwin, technical coordinators 2006 Northwest Forest Plan—the first 10 years (1994-2003): status and trends of populations

and nesting habitat for the marbled murrelet. General Technical Report PNW-GTR-650, USDA Forest Service, Pacific Northwest Research Station, Portland, Oregon.

Hull, C.L., G.W. Kaiser, C. Lougheed, L. Lougheed, S. Boyd, and F. Cook 2001 Interspecific variation in commuting distance of Marbled Murrelets (Brachyramphus

marmoratus: Ecological and energetic consequences of nesting further inland. Auk118:1036–1046.

International Panel on Climate Change (IPCC) 2007 Climate Change 2007: Impacts, Adaptation and Vulnerability. Cambridge University Press.

Cambridge, New York.

Jenkins A., J. Smallie, and M. Diamond 2010 Avian collisions with power lines: A global review of causes and mitigation with a South

African perspective. Bird Conservation International 20:263–278.

Jones, G.P., L.G. Pearlstine, and H.F. Percival 2006 An assessment of small unmanned aerial vehicles for wildlife research. Wildlife Society

Bulletin 34:750–758.

Katzner, T.E., D. Brandes, T. Miller, M. Lanzone, C. Maisonneuve, J.A. Tremblay, R. Mulvihill, G.T. Merovich, and D. Thompson 2012 Topography drives migratory flight altitude of golden eagles: implications for on-shore

wind energy development. Journal of Applied Ecology, 49: 1178–1186.

Kochert, M.N., K. Steenhof, C.L. Mcintyre, and E.H. Craig 2002 Golden Eagle (Aquila chrysaetos), The Birds of North America Online (A. Poole, Editor).

Cornell Lab of Ornithology. Ithaca, New York.

Lance, M.M., S.F. Pearson, M.G. Raphael, and T.D. Bloxton 2013 2012 Washington at-sea marbled murrelet population monitoring: research progress

report. Washington Department of Fish and Wildlife, Wildlife Science Division, Olympia, Washington, and USDA Forest Service Pacific Northwest Research Station, Olympia, Washington. 23 pp.



Lance, M.M., and S.F. Pearson 2016 Washington 2015 at-sea marbled murrelet population monitoring: research progress

report. Washington Department of Fish and Wildlife, Wildlife Science Division, Olympia, Washington.

Livingston, S.A., C.S. Todd, W.B. Krohn, and R.B. Owen, Jr. 1990 Habitat models for nesting bald eagles in Maine. Journal of Wildlife Management 54:644–

653.

Lorenz, T.J., M.G. Raphael, T.D. Bloxton, and P.G. Cunningham 2017 Low breeding propensity and wide-ranging movements by marbled murrelets in

Washington. Journal of Wildlife Management 81:306–321.

Lynch, D., G. Falxa, J. Baldwin, M.M. Lance, S.K. Nelson, S.F. Pearson, M.G. Raphael, C. Strong, and R. Young 2016 Marbled murrelet effectiveness monitoring, Northwest Forest Plan: 2015 summary

report. U.S. Fish and Wildlife Service, Washington Department of Fish and Wildlife, and U.S Forest Service. 19 pp.

Manly, B.F.J. 1991 Randomization and Monte Carlo methods in biology. Chapman and Hall, London, United

Kingdom.

Marzluff, J.M., S.T. Knick, M.S. Vekasy, L.S. Schueck, and T.J. Zarriello 1997 Spatial use and habitat selection of golden eagles in southwestern Idaho. Auk 114, 673–

687.

McClure, C., L. Martinson, and T. Allison 2018 Automated monitoring for birds in flight: Proof of concept with eagles at a wind power

facility. Biological Conservation 224: 26–33.

McShane, C., T. Hamer, H.R. Carter, G. Swartzman, V. Friesen, D. Ainley, R. Tressler, K. Nelson, A. Burger, L. Spear, T. Mohagen, R. Martin, L. Henkel, K. Prindle, C. Strong, and J. Keany 2004 Evaluation report for the 5-year status review of the marbled murrelet in Washington,

Oregon, and California. Unpublished report, prepared for U.S. Fish and Wildlife Service, by EDAW Inc., Seattle, Washington.

Melillo, J.M., T.C. Richmond, and G.W. Yohe, Eds. 2014 Climate change impacts in the United States: The Third National Climate Assessment. U.S.

Global Change Research Program, 841 pp.

Meyer, C.B., S.L. Miller, and C.J. Ralph 2004 Stand-scale habitat associations across a large geographic region of an old-growth

specialist, the marbled murrelet. Wilson Bulletin 116:197–210.



Miller, S.L., C.J. Ralph, M.G. Raphael, G. Strong, C. Thompson, J. Baldwin, and M.H. Huff 2006 At-sea monitoring of marbled murrelet population status and trend in the Northwest Plan

area. Pages 31–60 in: Huff, M.; Raphael, M.G.; Miller, S.L.; Nelson, S.K.; Baldwin, J.; technical coordinators. Northwest Forest Plan—the first 10 years (1994-2003): status and trends of populations and nesting habitat for the marbled murrelet. General Technical Report PNW-GTR-650. Portland, Oregon: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station, Portland, Oregon.

Millsap, B.A., and S.L. Vana 1984 Distribution of wintering golden eagles in the eastern United States. Wilson Bulletin

96:692–701.

Millsap, B., T. Breen, L. McConnell, T. Steffer, L. Philips, N. Douglass, and S. Taylor 2004 Comparative fecundity and survival of bald eagles fledged from suburban and rural natal

areas in Florida. Journal of Wildlife Management 68: 1018–1031.

Millsap, B.A., M.E. Cooper, and G. Holroyd 2007 Legal considerations. Pages. 437–449 in D.M. Bird and K.L. Bildstein (editors), Raptor

Research and Management Techniques. Hancock House, Surrey, British Columbia, Canada.

Millsap, B.A., G.S. Zimmerman, J.R. Sauer, R.M. Nielson, M. Otto, E. Bjerre, and R. Murphy 2013 Golden eagle population trends in the western United States: 1968–2010. Journal of

Wildlife Management 77:1436–1448.

Mojica, E.K., J.M. Meyers, B.A. Millsap, and K.L. Haley 2008 Migration of Florida sub-adult bald eagles. Wilson Bulletin 120:304–310.

Morris, W.F. and D.F. Doak 2002 Quantitative conservation biology: theory and practice of population viability analysis.

Sinauer Associates, Sunderland, Massachusetts.

Nagy, L., B. Gibson, K. Kosciuch, J. Taylor, B. Gunderman 2013 Whooping crane and sandhill crane behavior at an operating wind farm. Poster

presentation. National Wind Coordinating Collaborative. Wind Wildlife Research Meeting IX. Broomfield, Colorado. November 28-30, 2012.

Nelson, S.K. 1997 Marbled murrelet (Brachyramphus marmoratus). Pages 1–32 in A. Poole and G. Gill, (eds.)

Birds of North America, No. 276 Academy of Natural Sciences, Philadelphia, and American Ornithologists’ Union, Washington, DC.

Nelson, S.K., S.M. Desimone, M.A. Kappes, A.E. Burger, B.A. Cooper, and M. Ostwald 2013 Radar survey protocol for marbled murrelets at proposed land-based wind energy

developments. Unpublished Report, U.S. Fish and Wildlife Service, Olympia, Washington.



Nelson, S.K. and T.E. Hamer 1995 Nest Success and the Effects of Predation on marbled murrelets. Pages 89–98 in: C.J.

Ralph, G.L. Hunt Jr., M.G. Raphael, J.F. Piatt (Technical editors). Ecology and conservation of the marbled murrelet. General Technical Report PSW-GTR-152. Albany, California: Pacific Southwest Research Station, Forest Service, U.S. Department of Agriculture, Albany, California.

Pagel, J.E., K.J. Kritz, B.A. Millsap, R.K. Murphy, E.L. Kershner, and S. Covington 2013 Bald eagle and golden eagle mortalities at wind energy facilities in the contiguous United

States. Journal of Raptor Research 47: 311–315.

Palmer, R.S. 1988 Handbook of North American Birds. Vol. 5. Yale University Press, New Haven, Connecticut.

Peery, M.Z. and R.W. Henry 2010 Recovering marbled murrelets via corvid management: a Population Viability Analysis

Approach. Biological Conservation 143:2414–2424.

Peery, M.Z., B.H. Becker, and S.R. Beissinger 2006 Combining demographic and count-based approaches to identify source-sink dynamics:

An example using a threatened seabird. Ecological Applications 16:1516–1528.

Piatt, J.F., K.J. Kuletz, A.E. Burger, S.A. Hatch, V.L. Friesen, T.P. Birt, M.L. Arimitsu, G.S. Drew, A.M. Harding, K.S. Bixler 2007 Status review of the Marbled Murrelet (Brachyramphus marmoratus) in Alaska and British

Columbia. U.S. Geological Survey and Department of Interior, Reston, Virginia.

Pryor S.C. and R.J. Barthelmie 2010 Climate change impacts on wind energy: a review. Renewable and Sustainable Energy

Reviews 14:430–437.

Ralph, C.J., G.L. Hunt, M.G., Raphael, J.F. Piatt (technical editors) 1995 Ecology and conservation of the Marbled Murrelet. General Technical Report PSW-GTR-

152. Albany, California, Pacific Southwest Research Station, U.S. Department of Agriculture, Albany, California.

Raphael, M.G., D.M. Evans Mack, and B.A. Cooper 2002 Use of radar to investigate landscape-scale relationships between abundance of marbled

murrelets and nesting habitat. Condor 104:331–342.

Raphael, M.G., J. Baldwin, G.A. Falxa, M.H. Huff, M. Lance, S.L. Miller, S.F. Pearson, C.J. Ralph, C. Strong, and C. Thompson 2007 Regional population monitoring of the marbled murrelet: field and analytical methods.

General Technical Report PNW-GTR-716. U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station, Portland, Oregon.



Raphael, M.G., S.K. Nelson, P. Swedeen, M. Ostwald, K. Flotlin, S. Desimone, S. Horton, P. Harrison, D. Prenzlow Escene, and W. Jaross 2008 Recommendations and supporting analysis of conservation opportunities for the marbled

murrelet long-term conservation strategy. Washington State Department of Natural Resources, Olympia, Washington. Available online at: http://file.dnr.wa.gov/publications/lm_mamu_sci_team_report.pdf. Accessed June 2018.

Raphael, M.G., G.A. Falxa, D. Lynch, S.K. Nelson, S.F. Pearson, A.J. Shirk, and R.D. Young 2016 Status and trend of nesting habitat for the Marbled Murrelet under the Northwest Forest

Plan. Chapter 2 in: Falxa, G.A. and M.G. Raphael (Technical editors). 2016: Northwest Forest Plan—the first 20 years (1994-2013): Status and trend of Marbled Murrelet populations and nesting habitat. General Technical Report. PNW-GTR-933. U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station. Portland, Oregon.

Ridgley, R.S., T.F. Allnutt, T. Brooks, D.K. McNicol, D.W. Mehlman, B.E. Young, and J.R. Zook 2007 Digital distribution maps of the birds of the Western Hemisphere, version 3.0.

NatureServe, Arlington, Virginia.

Sanzenbacher, P.M., T.J. Mabee, and B.A. Cooper 2015 A radar and visual study of marbled murrelets at the proposed Skookumchuck Wind

Energy Project, summer 2013 and 2014. Alaska Biological Research, Inc. (ABR). Forest Grove, Oregon. Prepared for: RES America Developments, Inc. Broomfield, Colorado.

Sheppard, J.K., A. McGann, M. Lanzone, A. Walsh, M. Wallace, and R. Swaisgood 2014 Curtailing avian impacts with wind turbines using GSM/GPS tracking telemetry that

incorporates autonomous geofence alerts. San Diego Zoo Global and Cellular Tracking Technologies. Presentation at Wind Wildlife Research Meeting X on December 3–5, Broomfield, Colorado.

Smales, I., S. Muir, C. Meredith, and R. Baird 2013 A description of the biosis model to assess risk of bird collisions with wind turbines.

Wildlife Society Bulletin 37:59–65.

Thumthae, C. 2015 Optimum blade profiles for a variable-speed wind turbine in low wind area. Energy

Procedia 75: 651–657.

Tucker, V.A. 1996 A mathematical model of bird collisions with wind turbine rotors. Journal of Solar Energy

Engineering 118:253–262.

United States Department of Agriculture (USDA) 2018 Northwest Forest Plan. Forest Service – Pacific Northwest Region. Available Online at:

https://www.fs.usda.gov/detail/r6/landmanagement/planning/?cid=fsbdev2_026990. Accessed July 2018.



United States Environmental Protection Agency (USEPA) 2016a Ecoregions of North America. Available Online at: https://www.epa.gov/eco-

research/ecoregions-north-America. Accessed July 2018.

2016b Climate Change Indicators: Weather and Climate. Available Online at: https://www.epa.gov/climate-indicators/weather-climate. Accessed July 2018.

United States Fish and Wildlife Service (USFWS) 1997 Recovery plan for the threatened marbled murrelet (Brachyramphus marmoratus) in

Washington, Oregon, and California populations. Department of the Interior, Region 1. Portland, Oregon.

2012a Land-based Wind Energy Guidelines. Department of the Interior. Arlington, Virginia.

2012b Report on marbled murrelet recovery implementation team meeting and stakeholder workshop. USFWS, Lacey, Washington, April 17, 2012. 66 pp.

2013 Eagle Conservation Plan Guidance – Land-based Wind Energy. Version 2. U.S. Fish and Wildlife Service Division of Migratory Bird Management, Hadley, Massachusetts.

2016a Eagle permits; Revisions to regulations for eagle incidental take and take of eagle nests. Final Rule. Department of the Interior. Federal Register 81(242):91494–91554.

2016b Endangered and threatened wildlife and plants: determination of critical habitat for the marbled murrelet. U.S. Department of Interior. Final Determination. Federal Register 81(17): 51348-51370.

2016c Bald and Golden Eagles: Population demographics and estimation of sustainable take in the United States, 2016 update. Division of Migratory Bird Management, Washington D. C.

2016d Programmatic Environmental Impact Statement for the Eagle Rule Revision. Department of the Interior, Fall Church, Virginia D.C.

2017a Interim guidance for implementing the Endangered Species Act Compensatory Mitigation Policy. U.S. Department of Interior, Washington D.C. January. Available online at: https://www.fws.gov/endangered/improving_esa/cmp.html. Accessed June 2018.

2017b Biological Opinion 2017-2036 Puget Sound Treaty and Non-Treaty (All-Citizen) Salmon Fisheries. Puget Sound, Washington. Reference 01EWFW00-2016-F-1181. Lacy, Washington.

2018 Recommended best practices for communication tower design, siting, construction, operation, maintenance, and decommissioning, Migratory Bird Program. Fall Church, Virginia.

United States Fish and Wildlife Service (USFWS) and National Marine Fisheries Service (NMFS) 2016 Habitat Conservation Planning and Incidental Take Permit Processing Handbook.

Available Online at: https://www.fws.gov/endangered/esa-library/pdf/HCP_Handbook.pdf. Accessed May 2018.



Washington Department of Fish and Wildlife (WDFW) 2009 Wind Power Guidelines. Olympia, Washington.

2016 Long-term conservation strategy for the marbled murrelet. Draft Environmental Impact Statement. Olympia, Washington. Available online at: https://www.dnr.wa.gov/mm. Accessed May 2018.

2018 Priority Habitats and Species - bald eagle nest GIS request. Natural Heritage Program, Lacey, Washington. Received August 10.

Washington Department of Natural Resources (WDNR) 2016 Wildlife Survey Data Management. Corporate data-base. Olympia, Washington.

Watson, J. 1997 The Golden Eagle. Princeton University Press: London.

Watts, B.D., and E.K. Mojica 2012 Use of satellite transmitters to delineate bald eagle communal roosts within the Upper

Chesapeake Bay. Journal of Raptor Research 46, 121–128.

Western Regional Climate Center (WRCC) 2017 Recent Climate in the West. Available Online at: https://wrcc.dri.edu/. Accessed May

2018.

Whitfield, D.P. 2009 Collision avoidance of golden eagles at wind farms under the ‘Band’ collision risk model.

Report to Scottish Natural Heritage. Natural Research Ltd, Banchory, UK.

Wilk, R.J., M.G. Raphael, and T.D. Bloxton, Jr. 2016 Nesting habitat characteristics of marbled murrelets occurring in near-shore waters of the

Olympic Peninsula, Washington. Journal of Field Ornithology 87:162–175.

Wingfield, G. 1991 National Wildlife Federation eagle survey. Nebraska Game and Parks Commission,

Lincoln, Nebraska.

Zeiner, D.C., W.F. Laudenslayer, Jr., K.E. Mayer, and M. White, editors 1988-1990 California's Wildlife. Vol. I-III. California Department of Fish and Game,

Sacramento, California.



GLOSSARY

Adaptive Management – A formal process for: (a) evaluating the current resource status; (b) evaluating the effectiveness of rules and guidance necessary to meet the goals and objectives for the protection, maintenance, and enhancement of resources; and (c) based on the findings, making any necessary adjustments to management practices to achieve resource objectives.

Avoidance Probability – The likelihood or probability that a bird (or bat) would actively evade a wind turbine. Failure to avoid does not imply collision, as a bird may pass through an active rotor without being struck by a blade.

Canopy – The continuous cover of branches and foliage formed collectively by the crowns of adjacent trees and other woody growth.

Changed Circumstances – Changes in circumstances affecting a species or geographic area covered by [an HCP] that can reasonably be anticipated by [plan] developers and the Services and that can be planned for (e.g., the listing of new species, or a fire or other natural catastrophic event in areas prone to such events).” (50 CFR 17.3).

Code of Federal Regulations (CFR) – A codification of the general and permanent rules published in the Federal Register by the Executive departments and agencies of the federal government.

Collision Probability – The likelihood that a bird/turbine collision will occur, dependent on how the risk space (the air space around a turbine) is defined and the turbine structure in question (e.g., tower or rotor). For active rotors, collision probability as assessed by the Tucker (1996) model depends on bird air speed, wind speed, and the physical dimensions and dynamic properties of the rotor.

Collision Rate – The rate of an eagle colliding with a turbine per exposure in the hazardous area (fatalities per hours·kilometers2), where all collisions are considered to be fatal.

Compliance Monitoring – Monitoring to verify compliance with the terms and conditions of the ITP and HCP.

Covered Activities -– Activities for which the Applicant is seeking take authorization under ESA section 10(a)(1)(B). These Covered Activities include operation and maintenance of the Project (including emergency repairs and responses).

Critical Habitat, Federal – (1) The specific areas within the geographical area currently occupied by a species, at the time it is listed in accordance with the Act, on which are found those physical or biological features (i) essential to the conservation of the species and (ii) that may require special management considerations or protection, and (2) specific areas outside the geographical area occupied by a species at the time it is listed upon a determination by the Secretary of the Interior that such areas are essential for the conservation of the species. 50 CFR 402.02

Curtailment – Modifications to turbine operation including modifying the timing of turbine operation.



Demographic Stochasticity – The variability in numbers of individuals in a population arising from applying probabilities of birth and death to a finite sample. This is particularly an issue for small populations and can lead to high probabilities of extirpation.

Distinct Population Segment (DPS) – The combined marbled murrelet population located in California, Oregon, and Washington, representing that portion of the global population of marbled murrelets that is listed as threatened under the Endangered Species Act.

Eagle Management Unit (EMU) – A species specific area in which the project is located.

Effectiveness Monitoring – Monitoring conducted to determine whether the Habitat Conservation Plan conservation strategies have the anticipated results.

Endangered Species – A species that is in danger of extinction throughout all or a significant portion of its range. 50 CFR 402.02, 16 U.S.C. §1531 et seq. (1973)

Environmental Stochasticity – The variability in numbers of individuals in a population arising from environmental conditions (e.g., weather) which vary over time. These changes will have effects on survival or fecundity or both. Environmental variability is generally assumed to be random, even if there are underlying deterministic factors.

Extirpation – The elimination of a population from a particular area, as distinguished from extinction of a species.

Federally Listed – Species formally listed as a threatened or endangered species under the federal Endangered Species Act; designations are made by the U.S. Fish and Wildlife Service or National Marine Fisheries Service.

Fragmentation – The spatial arrangement of successional stages across the landscape as the result of disturbance; often used to refer specifically to the process of reducing the size and connectivity of adjacent stands (See Stand). Fragmentation of existing habitat increases the accessibility of nest sites to predators and isolates portions of the population.

Geographic Information System (GIS) – A computer system that stores and manipulates spatial data and can produce a variety of maps and analyses.

Habitat Conservation Plan (HCP) – A conservation plan required as part of an ESA Section 10(a)(1)(B) Incidental Take Permit application under the federal Endangered Species Act.

Harm – Defined by U.S. Fish and Wildlife Service regulations as “an act which actually kills or injures wildlife and may include significant habitat modification or degradation where it actually kills or injures wildlife by significantly impairing essential behavioral patterns including breeding, feeding or sheltering” (50 CFR 17.3).

Harass – Defined by U.S. Fish and Wildlife Service regulations as “an intentional or negligent act or omission which creates the likelihood of injury to wildlife by annoying it to such an extent as to significantly disrupt normal behavioral patterns which include, but are not limited to, breeding, feeding, or sheltering” (50 CFR 17.3).



Horizontal Interaction Probability – Probability that a murrelet will encounter a wind turbine under the assumptions: (1) that a single wind turbine occupies the space within a rectangle with width equal to the radar survey diameter (3 km) and height equal to turbine height; (2) murrelet flights are perpendicular to the plane of that rectangle; and (3) murrelet flights are uniformly distributed within the rectangle, both with respect to width and height.

IdentiFlight® – A machine vision technology that can be used to curtail turbines when eagles are at risk to reduce the probability of collision risk.

Incidental Take – The taking of a federally listed wildlife (animal) species, if the taking is incidental to, and not the purpose of, carrying out otherwise lawful activities. See also “Take.”

Incidental Take Permit (ITP) – Permit issued by the U.S. Fish and Wildlife Service pursuant to ESA section 10(a)(1)(B) that authorizes the incidental take of species protected by the ESA.

Land Based Wind Energy Guidelines (WEGs) – Voluntary guidelines developed by the USFWS in coordination with stakeholders that provide a tiered process for addressing wildlife conservation concerns at all stages of land-based wind energy development.

Landscape – Large regional units of lands that are viewed as a mosaic of communities, or a unit of land with separate plant communities or ecosystems forming ecological units with distinguishable structure, function, geomorphology, and disturbance regimes.

Local Area Population (LAP) – The population in the vicinity of a particular project or activity. (Federal Register Doc. 2016-29908) For the Skookumchuck Wind Energy Project, the LAP for the Project was buffered by 86 miles for bald eagles and 109 miles for golden eagles.

Matrix Model – A population model that accounts directly for the life history of an organism. Elements within the matrix represent age- or stage-specific vital rates such as survival and fecundity.

Northwest Forest Plan (NWFP) –A set of federal policies and guidelines governing land use on federal lands in the Pacific Northwest region of the United States. The series of policies cover 10 million hectares within Western Oregon and Washington as well as a small part of Northern California (USDA 2018).

Parameterization – The process of describing a model in terms of certain quantities or parameters.

Passage Rate – The number of birds that pass through a particular area in a specified amount of time. For marbled murrelets, passage rate was estimated from radar surveys.

Plan Area/Permit Area – The entirety of leased parcels proposed for the Project including 9,697 acres and the mitigation lands.

Population Viability Analysis (PVA) – Analysis to assess the likelihood that a population will persist over a particular period of time, or equivalently, the probability that a population will go extinct or become extirpated within a particular period.

Power Pole Retrofit – Actions to retrofit, reframe, or rebuild power poles to avian-safe designs.



Recovery Plan – A plan developed by a government agency, that if implemented is expected to result in the recovery of a threatened or endangered species to the extent that the species can be delisted from threatened or endangered status.

Re-power – The process of updating the Project components in order to update old equipment to align with impending technological advances in equipment.

Resource Equivalency Analysis (REA) – Developed by the U.S. Fish and Wildlife Service, an analysis to estimate the number of power pole retrofits necessary to offset the take of golden and bald eagles from energy development.

Riparian Area – Areas of land directly influenced by water or that influence water. Riparian areas usually have visible vegetative or physical characteristics reflecting the influence of water. Riversides and lake shores are typical riparian areas. Commonly referred to as “riparian zone.”

Salvage – The removal of snags, downed logs, windthrow, or dead and dying material.

Sensitive Species– A state designation. Species native to the state of Washington that are vulnerable or declining and are likely to become endangered or threatened in a significant portion of their ranges within the state without cooperative management or the removal of threats.

Stand – A group of trees that possess sufficient uniformity in composition, structure, age, spatial arrangement, or condition to distinguish them from adjacent groups.

Take – To harass, harm, pursue, hunt, shoot, wound, kill, trap, capture, or collect, or to attempt to harass, harm, pursue, hunt, shoot, wound, kill, trap, capture, or collect. 50 CFR 222.102.

Threatened Species – Any species that is likely to become an endangered species within the foreseeable future throughout all or a significant portion of its range. 50 CFR 402.02

Unmanned Aerial Vehicle (UAV or drone) – Emerging technologies for wildlife surveys relying on remote control.

Unforeseen Circumstances – Changes in circumstances affecting a species or geographic area covered by the HCP that could not reasonably have been anticipated by plan developers and the U. S. Fish and Wildlife Service at the time of the conservation plan’s negotiation and development and that result in a substantial and adverse change in the status of the covered species. 50 CFR 17.3.

Vertical Interaction Probability (VIP) – Calculated from the empirical distribution of murrelet flight heights recorded in the ABR radar study, a calculation of the probability of murrelet interaction with a turbine.

WDFW – The Washington Department of Fish and Wildlife works to preserve, protect and perpetuate fish, wildlife, and ecosystems while providing sustainable fish and wildlife recreational and commercial opportunities.

Windthrow – Trees blown down by wind; also called “blowdown.”

ATTA

CHM

ENT

A – R

adar

and

Visu

al S

tudy

of M

urre

lets

Sum

mer

s of 2

013

and

2014

A RADAR AND VISUAL STUDY OF MARBLED MURRELETS AT THE PROPOSED SKOOKUMCHUCK WIND ENERGY PROJECT,

SUMMER 2013 AND 2014

PETER M. SANZENBACHER, TODD J. MABEE, AND BRIAN A. COOPER

PREPARED FORRES AMERICA DEVELOPMENTS, INC.

BROOMFIELD, CO

PREPARED BYABR, INC.–ENVIRONMENTAL RESEARCH & SERVICES

FOREST GROVE, OR

Printed on recycled paper.

A RADAR AND VISUAL STUDY OF MARBLED MURRELETS AT THE PROPOSED SKOOKUMCHUCK WIND ENERGY PROJECT,

SUMMER 2013 AND 2014

FINAL REPORT

Prepared for

RES America Developments, Inc.11101 W. 120th Ave., Ste. 400

Broomfield, CO 80021

Prepared by

Peter M. Sanzenbacher, Todd J. Mabee, and Brian A. Cooper

ABR, Inc.—Environmental Research & Services

P.O. Box 249

Forest Grove, Oregon 97116

June 2015

iii Skookumchuck Marbled Murrelet Radar Study

EXECUTIVE SUMMARY

• This report presents the results of a radarand audiovisual (AV) study of MarbledMurrelets conducted during summer 2013and 2014 at the proposed SkookumchuckWind Energy Project (hereafter Project),located in Lewis and Thurston Counties, inwestern Washington.

• The exact size of the development, includingthe number, megawatt (MW) capacity, andspecific wind turbine model will be determinedcloser to the time of construction; however, thecurrent project design consists of 52 VestasV110 2.0 MW wind turbines with a combinedgenerating capacity of up to 104 MW. Themaximal height of the current proposedturbines would be 135 m.

• The primary goal of this study was to quantifyand characterize flight patterns of MarbledMurrelets at the Project and assess potentialrisk of murrelet collision fatalities at proposedturbines. The specific objectives were tocollect information on the number, flight paths,and flight altitudes of Marbled Murreletsflying over the proposed Project during thesummers of 2013 and 2014 and use those datato (1) calculate an exposure rate estimating thefrequency that murrelets would pass within theairspace occupied by the proposed turbines and(2) estimate the annual number of potentialcollision fatalities.

• Surveys were conducted at ten differentsampling stations during the summer breedingperiod of Marbled Murrelets (mid-May toearly August) during the morning activityperiod for murrelets (i.e., from 105 min beforesunrise to 75 min after sunrise). Each surveyincluded concurrent surveillance and verticalradar sampling and also a dedicatedaudiovisual observer. We conducted a total of50 surveys in 2013 and 70 surveys in 2014.

• We observed a total of 26 murrelet radartargets (pre-sunrise landward and seawardtargets) in 2013 (0−7 targets per station) and 47murrelet targets in 2014 (2−7 targets perstation). The mean pre-sunrise murreletpassage rate (landward + seaward targets/d)was 0.52 ± 0.11 targets/d in 2013 and 0.70 ±

0.06 targets/d in 2014. The overall passage rateaveraged across both seasons was 0.61 ± 0.09target/d.

• We did not detect any Marbled Murreletsduring AV observations in 2013 or 2014.

• A qualitative assessment of murrelet targetflight paths indicated the highest levels ofactivity occurred at Station 10. Flightdirections of murrelet targets were variable ateach station, but the percentage of landwardflying versus seaward flying murrelet targetswas 38% landward and 62% seaward in 2013and 40% landward and 60% seaward in 2014.

• We used vertical radar to measure flightaltitudes of 21 murrelet targets concurrentlydetected on surveillance radar. The mean flightaltitudes of these targets, measured relative tothe elevation of the Project ridgelines turbinestrings will be located, was 219.3 ± 34.6 mabove ground level (agl). Based on these flightaltitudes, 38.1% (8 of 21 targets) of targetsidentified on vertical radar flew below theheight of the proposed turbines (i.e., 135 magl).

• We modeled the estimated the number ofcollision fatalities (fatality rate), calculated asthe product of: (1) the exposure rate (i.e., thenumber of times that murrelets might flywithin the airspace occupied by turbines/d); (2)the avoidance probability (i.e., the probabilitythat a murrelet will detect and avoid enteringthe airspace containing the turbine); and (3) thefatality probability (i.e., the probability of afatal collision with a portion of the structureonce a murrelet is in the airspace occupied by aturbine. We also applied a series of adjustmentfactors to account for daily and seasonaldifferences in murrelet activity and arrive atseasonal and annual fatality estimates perturbine and across all turbines combined.

• The annual fatality rate at the Project, was0.0024–0.0619 murrelet fatalities/turbine/yrand 0.1238–3.2189 murrelet fatalities/yr acrossall 52 turbines at the Project. These estimatesof collision fatalities varied by approach (i.e.,side versus frontal) and avoidance rate(90–99%). Currently, there is little informationon collision avoidance rates for Marbled

Skookumchuck Marbled Murrelet Radar Study iv

Murrelets; however, based on the life history ofthe species and a review of the published dataon other crepuscular and nocturnally activeseabird species, the evidence suggests thatavoidance rates are likely high (>95%).Assuming a 99% avoidance rate the annualfatality rate was 0.0024–0.0048 fatalities/turbine/yr and 0.1238–0.2476 murreletfatalities/yr at all turbines.

• We used a Poisson distribution to estimate theprobability that one, two, three, etc. annualmurrelet collision fatalities based on avoidancefactors of 90–99%. For example, across thefull range of avoidance factors the probabilityof zero annual murrelet fatalities ranged from5–86%, whereas the probability of one annualmurrelet fatality ranged from 13–36%.Assuming a 99% avoidance rate theprobability of zero annual fatalities rangedfrom 74–86% and the probability of oneannual fatality ranged from 13–22%.

v Skookumchuck Marbled Murrelet Radar Study

TABLE OF CONTENTS

EXECUTIVE SUMMARY ..........................................................................................................................iiiACKNOWLEDGMENTS ...........................................................................................................................viiINTRODUCTION ......................................................................................................................................... 1OBJECTIVES................................................................................................................................................ 1STUDY AREA .............................................................................................................................................. 1METHODS.................................................................................................................................................... 2

STUDY DESIGN........................................................................................................................................ 2RADAR EQUIPMENT............................................................................................................................... 4DATA COLLECTION ............................................................................................................................... 4

RADAR DATA........................................................................................................................................ 6DATA ANALYSES.................................................................................................................................... 8

RADAR DATA........................................................................................................................................ 8MODELING FATALITY RATES ............................................................................................................. 9ESTIMATING THE LIKELIHOOD OF FATALITY LEVELS ............................................................... 9

RESULTS...................................................................................................................................................... 9PASSAGE RATES ..................................................................................................................................... 9FLIGHT ALTITUDES ............................................................................................................................. 14FATALITY RATES ................................................................................................................................. 15LIKELIHOOD OF FATALITIES ............................................................................................................ 15

DISCUSSION.............................................................................................................................................. 15TARGET IDENTIFICATION.................................................................................................................. 19PASSAGE RATES ................................................................................................................................... 19FLIGHT ALTITUDES ............................................................................................................................. 19FATALITY ESTIMATES ........................................................................................................................ 20

AVOIDANCE RATES........................................................................................................................... 20SOURCES OF VARIATION IN PASSAGE RATES AND FATALITY ESTIMATES ........................ 20

CONCLUSIONS ......................................................................................................................................... 21LITERATURE CITED................................................................................................................................ 21

LIST OF FIGURES

Figure 1. Map of the proposed Skookumchuck Wind Energy Project in Lewis and Thurston Counties, Washington............................................................................................................... 2

Figure 2. Locations of project lease areas and survey stations at the proposed Skookumchuck Wind Energy Project in Lewis and Thurston Counties, Washington....................................... 3

Figure 3. Surveillance and vertical radar lab used for studies of Marbled Murrelets at the Skookumchuck Wind Energy Project in Lewis and Thurston Counties, Washington, during summer 2013 and 2014 ................................................................................................. 6

Figure 4. Approximate murrelet-sampling airspace for the Furuno FR–1510 marine radar at the 1.5-km range setting, in surveillance and vertical antenna orientations, as determined by field trials with Rock Pigeons, which are similar in size to Marbled Murrelets ................. 7

Figure 5. Map indicating Marbled Murrelet target flight directions designated as landward and seaward at the proposed Skookumchuck Wind Energy Project in Lewis and Thurston Counties, Washington, during summer 2013 and 2014............................................................ 8

Skookumchuck Marbled Murrelet Radar Study vi

Figure 6. Major variables used in estimating possible fatality of Marbled Murrelets at wind turbines at the proposed Skookumchuck Wind Energy Project, Washington, during summer 2013 and 2014........................................................................................................... 11

Figure 7. Map showing the flight paths of Marbled Murrelet radar targets heading in landward and seaward directions observed before sunrise at each of ten radar sampling stations at the proposed Skookumchuck Wind Energy Project, Lewis and Thurston counties, Washington, during summer 2013 and 2014.......................................................................... 14

Figure 8. Location of Marbled Murrelet Conservation Zones 1 and 2 and the inland breeding limit of the Marbled Murrelet in Washington state. Adapted from USFWS.......................... 17

Figure 9. Geographic planning blocks evaluated for designation as Marbled Murrelet Management Areas ................................................................................................................. 18

LIST OF TABLES

Table 1. Location of sampling stations at the proposed Skookumchuck Wind Energy Project, Lewis and Thurston counties, Washington, during summer 2013 and 2014............................ 4

Table 2. Landward and seaward Marbled murrelet targets observed on radar before sunrise and number of audio-visual detections of Marbled Murrelets at the proposed Skookumchuck Wind Energy Project, Lewis and Thurston counties, Washington, during summer 2013 and 2014.................................................................................................................................... 5

Table 3. Estimated average daily exposure rates and daily fatality rates of Marbled Murrelets during the main breeding period at wind turbines at the proposed Skookumchuck Wind Energy Project, Lewis and Thurston counties, Washington................................................... 10

Table 4. Adjustment factors for modeling of seasonal and annual Marbled Murrelet fatalities per turbine at the proposed Skookumchuck Wind Energy Project, Lewis and Thurston counties, Washington.............................................................................................................. 12

Table 5. Summary of per turbine fatality rates and cumulative fatality rates for Marbled Murrelets at wind turbines during seasonal periods and across the entire year at the proposed Skookumchuck Wind Energy Project, Lewis and Thurston counties, Washington ............................................................................................................................. 13

Table 6. The estimated probability of different annual fatalities of Marbled Murrelets at all 52 wind turbines at the proposed Skookumchuck Wind Energy Project, Lewis and Thurston counties, Washington .............................................................................................. 16

vii Skookumchuck Marbled Murrelet Radar Study

LIST OF APPENDICES

Appendix 1. Information on modeling fatality rates ........................................................................... 27Appendix 2. Sampling dates and numbers of landward and seaward murrelet targets observed

on radar before sunrise, and number of audio-visual detections of Marbled Murrelets at the proposed Skookumchuck Wind Energy Project, Lewis and Thurston counties, Washington, during summer 2013 ................................................... 32

Appendix 3. Sampling dates and numbers of landward and seaward murrelet targets observed on radar before sunrise, and number of audio-visual detections of Marbled Murrelets at the proposed Skookumchuck Wind Energy Project, Lewis and Thurston counties, Washington, during summer 2014 ................................................... 34

Appendix 4. Information on seabird avoidance rates. ......................................................................... 36

ACKNOWLEDGMENTS

We thank RES America Developments, Inc. (RES) for funding this study and Sean Bell of RES forassistance with project coordination and logistics. Weyerhaeuser Company provided access to properties.At ABR, Inc. Matt Apling, Mike Davis, Melinda Malek, Jon Plissner, and Nate Schwab helped with radarand visual sampling; Susan Cooper and Delee Spiesschaert assisted with office logistics and projectcoordination; and Rich Blaha and Pam Odom helped with report preparation.

Skookumchuck Marbled Murrelet Radar Study viii

Introduction

1 Skookumchuck Marbled Murrelet Radar Study

INTRODUCTION

The Marbled Murrelet is a seabird thatinhabits nearshore waters along the Pacific Coastfrom Alaska to California. Murrelets fly up to ~80km inland to nest on limbs of old-growthconiferous trees throughout the southern portion ofthe species range in Washington, Oregon, andCalifornia (Nelson 1997). These populations of theMarbled Murrelet were listed as Threatened in1992 under the U.S. Endangered Species Act,primarily because of excessive loss andfragmentation of nesting habitat, as well asmortality associated with oil spills and gill-netfishing (USFWS 1992, 1997). The MarbledMurrelet also is classified as endangered at thestate level in California (CDFW 2014), threatenedat the state level in Washington and Oregon(ODFW 2014, WDFW 2014), and threatened inCanada (COSEWIC 2012).

Marbled Murrelets visit inland nesting areasprimarily during periods of low ambient light andfly at high speeds during these flights. Because oftheir secretive behaviors at inland sites and thedifficulty of locating their nests in large trees,multiple methods have been developed to studymurrelets including audio-visual (AV) and radarsurveys. The current ground-based Inland ForestSurvey Protocol (IFSP) for Marbled Murreletsdepends on the use of AV cues to detect birds inflight (Evans Mack et al. 2003). Collectinginformation on murrelets in this manner is difficult,because of the low light conditions during theirdawn and dusk peaks in inland activity and theirsmall size, cryptic coloration, rapid flight speed,and habitat preference for old-growth, closedcanopy forests. Further, because 85% of murreletdetections are auditory only (Paton et al. 1990) andmurrelets can call with varying frequency whenflying over survey stands, it is difficult todetermine the actual number of birds flying over asurvey area.

An alternate method, radar sampling, hasproven to be an excellent tool for observingMarbled Murrelets without some of the biasesassociated with AV surveys (Hamer et al. 1995;Burger 1997; Cooper and Blaha 2002; Bigger et al.2006). The main advantages of using radar forinventorying murrelets are that it works under alllight conditions and some inclement weather (fog,

light drizzle, but not rain); does not have theauditory bias of AV surveys; rapidly determinesmovement patterns; detects birds at low-use siteswhere they often are missed by AV observers(Cooper and Blaha 2002); focus AV efforts towardspecific areas of activity; samples a larger area (upto 1,500-m radius) than AV observers (up to a200-m radius). The major limitations of thetechnique are that it cannot be used in all standsbecause mountainous terrain creates radarshadows; and species identification errors arepossible. Radar sampling was selected for thestudy of Marbled Murrelets at the Project becauseof these advantages over AV surveys.

RES America Developments, Inc. (RES)proposes to develop the Skookumchuck WindEnergy Project (hereafter Project) in Lewis andThurston counties in western Washington (Figure1). There is potential Marbled Murrelet nestinghabitat on lands located inland from the Project andhistorical records of murrelet nesting adjacent tothe Project. Therefore, RES contracted ABR, Inc.(ABR) to conduct radar and AV studies during themurrelet breeding period in summer 2013 and 2014to determine if Marbled Murrelets fly over theproposed Project while traveling between thePacific Ocean and nesting locations further inlandand to assess potential collision risk to these birdsat the Project

OBJECTIVESThe primary goal of this study was to quantify

and characterize flight patterns of MarbledMurrelets at the Project and assess potential risk ofmurrelet collision fatalities at proposed turbines.The specific objectives were to collect informationon the number, flight paths, and flight altitudes ofMarbled Murrelets flying over the proposedProject during the summers of 2013 and 2014 anduse those data to (1) calculate an exposure rateestimating the frequency that murrelets would passwithin the airspace occupied by the proposedturbines and (2) estimate the annual number ofpotential collision fatalities.

STUDY AREA

The Project is located in western Washington~20–30 km (12.4–18.6 mi) east of Centralia andthe Interstate 5 corridor (Figure 1). The Project

Methods

Skookumchuck Marbled Murrelet Radar Study 2

consists of 4 different parcels that combined total~7,954 ha (19,654 ac) and are situated entirely onthe Vail Tree Farm, private land owned andmanaged by the Weyerhaeuser Company(WEYCO) for timber production (Figure 2).ABR focused survey efforts for this study on thetwo larger parcels of 4,088 ha and 2,049 ha. TheProject ridges range in elevation from ~450–1,050m above sea level (asl) and are separated bylower elevation stream-lined valleys, with theSkookumchuck River bisecting the project area.

The exact size of the development, includingthe number, megawatt (MW) capacity, and specificwind turbine model will be determined closer tothe time of construction; however, the currentproject design consists of 52 wind turbines with acombined generating capacity of up to 104 MW.Characteristics of the curent proposed wind

turbines, Vestas V110 2.0 MW turbines, include amonopole tower 80 m in height and three rotorblades each extending from a central hub with aradius of 55 m. Thus, the total maximal height ofeach turbine would be 135 m with a blade in thevertical position.

METHODS

STUDY DESIGNWe used marine radars and binoculars to

collect data on the passage rates, flight paths, flightbehaviors, and flight altitudes of MarbledMurrelets flying over the Project during the 2013and 2014 murrelet breeding seasons. Radar surveysfollowed protocols developed in previous studies(Cooper et al. 2001, 2006; Cooper and Blaha 2002;Raphael et al. 2002) and were also consistent with

Figure 1. Map of the proposed Skookumchuck Wind Energy Project in Lewis and Thurston counties, Washington.

Methods


recommendations for radar studies of murrelets inthe current Marbled Murrelet Inland Forest SurveyProtocol (IFSP; Evans Mack et al. 2003). The U.S.Fish and Wildlife Service (USFWS) is developingguidelines for radar studies of murrelets at windenergy developments. Although the guidelinescurrently are in draft form, we used them to helpdesign our studies, resulting in a number of notablechanges from our previous radar studies ofmurrelets at other locations. These changesincluded increased sampling intensity with a largernumber of sampling stations and visits per station,a dedicated visual observer in addition to the radaroperator, and the incorporation of additionalcorrection factors based on the best available datato adjust daily and seasonal passage rates.

We conducted radar observations for 5–7mornings at each of ten sampling stations frommid-May to early August each year (Tables 1 and2). Note that the number of sampling days perstation increased from 5 d in 2013 to 7 d in 2014 toreflect revisions in the draft USFWS surveyguidelines. We collected data during May to earlyAugust because that covers the period whenmurrelets are most active at inland sites andtherefore maximized our ability to detect murrelets(Cooper and Hamer 2003) while also minimizingcontamination of the data from migratory birds inthe spring and fall (Sanzenbacher and Cooper2008, Cooper and Mabee 2010). Radar surveysoccurred during the known peak of daily activityfor Marbled Murrelets, from 105 min before

Figure 2. Locations of project lease areas and survey stations at the proposed Skookumchuck Wind Energy Project in Lewis and Thurston counties, Washington.

Methods


sunrise to 75 min after sunrise (Burger 1997;Cooper and Hamer 2003; Cooper and Blaha 2002).

RADAR EQUIPMENTOur mobile radar setup consisted of two

marine radars mounted on a vehicle (Figure 3).One radar was operated in surveillance mode andscanned a 1.5 km radius area to record flight paths,passage rates, and flight speeds of seabird targets.The other radar was operated in vertical mode toscan an area roughly parallel with the proposeturbine string and measure flight altitudes oftargets. Figure 4 shows the approximate samplingairspace for the Furuno FR-1510 marine radar in a)surveillance and b) vertical mode at the 1.5-kmrange setting, as determined by field trials withRock Pigeons (Columba livia; Cooper et al. 2006),which are similar in size to murrelets.

The radar units were Furuno FR-1510 MKIIIs(X-band, 9.410 GHz, 2-m waveguide, 12 kWoutput) operated at a sampling range of 1.5 kmwith the pulse length set at 0.07 microseconds. Inorder to minimize ground clutter (areas obscuredon radar by vegetation or surrounding topography)the surveillance radar antenna was tilted upward at~10°, and we parked the vehicle in locations thatwere surrounded by low trees or low hills, a naturalradar fence (Eastwood 1967, Williams et al. 1972,and Cooper et al. 1991). To ensure that our radarunits perform to specifications, all ABR radars areperiodically maintained and tested by licensed

Furuno radar dealers. In addition, ABR tunes allradars seasonally and performs side-by-sidecomparisons annually to insure that all units collectcomparable information.

DATA COLLECTIONDuring each survey morning one observer

operated the surveillance and vertical radarsconcurrently while another observer conducted AVobservations within 50−100 m of the radar lab. Weentered all data into a Microsoft Excel workbookand checked data files for errors after eachsurvey. At the start of each survey we collectedenvironmental and weather data, includinginformation on wind speed and direction, cloudcover, ceiling height, visibility, temperature, andprecipitation. We used a Kestrel weatherinstrument to collect wind speed and temperaturedata (Nielsen-Kellerman, Birmingham, MI) and ahand-held compass to determine wind direction.

For each radar target that met the selectioncriteria for a murrelet target (see TargetIdentification below), we recorded the followingkey data: date; time; species, number of birds, andflight altitude (if identified by AV observer); flightdirection; flight behavior; flight path; velocity; anddistance from the radar. We recorded all radarimages with frame grabbers (Epiphan SystemsInc., Ottawa, ON) for review and archiving ofmurrelet targets and post-processing of flightaltitudes. Following surveys we used the timing,

Table 1. Location of sampling stations at the proposed Skookumchuck Wind Energy Project, Lewis and Thurston counties, Washington, during summer 2013 and 2014.

Latitude/longitude coordinates (WGS 84) Elevation (m above sea level) Station Easting Northing

1 122.57667 46.80674 563 2 122.55436 46.79197 649 3 122.53415 46.76403 833 4 122.51192 46.74569 756 5 122.66650 46.75219 570 6 122.63960 46.73961 767 7 122.59756 46.72499 730 8 122.57534 46.69175 850 9 122.54177 46.69756 870

10 122.51039 46.68846 1,027

Methods


Table 2. Landward and seaward Marbled murrelet targets observed on radar before sunrise and number of audio-visual detections of Marbled Murrelets (observed before or after sunrise) at the proposed Skookumchuck Wind Energy Project, Lewis and Thurston counties, Washington, during summer 2013 and 2014. Pre-sunrise murrelet targets

Survey year/ Radar station Surveys Landward1 Seaward1

Total

Audio-visual

detections

2013 1 5 1 3 4 0 2 5 1 1 2 0 3 5 0 1 1 0 4 5 0 3 3 0 5 5 0 2 2 0 6 5 2 0 2 0 7 5 0 2 2 0 8 5 0 0 0 0 9 5 2 1 3 0 10 5 4 3 7 0

Total: 50 10 16 26 0

2014 1 7 1 1 2 0 2 7 0 4 4 0 3 7 3 3 6 0 4 7 2 2 4 0 5 7 2 5 7 0 6 72 2 3 5 0 7 73 3 1 4 0 8 72,3 3 2 5 0 9 73 2 2 4 0 10 73

1 5 6 0 Total: 70 19 28 47 0

1 We classified targets as landward if they were flying 45 –224 and seaward if they were flying 225 –44 . 2 Unable to sample seaward targets during one morning in fall due to contamination from high numbers of avian migrants. 3 Unable to sample landward targets during one morning in spring due to contamination from high numbers of avian migrants.

Methods


direction, and distance of murrelet targets onsurveillance radar to locate these targets on therecorded vertical radar images and record flightaltitudes relative to the elevation of the radar lab.We then used digital elevation models to estimateflight altitudes of murrelet targets relative to wheremurrelet target flight paths crossed the ridgelineand proposed turbine strings.

RADAR DATA

Target IdentificationWe defined a Marbled Murrelet radar target

(hereafter murrelet target) as a radar echo thatrepresented one or more birds meeting the criteriadeveloped in previous studies (e.g., Hamer et al.1995, Cooper et al. 2001) including airspeed, targetsignature, flight characteristics, timing, and flightdirections of radar targets. All murrelet targets hadto meet or exceed a >64-km/h airspeed cut-off.

This approach has been shown to minimize thenumber of non-murrelet species and eliminateonly a small percentage (~3%) of MarbledMurrelets at sites within ~10 km of the coast(Cooper et al. 2001). Wind velocities anddirections used to calculate airspeed (i.e.,groundspeeds corrected for wind speed and relativedirection; Mabee et al. 2006) were measured froma nearby meteorological (met) tower in 2013 andSODAR instrument in 2014. Target signature (i.e.,size and appearance) was also used to distinguishMarbled Murrelets from other species. Forexample, Band-tailed Pigeons (Patagioenasfasciata) tend to travel in flocks and exhibit acharacteristic signature that is large and composedof multiple targets that repeatedly break apart, andthen coalesce. These distinct targets are easilydistinguished from a typical murrelet target and weeliminated them from our data set. Flight

Figure 3. Surveillance and vertical radar lab used for studies of Marbled Murrelets at the Skookumchuck Wind Energy Project in Lewis and Thurston counties, Washington, during summer 2013 and 2014.

Methods


Figure 4. Approximate murrelet-sampling airspace for the Furuno FR–1510 marine radar at the 1.5-km range setting, in a) surveillance and b) vertical antenna orientations, as determined by field trials with Rock Pigeons, which are similar in size to Marbled Murrelets. Note that the configuration of the radar beam within 250 m of the origin (i.e., the darkened area) was not determined.

b)

a)

Methods


characteristics, the overall pattern of how birdsmove across the landscape, can help todiscriminate among species and Marbled Murreletsgenerally exhibit a linear or arcing flight pathduring inland flights over non-habitat. Timing ofmovements also provides insight on speciesidentity and we did not analyze data collected aftersunrise because potential contamination from otherspecies (e.g., Band-tailed Pigeons) was too great;the majority of Marbled Murrelets fly into nestingstands before sunrise (Burger 1997, Cooper et al.2001) and a precedent for this method of datacollection has been set during previous radarstudies of murrelets (Burger 2001, Burger et al.2004, Sanzenbacher et al. 2014). We did, however,collect post-sunrise radar data in an attempt to getconcurrent AV observations of Marbled Murrelets.Flight directions, especially when considered inconjunction with time of year, are useful to

differentiate migratory birds from MarbledMurrelets. Based on the location of the Projectrelative to coastal areas and murrelet habitat furtherinland, we classified targets as landward if theirflight direction was between 45°−224 and seawardfor flight directions from 225°−44° (Cooper et al.2001, 2006; Sanzenbacher and Cooper 2008;Figure 5).

DATA ANALYSES

RADAR DATAWe used only pre-sunrise data to summarize

the number of murrelet targets observed duringeach survey morning and to compute the averagemorning murrelet passage rate (landward +seaward targets/d) for each sampling period. Forthe purpose of estimating fatality rates (see below)we also utilized correction factors to account for

Figure 5. Map indicating Marbled Murrelet target flight directions designated as landward (45°–224°)and seaward (225–44°) at the proposed Skookumchuck Wind Energy Project in Lewis and Thurston counties, Washington, during summer 2013 and 2014.

Results


post-sunrise murrelet targets and the flock size ofmurrelet targets (Table 3). Finally, we plotted theflight paths of all murrelet targets to assist in aqualitative assessment of potential flight corridorsand areas of concentrated activity.

We collected flight altitude data for all verticalradar targets that were concurrently confirmed asmurrelet targets on surveillance radar and provideflight altitudes relative to ground level wheretargets crossed ridgelines. We combined verticaltargets across years to calculate the averagemurrelet flight altitude and also the proportion oftargets that flew at or below the height of theproposed turbines ( 135 m above ground level[agl]). Data are presented as means ± StandardError (SE) unless otherwise noted.

MODELING FATALITY RATESWe have developed a model to estimate the

number of potential collision fatalities of murreletsat wind turbines as a risk-assessment tool. Theestimated number of collision fatalities (fatalityrate) is calculated as the product of: (1) theexposure rate (i.e., the number of times thatmurrelets might fly within the airspace occupied byturbines/d); (2) the avoidance probability (i.e., theprobability that a murrelet will detect and avoidentering the airspace containing the turbine); and(3) the fatality probability (i.e., the probability of afatal collision with a portion of the structure once amurrelet is in the airspace occupied by a turbine;Figure 6). The output from these calculations isthe unadjusted number of daily fatalities duringthe breeding period (Table 3). The final step is tomultiply this unadjusted daily fatality rate by theproduct of various adjustment factors (Table 4)that account for influential factors (e.g., number ofdays within each seasonal period, daytime andevening murrelet flights, and seasonal variation inmurrelet passage rates) not included in the originalmodel structure but important to provide a morerealistic estimate of fatalities. Fatality rates areestimated per turbine and per wind farm duringdifferent seasonal periods and across the entireyear (Table 5).

A detailed description of the fatality modelincluding its origin, review, use for other birdspecies and human-made structures, the modelstructure, adjustment factors, and the rationale

(scientific studies) used to justify the assumptionsused in the model are provided in Appendix 1.

ESTIMATING THE LIKELIHOOD OF FATALITY LEVELS

As an aid in interpreting the magnitude ofour adjusted fatality rates, we used a Poissondistribution (Rice 1995) to estimate the probabilitythat 1, 2, 3, etc. murrelets would be killed eachyear. The Poisson distribution describes thenumber of times a very rare occurrence happens ina large number of trials. If the expected number of fatalities per year (i.e., adjusted fatality rate)is , then the probability that there areexactly k fatalities (k = 0, 1, 2, ...10) is equal to:

,

wheree is the base of the natural logarithm (e =2.71828...)k is the number of murrelet fatalities each year—the probability of which is given by the functionk! is the factorial of k is a positive real number, equal to the expected

number of murrelet fatalities per year (i.e., adjustedfatality rate)

RESULTS

We conducted radar and AV observations at10 stations distributed along the 2 main ridgelinesproposed for development (Figure 2). The areagenerally was well-suited for radar and AVobservations and the 10 stations provided nearcomplete radar coverage of the ridgeline andproposed turbine strings at the Project. Weconducted a total of 50 surveys at the Projectduring May 24–August 1, 2013 and 70 surveysduring May 11–August 4, 2014 (Table 2). We alsoconducted an additional 4 surveys in 2013 and 9surveys in 2014 that were compromised because ofrain and therefore excluded from analyses.

PASSAGE RATESIn total, we observed 26 murrelet targets (i.e.,

pre-sunrise landward and seaward targets) in 2013

( )!

;kekfk λλλ

−

= ,

Results


Table 3. Estimated average daily exposure rates and daily fatality rates of Marbled Murrelets during the main breeding period at wind turbines at the proposed Skookumchuck Wind Energy Project, Lewis and Thurston counties, Washington. Rates based on radar data collected in summer 2013 and 2014. Values of particular importance are in boxes.

Variable/parameter for Vestas V110 2.0 MW TURBINE with 80 m hub height, 54 m blade lengths Side approach Frontal approach

PASSAGE RATE (PR) A) Mean pre-sunrise passage rate during breeding period (mid-May to early-August) 0.61 0.61 A1) Mean pre-sunrise landward targets/d 0.24 0.24 A2) Mean pre-sunrise seaward targets/d 0.37 0.37 B) Correction for post-sunrise targets (targets/day) = (A1/0.924) + (A2/0.380) 1.23 1.23 C) Mean number of birds/target 1.50 1.50 D) Daily passage rate (bird passes/day =B*C) 1.84 1.84 E) Fatality domain (days) 1 1

HORIZONTAL INTERACTION PROBABILITY (HIP) F) Turbine tower height (m) 80.0 80.0 G) Average turbine tower diameter (m) 5 5 H) Nacelle length (m) 10.4 10.4 I) Nacelle height (m) 4 4 J) Nacelle width (m) 3.9 3.9 K) Rotor and blade radius (m) 55 55 L) Average blade width (m) 3 3 M) Average blade thickness (front-to-back [m]) 2 2 N) Bird buffer zone (half of a murrelet wingspan, 0.205 m) 0.21 0.21 O) Minimal side profile area (m2) = (F*(G+(2*n))+((H+(2*N))*(I+(2*N)))+((2*(K+N))*(M+(2*N))) 761.69 P) Maximal font profile area (m2) = ((F-K)*(G+(2*N)))+ ( x (K+N)²) 9709.57 Q) Cross-sectional sampling area of radar at or below 135 m turbine height (= 3000 m * 135 m = 405,000 m²) 405,000 405,000 R) Minimal horizontal interaction probability (= O/Q) 0.00188 S) Maximal horizontal interaction probability (= P/Q) 0.02397

VERTICAL INTERACTION PROBABILITY (VIP) T) Proportion of murrelets flying turbine height (135 m above ground level) 0.38 0.38

EXPOSURE RATE (ER = PR*HIP*VIP) U) Daily exposure rate (bird passes/turbine/day = D*(M or N)*O, rounded to 4 decimal places) 0.0013 0.0168

FATALITY PROBABILITY (FP) V) Proportion of time turbines expected to operate during the morning activity period (0300-0800 h) 0.8256 0.8256 W) Probability of striking turbine if in airspace on a side approach 1.00 1.00 X) Probability of striking turbine if in airspace on frontal approach (turbine not in operation) 0.1059 0.1059 Y) Probability of striking turbine if in airspace on frontal approach (turbine in operation) 0.1625 0.1625 Z) Probability of fatality if striking turbine 1.00 1.00 AA1) Probability of fatality if an interaction on side approach and turbine not in operation (= (1-V)*W*Z) 0.1744

Results


Table 3. Continued.

Variable/parameter for Vestas V110 2.0 MW TURBINE with 80 m hub height, 54 m blade lengths Side approach Frontal approach

AA2) Probability of fatality if an interaction on side approach and turbine is in operation (= V*W*Z) 0.8257 AB1) Probability of fatality if an interaction on frontal approach and turbine not in operation (= (1-V)*X*Z) 0.0185 AB2) Probability of fatality if an interaction on frontal approach and turbine is in operation (= V*Y*Z) 0.1342

UNADJUSTED FATALITY RATE (= ER*FP) Daily fatality rate with 90% exhibiting collision avoidance (birds/turbine/day = P*(U or V)*0.1) 0.00013 0.00026 Daily fatality rate with 95% exhibiting collision avoidance (birds/turbine/day = P*(U or V)*0.05) 0.00006 0.00013 Daily fatality rate with 99% exhibiting collision avoidance (birds/turbine/day = P*(U or V)*0.01) 0.00001 0.00002

Figure 6. Major variables used in estimating possible fatality of Marbled Murrelets at wind turbines at the proposed Skookumchuck Wind Energy Project, Washington, during summer 2013 and 2014. See Tables 3 and 4 for details on calculations.

Passage rate(birds/time period)

Exposure rate(birds/time period)

Fatality rate(birds/time period)

Fatalityprobability

Interactionprobability

Avoidanceprobability

Results


Tabl

e 4.

A

djus

tmen

t fac

tors

for m

odel

ing

of se

ason

al a

nd a

nnua

l Mar

bled

Mur

rele

t fat

aliti

es p

er tu

rbin

e at

the

prop

osed

Sko

okum

chuc

k W

ind

Ener

gy P

roje

ct, L

ewis

and

Thu

rsto

n co

untie

s, W

ashi

ngto

n.

Sp

ring

trans

ition

Mai

n br

eedi

ng p

erio

d

Fall

trans

ition

Fa

ll m

olt

N

onbr

eedi

ng

Adj

ustm

ent f

acto

rs

M

arch

to

early

Apr

il

Apr

il M

ay

June

Ju

ly

early

A

ug

Mid

-Aug

to

mid

-Sep

t

mid

-Sep

t to

Oct

Nov

to F

eb

Num

ber o

f day

s per

per

iod

34

27

31

30

31

15

31

46

120

Day

time

fligh

t pro

porti

on1

1.00

1.

00

1.00

1.

00

1.86

1.

00

1.00

1.

00

1.00

Ev

enin

g fli

ght p

ropo

rtion

1 1.

00

1.00

1.

00

1.00

1.

29

1.00

1.

00

1.00

1.

00

Prop

ortio

n of

flig

hts o

utsi

de

mai

n br

eedi

ng p

erio

d 0.

20

1.00

1.

00

1.00

1.

00

1.00

0.

31

0.07

0.

24

Cum

ulat

ive

adju

stm

ent f

acto

r 6.

82

27.0

0 31

.00

30.0

0 73

.97

15.0

0 9.

57

3.07

28

.80

1 Adj

ustm

ent f

acto

r app

lied

to a

sing

le m

onth

(Jul

y) in

the

mod

el to

repr

esen

t the

chi

ck p

rovi

sion

ing

perio

d w

hen

dayt

ime

and

even

ing

inla

nd fl

ight

s occ

ur.

Results


Tabl

e 5.

Sum

mar

y of

per

turb

ine

fata

lity

rate

s and

cum

ulat

ive

fata

lity

rate

s for

Mar

bled

Mur

rele

ts a

t win

d tu

rbin

es d

urin

g se

ason

al p

erio

ds a

nd

acro

ss th

e en

tire

year

at t

he p

ropo

sed

Skoo

kum

chuc

k W

ind

Ener

gy P

roje

ct, L

ewis

and

Thu

rsto

n co

untie

s, W

ashi

ngto

n.

Perio

d

Fa

talit

y ra

te/tu

rbin

e

Proj

ect f

atal

ity ra

te

(acr

oss a

ll tu

rbin

es)

Avo

idan

ce

fact

or (%

) Si

de

appr

oach

Fr

onta

l ap

proa

ch

No.

turb

ines

Si

de

appr

oach

Fr

onta

l ap

proa

ch

Sprin

g tra

nsiti

on

(Mar

ch–e

arly

Apr

il)

90

0.00

09

0.00

18

52

0.04

61

0.09

22

95

0.

0004

0.

0009

52

0.

0213

0.

0461

99

0.00

01

0.00

01

52

0.00

35

0.00

71

Mai

n br

eedi

ng p

erio

d (A

pril–

early

Aug

ust)

90

0.02

47

0.04

94

52

1.28

32

2.56

65

95

0.

0114

0.

0247

52

0.

5923

1.

2832

99

0.00

19

0.00

38

52

0.09

87

0.19

74

Fall

trans

ition

(m

id-A

ugus

t–m

id-S

epte

mbe

r)

90

0.00

12

0.00

25

52

0.06

47

0.12

93

95

0.

0006

0.

0012

52

0.

0298

0.

0647

99

0.00

01

0.00

02

52

0.00

50

0.00

99

Fall

mol

t (m

id-S

epte

mbe

r–O

ctob

er)

90

0.00

04

0.00

08

52

0.02

07

0.04

15

95

0.

0002

0.

0004

52

0.

0096

0.

0207

99

<0.0

001

0.00

01

52

0.00

16

0.00

32

Non

bree

ding

(N

ovem

ber–

Febr

uary

) 90

0.

0037

0.

0075

52

0.

1947

0.

3894

95

0.00

17

0.00

37

52

0.08

99

0.19

47

99

0.

0003

0.

0006

52

0.

0150

0.

0300

Ann

ual f

atal

ity ra

te

(Jan

uary

–Dec

embe

r)

90

0.03

10

0.06

19

52

1.60

94

3.21

89

95

0.

0143

0.

0310

52

0.

7428

1.

6094

99

0.00

24

0.00

48

52

0.12

38

0.24

76

Results


and 47 in 2014 (Table 2; Figure 7). We detected atleast one murrelet target at all stations exceptstation 8 in 2013 and at all stations in 2014. Thenumber of murrelet targets per station ranged from0–7 in 2013 and from 2–7 in 2014. This equates toa murrelet passage rate (average number ofpre-sunrise landward + seaward targets/d) of 0.52 ±0.11 targets/d in 2013 (Appendix 2) and 0.70 ±0.06 targets/d in 2014 (Appendix 3). The overallpassage rate averaged across both seasons was 0.61± 0.09 target/d (Table 3). We did not detect anymurrelets during AV observations in 2013 or 2014;however, we did identify the presence ofBand-tailed Pigeons and Common Ravens at theProject, species that in some cases can appearsimilar to murrelet targets, but generally occurpost-sunrise.

A qualitative assessment of murrelet targetflight paths indicated that the highest levels ofactivity occurred at Station 10 (Figure 7), whichhad the highest number of murrelet targets in 2013and second highest number of murrelet targets in2014 (Table 2). Flight directions of murrelet targetswere variable at each station, but the combinedstation data each year resulted in 38%landward-flying targets and 62% seaward-flyingtargets in 2013 and 40% landward-flying targetsand 60% seaward- flying targets in 2014.

FLIGHT ALTITUDESFor murrelet targets observed concurrently on

surveillance radar and vertical radar we collectedflight altitudes of 6 murrelet targets in 2013 and 15murrelet targets in 2014. The average flight altitude

Figure 7. Map showing the flight paths of Marbled Murrelet radar targets heading in landward and seaward directions observed before sunrise at each of ten radar sampling stations at the proposed Skookumchuck Wind Energy Project, Lewis and Thurston counties, Washington,

Discussion


of all murrelet targets combined, relative to theProject ridgelines, was 219.3 ± 34.6 m agl. Basedon these flight altitudes, 38.1% (8 of 21 targets) ofmurrelet targets identified on vertical radar flewbelow the height of the proposed turbines (i.e.,

135 m agl).

FATALITY RATESAnnual fatality rates, our ultimate and focal

metric for assessing risk at the Project are producedby a combination of factors (Figure 6) and anadditional layer of adjustment factors, describedbelow, to account for the time murrelets are activeoutside the morning activity period and thebreeding season. The exposure rate (birds/turbine/d) was estimated by taking the murreletpassage rate; adjusting it for post-sunrise targetsand flock size; and accounting for both thehorizontal interaction probability (physicalcharacteristics of the turbines and the radarsampling area) and vertical interaction probability(% of murrelets flying turbine height; see Table3). The resulting daily exposure rate during thebreeding period was 0.0013 murrelet passes/turbine/d for a side approach and 0.0168 murreletpasses/turbine/d for a frontal approach (Table 3).

The next step was to adjust the exposure ratefor the fatality probabilities: % of time turbinesexpected to operate during the morning;probability of striking a turbine (with turbinesoperating/not operating); probability of fatality ifstriking turbine (with turbines operating/notoperating); and the avoidance probability (assumed90−99%) to derive the unadjusted fatality rate(birds/turbine/d; Table 3). The fatality rate in Table3 is specific to the morning activity period duringthe breeding season (when we conducted surveys)and hence is considered unadjusted because thisrate has not been adjusted for time outside theseperiods. The unadjusted fatality rate (murreletpasses/turbine/d) based on a range of avoidancefactors (90–99%) ranged from 0.00001−0.00026(Table 3).

Our ultimate metric, the annual fatality rate atthe Project, was 0.0024–0.0619 murrelet fatalities/turbine/yr and 0.1238–3.2189 murrelet fatalities/yracross all 52 turbines at the Project (Table 5).Estimates of collision fatalities varied by approach(i.e., side versus frontal) and avoidance rate.

LIKELIHOOD OF FATALITIESAn alternate way of examining the likelihood

of murrelet collisions with wind turbines at theProject is to use a Poisson distribution to estimatethe probability that 1, 2, 3, etc. murrelets would bekilled each year based on the annual fatality rate,frontal versus side approach, and a range ofavoidance factors (Table 6). Across the full rangeof avoidance rates (i.e., 90−99%) and approachesthe probability of 0 murrelet collisions/yr rangedfrom ~5–86% (0.0496–0.8569), whereas theprobability of 1 murrelet collision/yr ranged from~13–36% (0.1323–0.3567).

DISCUSSION

Understanding the relative abundance andmovement patterns of murrelet targets observed inthis study requires a broader understanding of theconservation status of Marbled Murrelets at aregional and local scale. In the Pacific Northwest,the Northwest Forest Plan and Marbled MurreletRecovery Plan include both population andhabitat-based monitoring programs for theMarbled Murrelet (USFWS 1997, Davis et al.2011). In this region, murrelet populations aredivided into five different Conservation Zoneswhere monitoring efforts focus on counting birdsin coastal waters, because murrelets are moredifficult to detect at inland sites (Miller et al.2012). The Marbled Murrelets that fly over theproposed Skookumchuck Wind Energy Projectwhen traveling to inland nesting habitat probablycome from populations in Zone 1, the Strait of Juande Fuca, Puget Sound, Hood Canal and the SanJuan Islands; and/or Zone 2, the Washington outercoast (Figure 8). At-sea survey data from 2013resulted in Marbled Murrelet population estimatesof 4,395 (95% confidence interval [CI] =2,276–6,740) in Zone 1 and 1,257 (95% CI =920–1,846) in Zone 2 (Pearson et al. 2014). Theestimated annual rate of decline for the 2001−2013period was -3.88 ± 1.73% in Zone 1 and -7.37 ±1.70% in Zone 2 (Pearson et al. 2014). Similarly,an analysis of changes in murrelet nesting habitatin the Pacific Northwest found a net loss of ~7% ofhigh quality nesting habitat from 1994−2007,primarily resulting from fire on federal lands andtimber harvest on private lands (Raphael et al.2011).

Discussion


Tabl

e 6.

The

estim

ated

pro

babi

lity

of d

iffer

ent a

nnua

l fat

aliti

es o

f Mar

bled

Mur

rele

ts a

t all

52 w

ind

turb

ines

at t

he p

ropo

sed

Skoo

kum

chuc

k W

ind

Ener

gy P

roje

ct, L

ewis

and

Thu

rsto

n co

untie

s, W

ashi

ngto

n.

90

% a

void

ance

95%

avo

idan

ce

99

% a

void

ance

Num

ber o

f fat

aliti

es

Side

app

roac

h Fr

onta

l ap

proa

ch

Si

de a

ppro

ach

Fron

tal

appr

oach

Side

app

roac

h Fr

onta

l ap

proa

ch

0 0.

2135

0.

0496

0.

4620

0.

2227

0.

8569

0.

7405

1

0.32

97

0.14

89

0.35

67

0.33

45

0.13

23

0.22

25

2 0.

2545

0.

2237

0.

1377

0.

2512

0.

0102

0.

0334

3

0.13

10

0.22

40

0.03

54

0.12

58

0.00

05

0.00

33

4 0.

0506

0.

1683

0.

0068

0.

0472

<0

.000

1 0.

0003

5

0.01

56

0.10

11

0.00

11

0.01

42

<0.0

001

<0.0

001

6 0.

0040

0.

0506

0.

0001

0.

0036

<0

.000

1 <0

.000

1 7

0.00

09

0.02

17

<0.0

001

0.00

08

<0.0

001

<0.0

001

8 0.

0002

0.

0082

<0

.000

1 0.

0001

<0

.000

1 <0

.000

1 9

<0.0

001

0.00

27

<0.0

001

<0.0

001

<0.0

001

<0.0

001

10

<0.0

001

0.00

08

<0.0

001

<0.0

001

<0.0

001

<0.0

001

Discussion


At a more local scale, the WashingtonDepartment of Natural Resources coordinatedefforts by a team of experts that ranked the value of17 blocks of state land (planning blocks) insouthwestern Washington for consideration aslong-term Marbled Murrelet management areas(Raphael et al. 2008). The closest planning block tothe Project, the Skookumchuck block (Figure 9),was ranked as low priority with the second lowestranking of all 17 blocks, based on habitatcharacteristic and the lack of known occupiednest sites. As a result of this low ranking, theSkookumchuck block was not recommended forconservation emphasis in the final report (Raphaelet al. 2008).

At the Project scale, information fromWEYCO and a literature review by ABR haveconcluded that there are no known nesting

locations of Marbled Murrelets in the Project(Sanzenbacher et al. 2011). A number of landsections, however, with documented historicaloccupied murrelet nesting habitat occur adjacent tothe Project (Sanzenbacher et al. 2011). Detectionsof murrelets in these nesting areas date back to1995−1997, but we are unaware of any more recentinformation on the status of murrelets at these sites.The relatively low passage rates of murrelets overthe Project during the current study in 2013 and2014 were consistent with the lack of knownmurrelet nesting within the Project, but thelikelihood that suitable murrelet nesting habitatexists inland from the project either on privatelands or National Forest Service lands (Figure 1).

Predictions of the effects of wind powerdevelopment on Marbled Murrelets and other birdsare hampered by both a lack of detailed knowledge

Figure 8. Location of Marbled Murrelet Conservation Zones 1 and 2 and the inland breeding limit of the Marbled Murrelet in Washington state. Adapted from USFWS (1997).

Discussion


Figure 9. Geographic planning blocks evaluated for designation as Marbled Murrelet Management Areas. Adapted from Raphael et al. (2008).

Discussion


about patterns of movement and behavior of birdsaround wind turbines and by the fact that theprecise relationship between bird abundance andbird fatalities at wind turbines is unknown. In thisstudy, we documented passage rates of MarbledMurrelets at the Project to describe their movementpatterns and to estimate the annual number ofcollision fatalities.

TARGET IDENTIFICATIONOne of the limitations of ornithological radar

is that it usually is difficult to identify radar targetsto species solely by flight characteristics.Identification to the species level is possible onlyif that species has flight characteristics and/or aunique timing of movements for a particularlocation. The Marbled Murrelet is one of the fewrapid-flying species present at some sites in thePacific Northwest that is active in the earliest partof the morning and thus has been successfullyidentified on radar at those locations (Hamer et al.1995; Burger 1997; Cooper et al. 2000, 2001).During the current study at the Project we did nothave any visual verification of the small number ofmurrelet targets detected by the radar and thus wewere unable to calculate an accuracy rate for radaridentification of murrelet radar targets. Weacknowledge that our Project radar data mayinclude contamination from non-murrelet targets;however, for the purposes of this study we assumedthat all pre-sunrise landward and seaward targetsmeeting all of our selection criteria were MarbledMurrelets because their airspeed, target signature,flight characteristics, timing, and flight directionswere consistent with known murrelet targets fromother studies.

PASSAGE RATESThe mean passage rate of murrelet targets

over the Project during the summers of 2013 and2014 was lower than passage rates observed at sitesadjacent to high quality nesting habitat inWashington, Oregon, and California. For instance,we documented a mean passage rate of 0.26landward targets/morning (corrected for post-sunrise movements but not flock size) at theProject, whereas the same metric observed at 12valleys in the Olympic Peninsula ranged from~30–150 landward targets/morning during

1996–2004 (Cooper et al. 2001, 2006). Similarly,at 30 sites in Oregon, most had counts of ~10−50landward targets/morning (range = 2–80 landwardtargets/morning; Cooper et al. 2000, Cooper andAugenfeld 2001). Counts at 10 sites in northernCalifornia averaged 42 landward targets/morningin 2003 and 54 landward targets/morning in 2004,with a range of 4–170 landward targets/morning(Cooper et al. 2005).

In contrast to these sites with high radarpassage rates, murrelet passage rates were muchlower at the Project and other proposed winddevelopment sites in Washington and California.For instance, the mean landward + seawardmurrelet passage rate (corrected for post-sunrisemovements but not flock size) at the Project was1.23 targets/morning compared to 1.59 targets/morning at the proposed Coyote Crest WindPower Project, located ~50 km to the west of theProject (Cooper and Mabee 2010). Similarly, thesame passage rate metric at the Radar Ridge site,~90 km to the southwest of the Project, was 1.89targets/morning, but the radar methods differed atthis site with results based on data collectedpre-sunrise and 30 mins post-sunrise (HamerEnvironmental 2009). In summary, the relativelylow passage rates of murrelets observed at theProject during both 2013 and 2014 were consistentwith the lack of murrelet nesting habitat at theProject. Presumably the murrelet targets detectedon radar were flying over the Project en route topotential nesting habitat on private or public landslocated further inland from the Project.

FLIGHT ALTITUDESThe vertical radar data from the Project

indicated that the majority of murrelet targets(62%) flew above the proposed turbine heights atthe Project (i.e., >135 m agl) and the average flightaltitude of these murrelet targets was 219 ± 34 magl. For comparison, the average flight altitude ofmurrelet targets at the Coyote Crest Wind PowerProject was 290 ± 68 m agl (n = 3 targets; Cooperand Mabee 2010) and for the Radar Ridge WindResource area was 307.9 ± 29.4 m agl (n = 36targets; Hamer Environmental 2009).

We chose to use site-specific flight altitudedata collected at the Project for our fatality models,because differences in flight altitudes among sites

Discussion


can occur. For example, among-site differences inflight altitudes of murrelets were observed at twovalleys in the Olympic Peninsula: the mean flightaltitude of murrelets was 246 ± 4.7 m agl (n = 282murrelets in three mornings) at a locationapproximately 6 km up the Queets Valley (Stumpfet al. 2011), compared with 142 ± 6 m agl (n = 309murrelets in eight mornings) at a locationapproximately 1.5 km up the Duckabush Valley(Cooper 2010). Similarly, in northern CaliforniaSanzenbacher et al. (2014) found that mean flightaltitudes differed between two sites, one locatedwithin 1 km of the coast (93 m agl and 98 m agl)and the other located 29 km inland (257 m agl).These comparisons provide evidence of among sitevariation in flight altitude of murrelets and furtherindicate that whenever possible site-specificinformation should be used to evaluate collisionrisk.

FATALITY ESTIMATESThe ultimate goal of this study was to assess

the potential risk of Marbled Murrelet collisionfatalities at the Project wind turbines. We usedinformation on passage rates of murrelets collectedover two years of study at the Project, combinedwith characteristics of the proposed development(e.g., number and dimensions of proposed windturbines) and a series of assumptions based on thelife history of Marbled Murrelets, to modelestimated annual collision fatalities. Collisionfatality estimates varied by approach (i.e., sideversus frontal) and avoidance rate (90–99%).Across the full range of avoidance rates the annualfatality rate at the Project was 0.0024–0.0619murrelet fatalities/turbine/yr and 0.1238–3.2189murrelet fatalities/yr at all 52 turbines combined.Based on a Poisson distribution that accounts forrare events over time the probability of zero annualmurrelet fatalities ranged from 5–86%, whereas theprobability of one annual murrelet fatality rangedfrom 13–36%.

AVOIDANCE RATESThe collision avoidance rate is a highly

influential factor in the model estimating collisionfatalities; however, no murrelet-specific data areavailable on the proportion of flights that do notresult in collisions with wind turbines because ofcollision-avoidance behavior. We know of only a

single study of murrelet avoidance behavior,which was at a transmission line and found zerofatalities, suggesting a high avoidance rate (Cooper2010). We agree with others (Chamberlain et al.2006, Fox et al. 2006) that species-specific,weather-specific, and site-specific avoidance dataare needed in models to increase the accuracy ofmurrelet fatality estimates. However, the currentlyavailable avoidance data from a number ofseabirds, while incomplete, is consistent with thenotion that substantial proportions of many speciesdetect and avoid large, human-made structures(Appendix 4). In particular, the ability to detect andavoid objects under low-light conditions makessense for Marbled Murrelets, which are adept atnavigating through forests near their nests duringlow light conditions and have good enough visualacuity to capture prey underwater at night (Nelson1997). Until murrelet-specific data on therelationship between exposure and fatality ratesare available for wind turbines, we continue to usea range of avoidance rates in our fatality models(i.e., 90%, 95%, and 99% avoidance) with thecurrent evidence suggesting that avoidance ratesare high (>95%). Assuming a collision avoidancerate of 99% the estimated annual fatality ratewas 0.0024–0.0048 fatalities/turbine/yr and0.1238–0.2476 murrelet fatalities/yr at all turbines;the probability of 0 murrelet collisions/yr rangedfrom 74–86% and the probability of 1 murreletcollision/yr ranged from 13–22%.

SOURCES OF VARIATION IN PASSAGE RATES AND FATALITY ESTIMATES

There are a number of factors that could affectmurrelet passage rates and fatality rate estimates.For instance, the inclusion of non-murrelet targetswould inflate our passage rate estimates, but weused murrelet targets observed only before sunrisealong with other target identification criteria formurrelets (e.g., flight speed, airspeed, targetsignature, flight characteristics, timing, and flightdirections) to help minimize the inclusion ofnon-murrelets in our data.

A statistical bias is that our model assumedthat passage rates of murrelets over the Projectwould not decline, even after a bird was killed (i.e.,we assumed sampling with replacement). Based ontwo summers of data collection there appears to be

Conclusions


a limited (but unknown) number of murrelets thatperiodically fly over the Project area. Assuming aclosed population (i.e., little or no mixing amonggeographic groupings of murrelets) and someflyway fidelity then the loss of even a single birdwould substantially reduce the average dailypassage rate. Thus, it is possible that our modelcould overestimate fatality rates over timefollowing a collision fatality. This issue deservesfurther investigation for future modeling efforts

Changes in ocean conditions, such as thosethat occur as the result of the El Niño–SouthernOscillation (ENSO) and the Pacific DecadalOscillation (PDO), have been linked to changes indiet, productivity, survival, and distribution ofMarbled Murrelets along the Pacific coast (Beckerand Beissinger 2003, Peery et al. 2006, Becker etal. 2007). During years with strong ENSO events itis likely that the widespread nesting failure ofMarbled Murrelets resulting from a depletion ofoceanic food resources also results in fewer inlandflights of murrelets, especially if the failure occursbefore the chick-rearing period. Further, there isevidence indicating that nonbreeding murrelets incentral California rarely fly inland during thebreeding season, which suggests that lowerradar-based counts should occur during years ofpoor breeding effort and that these counts arerepresentative indices of the potential breedingeffort in that area (Peery et al. 2004, Bigger et al.2006). Based on measurements of oceanographicconditions in 2013 and 2014 an ENSO event didnot occur prior to or during our studies andtherefore was not a factor affecting our radarcounts of murrelets at the Project.

Another factor that could cause interannualvariation in murrelet counts is population increasesor declines of murrelets. At-sea surveys conductedfrom 2001–2013 indicate a 4.65% annual declinein Marbled Murrelet populations in ConservationZones 1 and 2 combined (Pearson et al. 2014). Ifcontinued population monitoring substantiateslong-term population changes for this region thenfuture fatality modeling efforts may need toinclude adjustments based on the known rates ofdeclines.

CONCLUSIONS

We documented passage rates and flightaltitudes of Marbled Murrelets flying over theProject during radar and audiovisual surveys insummer 2013 and 2014. The passage rate ofmurrelet targets was low relative to sites in thePacific Northwest located near large areas ofmurrelet nesting habitat and was similar to passagerates from other studies at proposed wind energydevelopments in Washington (e.g., Coyote CrestWind Power Project). The low passage ratesobserved at the Project were consistent with thelack of murrelet nesting habitat at the Project andthe likelihood that some murrelets pass over theProject when traveling to nesting habitat furtherinland (e.g., National Forest Service lands to thesoutheast). Similar to other studies at inland sites inthe Pacific Northwest, the flight altitudes ofmurrelet targets at the Project indicated that amajority of murrelets (62%) flying over the Projectridgelines occurred above the height of theproposed turbines (>135 m agl). Based on theobserved passage rate and flight altitudes ofmurrelet targets at the Project there is a risk ofMarbled Murrelet collision fatalities at theproposed wind turbines, but with the assumption ofhigh collision avoidance rates by MarbledMurrelets (i.e., >95%) the estimated collision riskand associated fatality estimates were low.

LITERATURE CITED

Barbaree, B. A. 2012. Nesting season ecology ofMarbled Murrelets at a remote mainland fjordin southeast Alaska. MSc. Thesis, OregonState University, Corvallis, OR. 154 pp.

Becker, B. H., and S. R. Beissinger. 2003.Scale-dependent habitat selection by anearshore seabird, the Marbled Murrelet, in ahighly dynamic upwelling system. MarineEcology Progress Series 256: 243–255.

Becker, B. H., Z. M. Peery, and S. R. Beissinger.2007. Ocean climate and prey availabilityaffect the trophic level and reproductivesuccess of the Marbled Murrelet, anendangered seabird. Marine Ecology ProgressSeries 329: 267–279.

Literature Cited


Bigger, D., M. Z. Peery, J. Baldwin, S. Chinnici,and S. P. Courtney. 2006. Power to detecttrends in Marbled Murrelet breedingpopulations using audiovisual and radarsurveys. Journal of Wildlife Management 70:493–504.

Bloxton, Jr., T. D., and M. G. Raphael. 2008.Breeding ecology of the Marbled Murrelet inWashington State: Five year project summary(2004−2008). Unpublished report, USDAForest Service, Pacific Northwest ResearchStation, Olympia, WA. 41pp.

Bradley, R. W., F. Cooke, L. W. Lougheed, andW. S. Boyd. 2004. Inferring breeding successthrough radiotelemetry in the MarbledMurrelet. Journal of Wildlife Management 68:318−331.

Burger, A. E. 1997. Behavior and numbers ofMarbled Murrelets measured with radar.Journal of Field Ornithology 68: 208–223.

Burger, A. E. 2001. Using radar to estimatepopulations and assess habitat associations ofMarbled Murrelets. Journal of WildlifeManagement 65: 696–715.

Burger, A. E., T. A. Chatwin,, S. A. Culler, N. P.Holmes, I. A. Manley, M. H. Mather, B. K.Schroeder, J. D. Steventon, J. E. Duncan, P.Arcese, and E. Selak. 2004. Application ofradar surveys in the management ofnesting habitat of Marbled MurreletsBrachyrhamphus marmoratus. MarineOrnithology 32: 1–11.

California Department of Fish and Wildlife(CDFW). 2014. State and federally listedendangered and threatened animals ofCalifornia. http://www.dfg.ca.gov/biogeodata/cnddb/pdfs/TEAnimals.pdf. AccessedNovember 2014

Chamberlain, D. E., M. R. Rehfisch, A. D. Fox,M. Desholm, and S. J. Anthony. 2006. Theeffect of avoidance rates on bird mortalitypredictions made by wind turbine collisionrisk models. Ibis 148: 198–202.

Committee on the Status of Endangered Wildlife inCanada (COSEWIC). 2012. Listing of theconservation status of species at risk inCanada.

Cooper, B. A. 2010. Memorandum regarding fieldand data summary for Duckabush AvoidanceStudy (Attachment B of Appendix J). In:TetraTech EC. Final Environmental ImpactStatement for the Everpower Coyote CrestWind Park. Chehalis, WA: Lewis CountyCommunity Development.

Cooper, B. A., and K. H. Augenfeld. 2001. Radarsurveys for Marbled Murrelets in the ElliottState Forest, Oregon, 2001. Unpublishedreport prepared for Oregon Department ofForestry, Salem, OR, by ABR, Inc., ForestGrove, OR. 24 pp.

Cooper, B. A., and R. J. Blaha. 2002. Comparisonof radar and audio-visual counts of MarbledMurrelets during inland forest surveys.Wildlife Society Bulletin 30: 1182–1194.

Cooper, B. A., R. J. Blaha, and J. H. Plissner. 2005.Determining local population trends ofMarbled Murrelets in that portion ofConservation Zone Four North of the PalcoHCP Area. 2004. Unpublished. reportprepared for Pacific Lumber Company(Palco), U.S. Fish and Wildlife Service, andCalifornia Department of Fish and Game,Scotia, CA, by ABR, Inc., Forest Grove, OR.26 pp.

Cooper, B. A., and R. H. Day. 2004a. Results ofendangered bird and bat surveys at theproposed Kaheawa Pastures Wind EnergyFacility, Maui Island, Hawaii, fall 2004.Unpublished report prepared for KaheawaWindpower LLC, Makawao, HI, and UPCWind Management LLC, Newton, MA, byABR, Inc., Forest Grove, OR, and Fairbanks,AK. 16 pp.

Cooper, B. A., and R. H. Day. 2004b. Modelingannual seabird use and fatality at the proposedKaheawa Pastures Wind Energy Facility,Maui Island, Hawaii, fall 2004. Unpublishedreport prepared for Kaheawa Windpower

Literature Cited


LLC, Makawao, HI, and UPC WindManagement LLC, Newton, MA, by ABR,Inc., Forest Grove, OR, and Fairbanks, AK.7 pp.

Cooper, B. A., and R. H. Day. 2009. Radar andvisual studies of seabirds at the proposedKWP II down-road alternative wind energyfacility, Maui Island, Hawaii, summer 2009.Unpublished report prepared for FirstWind,LLC., Newton, MA by ABR, Inc.—Environmental Research & Services, ForestGrove, OR. 33 pp.

Cooper, B. A., R. H. Day, and J. H. Plissner. 2007.Radar and audio-visual studies of HawaiianPetrels near proposed meteorological towersand wind turbines on northwestern LanaiIsland, May−July 2007. Unpublished reportprepared for KC Environmental, Inc.,Makawao, HI, and TetraTech EC, Portland,OR, by ABR, Inc.—Environmental Research& Services, Forest Grove, OR, and Fairbanks,AK. 44 pp. Appendix 3 in Draft HabitatConservation Plan for the construction andoperation of the Lanai metereological towers,Lanai, Hawaii. Prepared for Castle andCooke, Lanai City, HI, by TetraTech EC,Honolulu, HI. 54 pp. + appendices.

Cooper, B. A., R. H. Day, R. J. Ritchie, and C. L.Cranor. 1991. An improved marine radarsystem for studies of bird migration. Journalof Field Ornithology 62: 367–377.

Cooper, B. A., and T. E. Hamer. 2003. Use of radarfor Marbled Murrelet Surveys, Appendix G.In Evans, D. R., W. P. Ritchie, S. K. Nelson,E. Kuo-Harrison, P. Harrison, and T. E. Hamer(eds.). Methods for surveying marbledmurrelets in forests: an update to the protocolfor land management and research. Publishedby the Pacific Seabird Group. 64 pp.http://www.pacificseabirdgroup.org/publications/mamuforms/mamuprotocol.html.Accessed November 2014.

Cooper, B. A., and T. J. Mabee. 2010. A radarstudy of marbled murrelets at the proposedCoyote Crest Wind Power Project, Summer2008 and 2009. Unpublished report prepared

for Tetra Tech EC, Portland, OR andEverpower Wind Holdings Inc. and Pe EllNorth LLC, Portland, OR, by ABR, Inc.,Forest Grove, OR. 34 pp.

Cooper, B. A., M. G. Raphael, and D. M. Evans.2001. Radar-based monitoring of marbledmurrelets. Condor 103: 219–229.

Cooper, B. A., M. G. Raphael, and M. Z. Peery.2006. Trends in radar-based counts ofMarbled Murrelets on the Olympic Peninsula,Washington, 1996–2004. Condor 108:936–947.

Cooper, B. A., P. M. Sanzenbacher, and R. H. Day.2011. Radar and visual studies of seabirds atthe proposed Kawailoa Wind Energy Facility,Oahu Island, Hawaii, 2009. Unpublishedreport prepared for First Wind LLC, Newton,MA by ABR, Inc., Forest Grove, OR, andFairbanks, AK. 32 pp. Appendix 3 inKawailoa Wind Power Draft HabitatConservation Plan. Prepared by SWCA,Honolulu, HI, for Kawailoa Wind PowerLLC, Honolulu, HI. 144 pp + appendices.

Cooper, B. A., C. Strong, and L. Folliard. 2000.Radar-based monitoring of Marbled Murreletsin Oregon, 1996–1999. Unpublished reportprepared for U.S. Fish and Wildlife Service,Portland, OR, by ABR, Inc., Forest Grove,OR. 46 pp.

Davis, R., G. Falxa, E. Grinspoon, G. Harris, S. H.Lanigan, M. Moeur, and S. Mohoric. 2011.Northwest Forest Plan the first 15 years[1994-2008]: monitoring the Northwest ForestPlan - fifteen year summary of key findings.Technical Paper R6-RPM-TP-03-2011.Portland, OR: U.S. Department ofAgriculture, Forest Service, Pacific NorthwestRegion.

Day, R. H., and B. A. Cooper. 1999. Results ofendangered bird and bat surveys at theproposed Kaheawa Pastures Windfarm onMaui Island, Hawaii, summer 1999.Unpublished report prepared for Zond Pacific,Wailuku, HI, by ABR, Inc., Fairbanks, AK,and Forest Grove, OR. 26 pp.

Literature Cited


Day, R. H., and B. A. Cooper. 2008. Results ofendangered seabird and Hawaiian hoary batsurveys on northern Oahu Island, Hawaii,October 2007 and July 2008. Unpublishedreport prepared for FirstWind, LLC, Newton,MA by ABR, Inc.—Environmental Researchand Services, Fairbanks, AK. 35 pp.

Day, R. H., A. K. Prichard, and J. R. Rose. 2005.Migration and collision avoidance of eidersand other birds at Northstar Island, Alaska,2001–2004. Unpublished final report preparedfor BP Exploration (Alaska), Inc., Anchorage,AK, by ABR, Inc.—Environmental Researchand Services, Fairbanks, AK. 142 pp.

Desholm, M., A. D. Fox, P. D. L. Beasley, and J.Kahlert. 2006. Remote techniques forcounting and estimating the number ofbird-wind turbine collisions at sea: a review.Ibis 148: 76–89.

Desholm, M. and J. Kahlert. 2005. Avian collisionrisk at an offshore windfarm. Biology Letters1: 296–298.

Dirksen, S. E., A. L. Spaans, and J. Winden. 1998.Nocturnal collision risks with wind turbines intidal and semi-offshore areas. Pages 99–108in Proceedings of International Workshop onWind Energy and Landscape, Genua, 26–27July 1997. Balkema, Rotterdam.

Eastwood, E. 1967. Radar ornithology. Methuenand Co., Ltd., London, UK. 278 pp.

Evans Mack, D. M., W. P. Ritchie, S. K. Nelson, E.Kuo-Harrison, P. Harrison, and T. E. Hamer.2003. Methods for surveying MarbledMurrelets in forests: a revised protocol forland management and research. PacificSeabird Group unpublished document.www.pacificseabirdgroup.org/publications/PSG_TechPub2_MAMU_ISP.pdf. AccessedNovember 2014.

Everaert, M. and E. W. M. Stienen. 2007. Impact ofwind turbines on birds in Zeebrugge(Belgium). Biodiversity and Conservation 16:2,259–3,345.

Fox, A. D., M. Desholm, J. Kahlert, T. K.Christensen, and I. K. Petersen. 2006.Information needs to support environmentalimpact assessment of the effects of Europeanmarine offshore wind farms on birds. Ibis 148:129–144.

Hamer Environmental. 2009. Use of radar todetermine passage rates and heightdistribution of Marbled Murrelets at theproposed Radar Ridge Wind Resource Area,Pacific County, Washington. UnpublishedFinal Report prepared for Energy Northwest,Richland, WA, by Hamer Environmental L.P.,Mount Vernon, WA. 54 pp.

Hamer, T. E., and S. K. Nelson. 1995. Nestingchronology of the Marbled Murrelet In Ralph,C. J., G. L. Hunt, J. Piatt, and M. Raphael(eds.). Conservation Assessment for theMarbled Murrelet. U.S.D.A. Forest Service,Redwood Sciences Laboratory, Arcata, CA.

Hamer, T. E., B. A. Cooper, and C. J. Ralph. 1995.Use of radar to study the movements ofMarbled Murrelets at inland sites.Northwestern Naturalist 76: 73–78.

Hebert, P. N., and R. T. Golightly. 2006.Movements, nesting, and response toanthropogenic disturbance of marbledmurrelets (Brachyramphus marmoratus) inRedwood National and State Parks,California. Unpublished report, Department ofWildlife, Humboldt State University, Arcata,CA and California Department of Fish andGame report 2006-02, Sacramento, CA.

Jones, P. H. 2001. The Marbled Murrelets of theCaren Range and Middlepoint Bight. WesternCanada Wilderness Committee, Vancouver,BC.

Mabee, T. J., J. H. Plissner, B. A. Cooper, and D. P.Young. 2006. Nocturnal bird migration overan Appalachian Ridge at a proposed windpower project. Wildlife Society Bulletin 34:682–690.

Literature Cited


Miller, S. L., M. G. Raphael, G. A. Falxa, C.Strong, J. Baldwin, T. Bloxton, B. M.Galleher, M. Lance, D. Lynch, S. F. Pearson,C. J. Ralph, and R. D. Young. 2012. Recentpopulation decline of the Marbled Murrelet inthe Pacific Northwest. Condor 114: 771−781.

Naslund, N. L. 1993. Why do Marbled Murreletsattend old-growth forest nesting areasyear-round? Auk 110: 594–602.

National Oceanic and Atmospheric Administration(NOAA). 2014. National Climatic DataCenter El Niño and La Niña website.http://www.ncdc.noaa.gov/teleconnections/enso/index.php. Accessed November 2014.

Nelson, S. K. 1997. Marbled Murrelet(Brachyramphus marmoratus). In A. Pooleand F. Gill, eds. The Birds of North America,No. 276. Academy of Natural Sciences,Philadelphia, PA, and AmericanOrnithologists’ Union, Washington, DC.32 pp.

Nelson, S. K. and T. E. Hamer. 1995. Nestingbiology and behavior of the MarbledMurrelet. Pages 57-68 in C. J. Ralph, G. L.Hunt, Jr., M. G. Raphael, and J. F. Piatt, eds.Ecology and conservation of the MarbledMurrelet. USDA Forest Service GeneralTechnical Report PSW-GTR-152, Albany,CA.

Oregon Department of Fish and Wildlife (ODFW).2014. Oregon list of threatened andendangered fish and wildlife species.http://www.dfw.state.or.us/wildlife/diversity/species/threatened_endangered_candidate_list.asp. Accessed November 2014.

Paton, P. W., C. J. Ralph, H. R. Carter, and S. K.Nelson. 1990. Surveying Marbled Murreletsat inland forest sites: a guide. USDA ForestService, Pacific Southwest ResearchLaboratory, Arcata, CA. Gen. Tech. Rep.PSW-120. 9 pp.

Pearson, S. F., M. M. Lance, and M. G. Raphael.2014. Washington 2013 at-sea marbledmurrelet population monitoring: ResearchProgress Report. Washington Department ofFish and Wildlife, Wildlife Science Divisionand USDA Forest Service Pacific NorthwestResearch Station, Olympia, WA.

Peery, M. Z., S. R. Beissinger, B. H. Becker, E.Burkett, and S. H. Newman. 2004. Inlandflight behavior of Marbled Murrelets:implications for population monitoring withradar. Condor 106: 344–353.

Peery, M. Z., S. R. Beissinger, E. Burkett, and S. H.Newman. 2006. Local survival rates ofMarbled Murrelets: roles of oceanographicconditions, sex, and radio-tagging. Journal ofWildlife Management 70: 78–88.

Raphael, M. G., S. K. Nelson, P. Swedeen, M.Ostwald, K. Flotlin, S. Desimone, S. Horton,P. Harrison, D. Prenzlow Escene, and W.Jaross. 2008. Recommendations andsupporting analysis of conservationopportunities for Marbled Murrelet long-termconservation strategy. Washington State Dept.Natural Resources, Olympia, WA. 337 pp.

Raphael, M. G., D. Evans Mack, and B. A. Cooper.2002. Use of radar to investigate landscape-scale relationships between abundance ofMarbled Murrelets and nesting habitat.Condor 104: 331–342.

Raphael, M. G., G. A. Falxa, K. M. Dugger, B. M.Galleher, D. Lynch, S. L. Miller, S. K. Nelson,and R. D. Young. 2011. Northwest ForestPlan—the first 15 years (1994−2008): statusand trend of nesting habitat for the marbledmurrelet. General Technical ReportPNW-GTR-848. Portland, OR: U.S.Department of Agriculture, Forest Service,Pacific Northwest Research Station. 52 p.

Rice, J. A. 1995. Mathematical statistics and dataanalysis, 2nd edition. Duxbury Press,Belmont, CA. 601 pp.

Literature Cited


Sanzenbacher, P. M., and B. A. Cooper. 2008. Aradar study of Marbled Murrelets at theproposed Bear River Windpark, California,summer 2007. Unpublished report preparedfor Shell Wind Energy Inc., Houston, TX, byABR Inc., Forest Grove, OR. 29 pp.

Sanzenbacher, P. M., B. A. Cooper, T. J. Mabee.2011. Site characterization study and habitatmapping for the proposed SkookumchuckWind Energy Project, Lewis and Thurstoncounties, Washington. Unpublished reportprepared for RES America Developments,Inc., Portland, OR, by ABR,Inc.—Environmental Research and Services,Forest Grove, OR. 66 pp.

Sanzenbacher, P. M., B. A. Cooper, J. H. Plissner,and J. Bond. 2014. Intra-annual patterns inpassage rates and flight altitudes of MarbledMurrelets Brachyramphus marmoratus atinland sites in northern California. MarineOrnithology 42: 169−174.

Stumpf, J. P., N. Denis, T. E. Hamer, G. Johnson,and J. Verschuyl. 2011. Flight heightdistribution and collision risk of the MarbledMurrelet Brachyramphus marmoratus:methodology and preliminary results. MarineOrnithology 39: 123−128.

TetraTech EC. 2008. DRAFT Habitat ConservationPlan for the construction and operation ofLanai met towers, Lanai, Hawaii (RevisedFebruary 8, 2008, TTEC-PTLD-2008-080).Unpubl. report by Tetratech EC, Honolulu, HIfor Castle and Cooke LLC, Lanai City, HI. 52pp. + appendices. http://www.state.hi.us/dlnr/dofaw/pubs/Lana'i_Met_Towers_HCP.pdf.Accessed November 2014.

Tucker, V. A. 1996. A mathematical model of birdcollisions with wind turbine rotors. ASMEJournal of Solar Energy Engineering 118:253–262.

U.S. Fish and Wildlife Service (USFWS). 1992.Endangered and threatened wildlife andplants: determination of threatened status forthe Washington, Oregon, and Californiapopulation of the Marbled Murrelet. FederalRegister 61: 26,256–26,320.

U.S. Fish and Wildlife Service (USFWS). 1997.Recovery plan for the Marbled Murrelet(Washington, Oregon, and Californiapopulation). U.S. Fish and Wildlife Service,Portland, OR. 203 pp.

U.S. Fish and Wildlife Service (USFWS). 2011.Notice of application to renew incidental takepermit for meteorological towers on Lanai,HI. Federal Register Volume 76, Number 235.Pages 76,432–76,434.

Washington Department of Fish and Wildlife(WDFW). 2014. State threatened species.http://wdfw.wa.gov/conservation/endangered/. Accessed November 2014.

Williams, T. C., J. Settel, P. O’Mahoney, and J. M.Williams. 1972. An ornithological radar.American Birds 26: 555–557.

Winkelman, J. E. 1995. Bird/wind turbineinvestigations in Europe. Pages 43–47 and110–140 in LGL Ltd. ed. Proceedings ofNational Avian-Wind Power PlanningMeeting I, Lakewood, CO.


Appendix 1. Information on modeling fatality rates.

We have developed a model to estimate the number of potential collision fatalities of seabirds at

manmade structures (i.e., met towers and wind turbines) as a risk-assessment tool. This model was

developed for studies of Marbled Murrelets (Brachyramphus marmoratus) in the Pacific Northwest and

Hawaiian Petrel (Pterodroma sandwichensis) and Newell’s Shearwater (Puffinus auricularis newelli) in

the Hawaiian Islands. The model was first developed in 2004 and has continued to evolve to incorporate

additional model components and aspects of the species life histories. The model has undergone extensive

review including the following: internal review at ABR; external review by state and federal agencies

including California Department of Fish and Wildlife, Washington Department of Fish and Wildlife, and

U.S. Fish and Wildlife Service; and external review by expert seabird biologists (K. Nelson, Oregon State

University; M. Raphael, USDA Forest Service) and statisticians (C. Nations, WEST, Inc.; Tetra Tech, Inc.

staff). To date this model has been used to estimate fatalities of murrelets at two other proposed wind

energy facilities in California and Washington (Sanzenbacher and Cooper 2008, Cooper and Mabee 2010)

as well as fatalities of Hawaiian seabirds at more than four different proposed wind energy facilities

(Cooper and Day 2004b, 2009; Cooper et al. 2007, 2011, Day and Cooper 2008).

The estimated fatality rate is calculated as the product of: (1) the exposure rate (i.e., the number of

times that birds might fly within the airspace occupied by turbines); (2) the avoidance probability (i.e., the

probability that a bird will detect and avoid entering the airspace containing the turbine); and (3) the

fatality probability (i.e., the probability of a fatal collision with a portion of the structure once a bird is in

the airspace occupied by a turbine).

EXPOSURE RATE

The exposure rate represents the number of times that murrelets fly through the airspace occupied by

a turbine over a given time period. To calculate the exposure rate we combined information on the passage

rate of murrelets in the project area (adjusted for post-sunrise targets and flock size) with the horizontal

and vertical probabilities of interaction with a turbine.

A post-sunrise adjustment was needed because we used only pre-sunrise data to compute the morning

murrelet passage rate (landward + seaward targets/d) and some murrelets move after sunrise. Radar studies

in Washington, Oregon, and California (B. Cooper, unpubl. data; n = 17,750 landward targets and n =

12,863 seaward targets) found that on average, 92.4% of morning landward passes of murrelets occur prior


to sunrise (i.e., 7.6% occur after sunrise) and 38.0% of seaward passes occur prior to sunrise (i.e., 62.0%

occur after sunrise). Thus, to account for postsunrise passes, we calculated the sum of the pre-sunrise

landward rate divided by 0.924, plus the pre-sunrise seaward rate divided by 0.380. A similar technique

was used during studies at the proposed Coyote Crest Wind Power Project (Cooper and Mabee 2010). A

flock size adjustment was also needed because previous radar studies with visual confirmation of radar

targets recorded 1.5 murrelets/target (n = 370 targets from Northern California, Oregon and Washington

during 1996–2004; B. Cooper, unpubl. data). The sum of these factors then equaled the total number of

murrelet targets for the morning.

Interaction probabilities consist of horizontal and vertical components. The horizontal- interaction

probability is the probability that a murrelet target will pass through the area (front or side view) of a

turbine and the two-dimensional area sampled by the radar screen (Table 3). Thus, we assumed that the

probability of exposure was equal to the fraction of sampled air space that was occupied by the turbines

and also assumed that there was a uniform distribution of murrelets in the sampled airspace. The

vertical-interaction probability is the probability that a murrelet target will be flying at an altitude low

enough that it might pass through the airspace occupied by a proposed turbine. This probability was

calculated from the 21 murrelet targets observed concurrently on surveillance and vertical radar during

this study, which indicated that 38% of murrelet targets (i.e., 8/21 vertical radar targets) flew at or below

135 m agl.

AVOIDANCE PROBABILITY

Avoidance probability is the probability that a bird will see a turbine and change flight direction,

flight altitude, or both so that it completely avoids flying through the space occupied by it. Currently, no

empirically-derived estimates of avoidance rates are available for Marbled Murrelets at wind turbines,

however, a radar and visual study of murrelet avoidance behavior at a transmission line suggested a high

avoidance rate (100% avoidance; Cooper 2010). Clearly, the ability of murrelets to access nest sites in

old-growth forest stands during periods of low light levels (i.e., dawn and dusk) indicates that murrelets

can avoid obstacles during these inland flights and is suggestive of high avoidance rates at human-made

structures. Data on other seabird species active during crepuscular and nocturnal periods also suggests high

avoidance rates approaching or greater than 99% (Winkelman 1995, Dirksen et al. 1998, Desholm and

Kahlert 2005, Desholm et al. 2006, Chamberlain et al. 2006). Thus, we assumed a conservative range of

avoidance rates (90%, 95%, and 99%) that have been observed for other seabirds for our fatality model but


speculate that murrelets exhibit similarly high avoidance rates (i.e., 95%) as these other seabirds with

avoidance data. See Appendix 4 for a summary of available evidence on murrelet and other seabird

avoidance rates.

FATALITY PROBABILITY

The fatality-probability is derived as the product of: (1) the probability of encountering an active (i.e.,

operating) turbine; (2) the probability of dying if it collides with the turbine; and (3) the probability of

colliding with the turbine if the bird enters the rotor-swept airspace (i.e., are there gaps big enough for

birds to fly through the blades without hitting them). To calculate the first probability, we used data on

average wind speeds collected during mid-May through early August in 2013 and 2014 at the 80-m level

from the met tower (2013) and SODAR (2014) located at the Project. These wind speed data were used to

estimate the average proportion of time (82.3%) that the turbines would be spinning each morning from

0300 h to 0800 h, the approximate peak period of daily inland murrelet passes during the breeding season

(Nelson 1997). We used specifications for the Vestas V110 2.0 MW turbine model for turbine dimensions,

rotor swept area, rotational speed (4.8–12.1 rpm), cut-in speed (3.0 m/s), and cut-out speed (20.0 m/s).

Because a murrelet hitting the proposed turbines will have a very high probability of actually dying, we

used an estimate of 100% for the second fatality-probability parameter. The third probability (i.e., striking

the structure) needed to be calculated in two ways for both active and inactive turbines, because a bird

approaching a wind turbine from the side has essentially a 100% probability of collision, while a bird

approaching from the back or front of a turbine may pass through the rotor-swept area without colliding

with a blade. Following Tucker (1996) we recalculated the collision probability for a murrelet flying

through the blades of the modern turbines planned for the Project. We estimated that a murrelet

approaching from the back or front of an active turbine has only a 16.3% chance of hitting a blade when

passing through the rotor swept area (Table 3). This calculation was based on the length of a murrelet (25

cm; Nelson 1997); the average flight speed of murrelets (mean ± SE velocity = 53 ± 0.1 mph, n = 42,832

murrelet targets; B. Cooper, unpubl. data); and the time that it took a 25-cm-long murrelet to travel

completely past a 3.4-m-wide turbine blade (combined blade width and murrelet wingspan) spinning at the

maximum rotor speed (12.1 rpm). Thus, these calculations indicated that 16.3% of the disk of the

rotor-swept area was occupied by a blade sometime during the length of time (i.e., 0.15 sec) that it would

take a murrelet to fly completely past a rotor blade (i.e., to fly 2.25 m) of an active turbine. We also

calculated collision probabilities based on inactive turbines with stationary rotors (Table 3).


ADJUSTMENT FACTORS

To calculate seasonal and annual fatality rate estimates (Table 5) it was necessary to combine the

daily fatality rate with a number of additional assumptions (Table 4). Specifically, we used information on

the life history of the Marbled Murrelet to account for daily and seasonal differences in the inland flight

activity of murrelets. For instance, it was necessary to account for inland flights of murrelets outside the

dawn activity period, during daylight hours (daytime flight proportion) and evening hours (evening flight

proportion), when some breeding murrelets make additional trips to their nest to feed chicks. To calculate

the daytime flight proportion we used the number of inland flights during mid-day, accounting for the

proportion of flights by breeding versus nonbreeding birds. Previous studies found that the proportion of

inland murrelet flights that were breeding birds ranged from 11−79% (mean = 49%) depending on the

location and year (Hamer and Nelson 1995, Bradley et al. 2004, Peery et al. 2004, Hebert and Golightly

2006, Barbaree 2012). Based on the currently available data the number of provisioning flights at mid-day,

subtracting dawn and dusk flights, ranges from 0−7 flights with a median of 3.5 flights (Nelson and Hamer

1995, Nelson 1997, Jones 2001, Peery et al. 2004, Bloxton and Raphael 2008). Therefore we determined

the daytime flight proportion over a 31-day chick provisioning period was (3.5 * 0.49)/2 = 0.86. We divide

by 2 to account for pairs. The evening flight proportion was based on numerous studies that found the

proportion of evening flights ranged from 0.16−0.39 (average = 0.29) morning flights (e.g., Hamer et al.

1995, Burger 1997, Burger 2001, Cooper, unpubl. data). We used the approach of applying both

adjustment factors specific to chick provisioning to a single month (July) in our model (Table 3) to

represent the chick-rearing period that generally occurs over a 31-day span. We did not account for

potential differences in avoidance rates based on light levels or visibility. Clearly, there could be higher

avoidance rates of turbines during high light/visibility conditions, however, there currently are no data

available on these variables, so we used the same range of avoidance rates across all periods.

We also calculated seasonal adjustment factors to account for differences in murrelet passage rates at

inland sites across the year. Specifically, if we conducted surveys from mid-May through early August

then we assume that the murrelet passage rate and resulting daily fatality rate from these data represent

averages for the main breeding period. However, both radar and AV studies have shown a decrease in

activity levels of murrelets at inland sites outside these summer breeding months (Nelson 1997,

Sanzenbacher et al. 2014). Using data from a full year of murrelet radar surveys at three high-use sites in

northern California (Sanzenbacher et al. 2014), we determined the proportion of flights relative to the main

breeding period that occur during the spring transition period (0.20), fall transition period (0.31), fall molt


period (0.07), and winter period (0.24; Table 3). We decided not to use the radar survey data from the

proposed Radar Ridge Wind Project (Hamer Environmental 2009) because of differences in how seasonal

periods were designated, as well as suspected contamination from non-murrelets in the radar data. We also

did not use the AV data from Naslund (1993) because those studies were not designed to quantify seasonal

differences in murrelet activity.

FATALITY RATE

The product of all the above calculations is the fatality rate, the final metric used for risk assessment at

wind-energy projects. We calculate this fatality rate on a per turbine basis and also across all turbines at a

project. Finally, as an aid in interpreting the magnitude of fatality rates we also use a Poisson distribution

(Rice 1995) to express the probability of different numbers of fatalities occurring each year (0, 1, 2,

3…fatalities). We use a Poisson distribution in this case because it describes the number of times a very

rare occurrence happens in a large number of trials.


Appendix 2. Sampling dates and numbers of landward and seaward murrelet targets observed on radar before sunrise, and number of audio-visual detections of Marbled Murrelets (observed before or after sunrise) at the proposed Skookumchuck Wind Energy Project, Lewis and Thurston counties, Washington, during summer 2013.

Pre-sunrise murrelet radar targets Radarstation Date

Sampling hours

Sunrisetime

Landward targets1

Seaward targets1

Totaltargets

Audio-visualdetections

1 5/26/2013 0341–0641 0526 0 1 1 0 6/19/2013 0333–0633 0518 0 0 0 0 6/26/2013 0335–0635 0520 0 0 0 0

7/13/2013 0347–0647 0532 0 2 2 0 7/27/2013 0402–0702 0547 1 0 1 0

2 5/25/2013 0342–0342 0527 0 0 0 0 6/18/2013 0333–0633 0518 0 0 0 0 6/25/2013 0334–0634 0519 0 0 0 0 7/10/2013 0344–0644 0529 1 1 2 0 7/23/2013 0357–0657 0542 0 0 0 0

3 5/24/2013 0343–0643 0528 0 0 0 0 6/17/2013 0332–0632 0517 0 0 0 0 7/02/2013 0338–0638 0523 0 0 0 0 7/12/2013 0346–0646 0531 0 0 0 0 7/25/2013 0359–0659 0544 0 1 1 0

4 5/30/2013 0338–0638 0523 0 1 1 0 6/16/2013 0332–0632 0517 0 0 0 0 6/29/2013 0336–0636 0521 0 0 0 0 7/11/2013 0345–0645 0530 0 1 1 0 7/28/2013 0403–0703 0548 0 1 1 0

5 6/05/2013 0335–0635 0520 0 0 0 0 6/28/2013 0336–0636 0521 0 0 0 0 7/04/2013 0339–0639 0524 0 0 0 0 7/15/2013 0349–0649 0534 0 2 2 0 7/22/2013 0356–0656 0541 0 0 0 0

6 5/31/2013 0337–0632 522 0 0 0 0 6/14/2013 0332–0632 517 1 0 1 0 7/01/2013 0337–0637 522 0 0 0 0 7/24/2013 0358–0658 543 1 0 1 0 8/01/2013 0408–0708 553 0 0 0 0 7 6/01/2013 0337–0637 522 0 1 1 0 6/15/2013 0332–0632 517 0 1 1 0 6/30/2013 0337–0637 522 0 0 0 0 7/16/2013 0350–0650 535 0 0 0 0 7/26/2013 0400–0700 545 0 0 0 0


Appendix 2. Continued.

Pre-sunrise murrelet radar targets Radarstation Date

Sampling hours

Sunrisetime

Landward targets1

Seaward targets1

Totaltargets


8 6/04/2013 0335–0635 520 0 0 0 0 6/21/2013 0333–0633 518 0 0 0 0 7/03/2013 0338–0638 523 0 0 0 0 7/17/2013 0351–0651 536 0 0 0 0 7/29/2013 0404-0704 549 0 0 0 0

9 6/03/2013 0336–0636 521 0 0 0 0 6/20/2013 0333–0633 518 0 0 0 0 7/09/2013 0343–0643 528 1 0 1 0 7/19/2013 0353–0653 538 1 0 1 0 7/31/2013 0406–0706 551 0 1 1 0

10 6/02/2013 0336–0636 521 1 2 3 0 6/22/2013 0333–0633 518 0 0 0 0 7/08/2013 0342–0642 527 1 1 2 0 7/18/2013 0352–0652 537 2 0 2 0 7/30/2013 0405–0705 550 0 0 0 0

Total Targets: 10 14 24 0 Average targets/day (mean ± SE): 0.20 0.06 0.32 ± 0.08 0.52 ± 0.11 0

1 We classified targets as landward if they were heading between 45 –224 and seaward if they were heading between 225 –44 .


Appendix 3. Sampling dates and numbers of landward and seaward murrelet targets observed on radar before sunrise, and number of audio-visual detections of Marbled Murrelets (observed before or after sunrise) at the proposed Skookumchuck Wind Energy Project, Lewis and Thurston counties, Washington, during summer 2014.

Presunrise murrelet radar targets Radarstation Date

Sampling hours

Sunrisetime

Landward targets1

Seaward targets1

Totaltargets


1 5/17/2014 0349–0649 0534 0 1 1 0 5/31/2014 0336–0636 0521 0 0 0 0 6/16/2014 0331–0631 0516 1 0 1 0 6/21/2014 0332–0632 0517 0 0 0 0 7/14/2014 0346–0646 0531 0 0 0 0 7/19/2014 0351–0651 0536 0 0 0 0 8/03/2014 0409–0709 0554 0 0 0 0

2 5/15/2014 0351–0651 0536 0 1 1 0 5/27/2014 0339–0639 0524 0 1 1 0 6/10/2014 0332–0632 0517 0 2 2 0 6/19/2014 0332–0632 0517 0 0 0 0 7/08/2014 0341–0641 0526 0 0 0 0 7/15/2014 0347–0641 0532 0 0 0 0 7/26/2014 0359–0659 0544 0 0 0 0

3 5/13/2014 0354–0654 0539 0 2 2 0 6/02/2014 0335–0635 0520 2 1 3 0 6/11/2014 0332–0632 0517 1 0 1 0 6/20/2014 0332–0632 0517 0 0 0 0 7/10/2014 0343–0643 0528 0 0 0 0 7/16/2014 0348–0648 0533 0 0 0 0 7/28/2014 0402–0602 0547 0 0 0 0

4 5/19/2014 0347–0647 0532 0 1 1 0 5/29/2014 0338–0636 0523 1 0 1 0 6/05/2014 0334–0639 0519 1 1 2 0 6/18/2014 0331–0644 0516 0 0 0 0 7/02/2014 0337–0637 0522 0 0 0 0 7/9/2014 0342–0642 0527 0 0 0 0 7/25/2014 0358–0658 0543 0 0 0 0

5 5/11/2014 0357–0657 0542 0 2 2 0 5/28/2014 0339–0639 0524 1 0 1 0 6/10/2014 0332–0632 0517 0 1 1 0 6/18/2014 0331–0621 0516 0 1 1 0 7/01/2014 0336–0636 0521 0 0 0 0 7/15/2014 0347–0647 0532 0 0 0 0 7/30/2014 0404–0704 0549 1 1 2 0


Appendix 3. Continued.

Presunrise murrelet radar targets Radarstation Date

Sampling hours

Sunrisetime

Landward targets1

Seaward targets1

Totaltargets


6 5/12/2014 0355–0655 540 1 0 1 0 6/01/2014 0336–0636 521 0 0 0 0 6/12/2014 0332–0632 517 0 2 2 0 6/19/2014 0332–0632 517 0 1 1 0 7/03/2014 0337–0637 522 0 0 0 0 7/16/2014 0348–0648 533 0 0 0 0 7/29/2014 0403–0703 548 1 0 1 0 7 5/21/2014 0345–0645 530 1 0 1 0 6/06/2014 0333–0633 518 0 1 1 0 6/14/2014 0331–0631 516 0 0 0 0 6/22/2014 0332–0632 517 0 0 0 0 7/13/2014 0345–0645 530 0 0 0 0 7/17/2014 0349–0649 534 2 0 2 0 7/27/2014 0400–0700 545 0 0 0 0

8 5/16/2014 0350–0650 535 0 0 0 0 6/03/2014 0335–0635 520 0 0 0 0 6/12/2014 0332–0632 517 0 1 1 0 6/28/2014 0350–0650 519 3 0 3 0 7/14/2014 0335–0635 531 0 0 0 0 7/22/2014 0332–0632 540 0 1 1 0 8/04/2014 0350–0650 555 0 0 0 0

9 5/20/2014 0346–0646 531 1 0 1 0 6/04/2014 0334–0634 519 0 1 1 0 6/17/2014 0331–0631 516 0 1 1 0 6/30/2014 0336–0636 521 0 0 0 0 7/11/2014 0344–0644 529 1 0 1 0 7/17/2014 0349–0649 534 0 0 0 0 7/31/2014 0405–0705 550 0 0 0 0

10 5/14/2014 0353–0653 538 0 1 1 0 5/30/2014 0337–0637 522 0 1 1 0 6/11/2014 0332–0632 517 0 0 0 0 6/23/2014 0332–0632 517 0 0 0 0 7/12/2014 0344–0644 529 0 1 1 0 7/18/2014 0350–0650 535 1 0 1 0 8/2/2014 0408–0708 553 0 2 2 0

Total Targets: 19 28 47 0 Average targets/day (mean ± SE): 0.29 ± 0.07 0.41 ± 0.08 0.70 ± 0.06 0

1 We classified targets as landward if they were flying 45 –224 and seaward if they were flying 225 –44 .


Appendix 4. Information on seabird avoidance rates.

Currently, no empirically derived estimates of avoidance rates are available for Marbled

Murrelets (Brachyramphus marmoratus) at wind turbines, because there has been no opportunity

to study their flight behavior near operational wind turbines. The only available murrelet

avoidance information we are aware of is a radar and visual study of avoidance behavior at a

transmission line corridor across the Duckabush Valley, Washington (Cooper 2010). In that

study, all 423 murrelet targets observed on radar successfully crossed the transmission line

corridor. Flight altitude measurements of murrelet radar targets suggested that 18.6% of those

targets (i.e., ~79 targets) were at or below the 55-m height of the transmission lines and thus at

risk of collision. Of those 79 targets, 10 should have intersected one of the lines by chance alone.

Zero of the 79 targets that could have hit a structure did so; thus the conclusion was that all 10 of

the targets that would have hit by chance alone took evasive action and avoided the transmission

lines (i.e., avoidance was 100% for this study, however, a confidence interval could not be

calculated).

There is evidence that other species of birds also detect and avoid human-made structures in

low-light conditions (Winkelman 1995, Dirksen et al. 1998, Desholm and Kahlert 2005, Desholm

et al. 2006). For seabirds, there are data available that suggest the behavioral-avoidance rate of

Hawaiian Petrels (Pterodroma sandwichensis) and Newell’s Shearwaters (Puffinus auricularis

newelli) near powerlines during crepuscular and nocturnal periods is high (Cooper and Day

1998). For example, across all 207 Hawaiian Petrels observed flying within 150 m of

transmission lines on Kauai, 40 exhibited behavioral responses; of those 40 birds that exhibited

collision-avoidance responses, none (0%) collided with a transmission line. Thus, the

collision-avoidance rate for Hawaiian Petrels in the study was 100% (i.e., 40 of 40 interactions

resulted in collision avoidance). Across all 392 Newell’s Shearwaters observed flying within 150

m of transmission lines, 29 exhibited behavioral responses; of those 29 birds that exhibited

collision-avoidance responses, none (0%) collided with a transmission line. Thus, the

collision-avoidance rate for Newell’s Shearwaters in the study was 100% (i.e., 29 of 29

interactions resulted in successful collision avoidance). There also is some information available


on collision-avoidance of Hawaiian Petrels on Lanai, where the behavior of petrels was studied as

they approached large communication towers near their breeding colony (TetraTech 2008). In

that study, all 20 (100%) of the Hawaiian Petrels seen on a collision-course toward

communication towers at night exhibited avoidance behavior and avoided collision.

Additional indirect data on collision avoidance of seabirds are available from studies

associated with the operational KWP I wind energy facility on Maui and the six meteorological

towers on Lanai. Based on fatality searches and observations during the first five years of

operation at the 20 wind turbines and three met towers at the KWP I facility, the estimated total

annual take was 0.93 Hawaiian Petrels and 0 Newell’s Shearwater fatalities per year (KWP II

2011). Cooper and Day (2004b) used similar methods as the current study to model seabird

fatality for the KWP I wind turbines, based on passage rates from radar studies at the site (Day

and Cooper 1999; Cooper and Day 2004a, 2004b). They estimated that the combined annual

fatality of Hawaiian Petrels and Newell’s Shearwaters at the KWP I turbines would be ~3–18

birds/yr with a 50% avoidance rate, ~1–2 birds/yr with a 95% avoidance rate, and <1 bird/yr with

a 99% avoidance rate. Thus the fatality model that used a 99% avoidance value was a closer fit

with the measured fatality rates than were the fatality estimates based on lower avoidance rates.

Similarly, 0 Hawaiian Petrels were found in five years of fatality searches at 1–6 met towers on

Lanai (USFWS 2011; A. Oller, Tetra Tech, pers. comm.), which fit the preconstruction fatality

estimates based upon radar data and a >99%avoidance factor (i.e., <0.07–0.77 petrels/met

tower/yr with an assumption of 99% avoidance; Cooper et al. 2008). Thus, the two wind energy

projects in Hawaii with preconstruction fatality estimates and post-construction fatality data both

suggest that fatality models based on an assumption that 99% of petrels avoided structures (i.e.,

wind turbines and met towers) produced more realistic estimates of fatality than did models using

lower avoidance values.

A further example of avoidance behavior in seabirds comes from studies of seaducks in

Europe, where those birds were found to detect and avoid turbines >95% of the time (Desholm et

al. 2006). Natural anti-collision behavior (especially alteration of flight directions) is also seen in

migrating Common and King eiders (Somateria mollissima and S. fischeri) approaching

human-made structures in the Beaufort Sea off of Alaska (Day et al. 2005) and in diving ducks

approaching offshore wind parks in Europe (Dirksen et al. 1998). Collision-avoidance rates


around wind turbines are also high for Common Eiders in the daytime (Desholm and Kahlert

2005), Common Terns (Sterna hirundo) and Sandwich Terns (Sterna sandvicensis) during the

daytime (>99%, Everaert and Stienen 2007), and gulls (Larus spp.) in the daytime (>99%; Painter

et al. 1999, cited in Chamberlain et al. 2006).

In summary, the currently available avoidance data for Marbled Murrelets and a variety of

other seabirds, while incomplete, are consistent with the notion that substantial proportions of

these birds detect and avoid large, human-made structures with the weight-of-evidence suggesting

high avoidance rates (i.e., >95%). Clearly, however, there is a need for further studies to collect

additional species-specific, weather-specific, structure-specific, and site-specific data on

avoidance, particularly for efforts to model collision fatalities at human-made structures.

ATTA

CHM

ENT

B – A

lask

a Bi

olog

ical

Res

earc

h, In

c. P

re-c

onst

ruct

ion

Stud

ies o

f Av

ian

Use

of th

e Pl

an A

rea

from

Win

ter 2

014

thro

ugh

Fall

2015

BASELINE STUDIES OF AVIAN ACTIVITY AT THE PROPOSED SKOOKUMCHUCK WIND ENERGY PROJECT, 2014–2015

PETER M. SANZENBACHERTODD J. MABEE

BRIAN A. COOPER

PREPARED FORRES AMERICA DEVELOPMENTS, INC.

BROOMFIELD, CO

PREPARED BYABR, INC.–ENVIRONMENTAL RESEARCH & SERVICES

FOREST GROVE, OR

DRAFT

Printed on recycled paper.

BASELINE STUDIES OF AVIAN ACTIVITY AT THE PROPOSED SKOOKUMCHUCK WIND ENERGY PROJECT, 2014–2015

DRAFT REPORT

Prepared for

RES America Developments, Inc.11101 W. 120th Ave., Ste. 400

Broomfield, CO 80021

Prepared by

Peter M. Sanzenbacher, Todd J. Mabee, and Brian A. Cooper

ABR, Inc.—Environmental Research & Services

P.O. Box 249

Forest Grove, Oregon 97116

February 2015

iii Skookumchuck Avian Study

EXECUTIVE SUMMARY

• RES America Developments, Inc. (RES)proposes to develop the Skookumchuck WindEnergy Project (Project) in Lewis and Thurstoncounties, Washington. The actual size of theProject will be determined closer to the time ofconstruction but the current project designconsists of 52 wind turbines, each with anameplate capacity up to 2.0 megawatts(MW), for a total production capacity of up toapproximately 104 MW. ABR, Inc. (ABR)conducted a full year of pre-constructionstudies of avian use of the project area fromwinter 2014 through fall 2015.

• The primary goal of the study was to obtaininformation on the annual spatial and temporaluse of all birds in the Project area. The specificobjective was to conduct avian use surveysusing point-count methods to describe therelative abundance, distribution, and flightcharacteristics of birds in the Project andderive standard exposure indices for estimatingpotential risk of collision.

• ABR made a total of 36 visits to the site andconducted a total of 291 individual 20-minpoint count surveys at 9 survey stations. A totalof 68 species were detected on surveys. Weobserved the highest number of species duringthe spring (n = 51 species), followed bysummer (44 species), fall (43 species), andwinter (26 species).

• We calculated a range of metrics describingbird use at the site including mean use, speciescomposition, and frequency of occurrence foreach bird group and species. To help assessrisk, an exposure index was calculated basedon the product of a species’ mean use, thepercentage of time spent flying, and thepercentage of time observed flying within therotor swept area (RSA; 25–135 m aboveground level).

• Species of special interest recognized by theU.S. Fish & Wildlife Service and theWashington Department of Fish and Wildlifethat were observed at the Project included:Bald Eagle (federal Bald and Golden EagleProtection Act, Federal Species of Concern,State Sensitive species; n = 25 individuals [21

point count and 4 incidental observations]),Peregrine Falcon (federal Species of Concern,State Sensitive species; n = 4 individuals [1point count and 3 incidental observations]),Northern Goshawk (State Candidate species;n = 1 individual observed incidentally),Olive-sided Flycatcher (Federal Species ofConcern; n = 11 individuals), PileatedWoodpecker (State Candidate species; n = 3individuals), and Vaux’s Swift (StateCandidate species; n = 8 individuals). Of note,two adult Golden Eagles (Federal Bald andGolden Eagle Protection Act, State Candidatespecies) also were recorded as incidentalobservations outside the Project leased areaduring other studies we conducted for theProject prior to the avian use study.

• Raptors generally were uncommon at theProject and we observed a total of 10 differentspecies. Turkey Vulture, Red-tailed Hawk andBald Eagle were the three most commonspecies. Red-tailed Hawks and Bald Eagleswere observed during all four seasons, but allother raptor species were not observed in oneor more seasons. Annual raptor mean use(0.52 birds/point count) ranked third amongthe various avian species groups, behindpasserines and “other birds”. Seasonally, raptoruse was significantly higher in summer than inwinter. Raptor use also was significantlyhigher at survey point #1 on the north end ofthe Project than at all other survey points.

• Approximately 64% of all raptor flightsoccurred within the RSA. Raptors generallyhad low exposure indices, ranging from 0.136in winter to 0.429 in fall and averaging 0.293for the year. Most individual species (i.e., 9 of11 species) exhibited an exposure index of zero(0) in one or more seasons. Of the three mostnumerous raptor species, Turkey Vultureexhibited the highest annual exposure index(0.088), followed by Bald Eagle (0.063) andRed-tailed Hawk (0.053). The two raptorspecies of special interest (i.e., Bald Eagle andPeregrine Falcon) had variable exposureindices among seasons. Specifically, exposureindices for Bald Eagle were 0.074 in spring,0.019 in summer, 0.013 in fall, and 0.121 inwinter and exposure indices for Peregrine

Skookumchuck Avian Study iv

Falcon were 0.011 in spring, 0.000 in summer,0.000 in fall, and 0.000 in winter.

• Passerines were the dominant speciesassemblage across all seasons and accountedfor 66% of all species observed during thestudy (n = 45 species of passerines). The fourmost abundant species observed at the Projectwere all passerines and included Dark-eyedJunco (12.0% of all individuals), AmericanRobin (6.1%), Common Raven (5.8%), andWhite-crowned Sparrows (5.3%). Annualpasserine mean use (9.37 birds/point count)was considerably higher than that of all otheravian species groups. Mean use of passerinesdiffered seasonally, with significantly loweruse in winter than in the other three seasons.

• Approximately 30% of all passerine flightsoccurred within the RSA. Exposure indices forall passerines combined ranged from 0.428 inwinter to 4.324 in fall, averaging 1.748 for theyear. Most individual passerine species (i.e., 31of 45 species) had an annual exposure index ofzero (0), because they either were neverobserved flying in the RSA or were notobserved flying at all. Common Raven andWestern Bluebird were the only two passerinespecies with an exposure index>0 in all fourseasons. The single species of special interestwe observed (i.e., Olive-sided Flycatcher) hadexposure indices=0 in all four seasons.

• Gamebird and waterbird diversity was low,with Sooty Grouse and Canada Goose,respectively, comprising the only species intwo those species groups. The “other birds”species group included low numbers of doves,woodpeckers, hummingbirds, and swifts.

• The annual exposure index was 0, 0.087, and0.134 for gamebirds, waterbirds, and “otherbirds”, respectively. All gamebirds, as well asmost “other bird” species (i.e., 8 of 10 species)had an annual exposure index of zero (0),because they either were never observed flyingin the RSA or were not observed flying at all.Band-tailed Pigeon was the only “other bird”species with an exposure index>0 in all fourseasons. Two of the species in the “other bird”category, Pileated Woodpecker and Vaux’sSwift, were species of interest. The Pileated

Woodpecker had exposure indices of zero (0)in all four seasons and the Vaux’s Swift hadexposure indices of zero (0) in spring, summer,and winter and an index of 0.026 in fall.

• The likelihood for raptor collisions at theproposed Project is expected to be low andwithin the range of fatalities observed atoperational wind projects in the PacificNorthwest. This prediction is based on thefollowing factors: typical raptor speciesdiversity for the region, low mean use, lowfrequency of occurrence, moderate occurrencewithin the RSA, and low exposure index. Theraptor species that would be expected to havethe highest risk of collision would beRed-tailed Hawk, based on their collisionhistory at other new generation wind projectsand because of their documented use of theproject area. Based on low but consistentnumbers of Bald Eagle observations in theproject area and the presence of occupied eaglenest sites outside the Project, additionalinformation on Bald Eagles in the area isneeded to better evaluate potential risk to thisspecies.

• The likelihood for passerine, gamebird,waterbird and “other bird” collisions at theproposed Project is expected to be low andwithin the range of fatalities observed atoperational wind projects in the PacificNorthwest, based on the following factors:typical species diversity, low-to-moderatemean use, high frequency of occurrence, lowoccurrence w/in the RSA, and low exposureindex. The composition of the fatalities couldbe expected to be similar to those recorded atother new generation facilities in the regionwith similar habitats.

v Skookumchuck Avian Study

TABLE OF CONTENTS

Executive Summary......................................................................................................................................iiiList of Figures................................................................................................................................................ vList of Tables ................................................................................................................................................. vList of Appendices......................................................................................................................................... vAcknowledgments ........................................................................................................................................ viIntroduction.................................................................................................................................................... 1Objectives ...................................................................................................................................................... 2Study Area ..................................................................................................................................................... 2Methods ......................................................................................................................................................... 2

Study Design and Protocols ................................................................................................................... 2Avian Use Surveys................................................................................................................................. 2

Survey Methods ................................................................................................................................ 2Survey Schedule ............................................................................................................................... 4Incidental Observations .................................................................................................................... 4

Data Analysis ......................................................................................................................................... 4Results............................................................................................................................................................ 5

Point Count Surveys............................................................................................................................... 5Species Composition......................................................................................................................... 5Mean Use, Percent Composition, and Frequency of Occurrence ................................................... 10Flight Characteristics ...................................................................................................................... 17Exposure Index ............................................................................................................................... 20Incidental Observations .................................................................................................................. 37

Discussion.................................................................................................................................................... 37Direct Impacts ...................................................................................................................................... 38

Raptors ............................................................................................................................................ 38Passerines........................................................................................................................................ 41Gamebirds, Waterbirds and other Species...................................................................................... 42Sensitive Species............................................................................................................................. 43

Indirect Impacts.................................................................................................................................... 43Conclusions.................................................................................................................................................. 44Literature Cited............................................................................................................................................ 44

LIST OF FIGURES

Figure 1. Vicinity map of the proposed Skookumchuck Wind Energy Project, Lewis and Thurston counties, Washington ................................................................................................ 1

Figure 2. Map of diurnal avian point count stations at the proposed Skookumchuck Wind Energy Project, Lewis and Thurston counties, Washington..................................................... 3

Figure 3. Summary of seasonal mean use for raptors, passerines, and all birds at the proposed Skookumchuck Wind Energy Project in Lewis and Thurston counties, Washington, during 2014–2015................................................................................................................... 14

Figure 4. Summary of annual mean use by point count station for raptors and passerines at the proposed Skookumchuck Wind Energy Project in Lewis and Thurston counties, Washington, during 2014–2015.............................................................................................. 21

Skookumchuck Avian Study vi

Figure 5. Flight paths of all buteos, harriers, osprey and vultures observed at point count survey stations 1–4 in the proposed Skookumchuck Wind Energy Project in Lewis and Thurston counties, Washington, during 2014–2015............................................................... 22

Figure 6. Flight paths of all buteos, harriers, osprey and vultures observed at point count survey stations 5–9 in the proposed Skookumchuck Wind Energy Project in Lewis and Thurston counties, Washington, during 2014–2015............................................................... 23

Figure 7. Bald Eagle flight paths observed at point count survey stations 1–4 in the proposed Skookumchuck Wind Energy Project in Lewis and Thurston counties, Washington, during 2014–2015................................................................................................................... 24

Figure 8. Bald Eagle flight paths observed at point count survey stations 5–9 in the proposed Skookumchuck Wind Energy Project in Lewis and Thurston counties, Washington, during 2014–2015................................................................................................................... 25

Figure 9. Flight paths of all accipiters and falcons observed at point count survey stations 1–4 in the proposed Skookumchuck Wind Energy Project in Lewis and Thurston counties, Washington, during 2014–2015.............................................................................................. 26

Figure 10. Flight paths of all accipiters and falcons observed at point count survey stations 5–9 in the proposed Skookumchuck Wind Energy Project in Lewis and Thurston counties, Washington, during 2014–2015.............................................................................................. 27

Figure 11. Summary of estimated raptor mean use values recorded at wind projects throughout the U.S. ................................................................................................................................... 39

LIST OF TABLES

Table 1. Avian species and groups recorded within 800 m of bird survey points at the proposed Skookumchuck Wind Energy Project located in Lewis and Thurston counties, Washington, 2014–2015 ........................................................................................................... 6

Table 2. Avian species observed during incidental observations that were species of special interest, including eagles, or species not observed during point count surveys at the proposed Skookumchuck Wind Energy Project in Lewis and Thurston counties, Washington, 2014–2015 ........................................................................................................... 9

Table 3. Species of special interest observed during the baseline avian use study at the proposed Skookumchuck Wind Energy Project in Lewis and Thurston counties, Washington, 2014–2015 ........................................................................................................... 9

Table 4. Estimated mean use of avian species observed during point count surveys at the proposed Skookumchuck Wind Energy Project located in Lewis and Thurston counties, Washington, 2014–2015 ......................................................................................................... 11

Table 5. Estimated percent composition of avian species observed during point count surveys at the proposed Skookumchuck Wind Energy Project located in Lewis and Thurston counties, Washington, 2014–2015.......................................................................................... 15

Table 6. Estimated frequency of occurrence of avian species observed during point count surveys at the proposed Skookumchuck Wind Energy Project located in Lewis and Thurston counties, Washington, 2014–2015 .......................................................................... 18

Table 7. Overall flight height characteristics and percentage of avian species observed flying within 800 m of survey points during point count surveys at the proposed Skookumchuck Wind Energy Project located in Lewis and Thurston counties, Washington, 2014–2015 ..... 28

vii Skookumchuck Avian Study

Table 8. Seasonal percentages of avian species observed flying at altitudes relative to the rotor-swept area within 800 m of survey points during point count surveys at proposed Skookumchuck Wind Energy Project located in Lewis and Thurston counties, Washington, 2014–2015.......................................................................................... 32

Table 9. Seasonal and annual exposure indices calculated for avian species observed within 800 m of survey points during point count surveys at the proposed Skookumchuck Wind Energy Project located in Lewis and Thurston counties, Washington, 2014–2015 ..... 35

LIST OF APPENDICES

Appendix 1. Examples of raptor use estimates and estimated raptor collision fatalities from wind power developments in the western United States ................................................ 48

Appendix 2. Data sources and full citations for summary of raptor mean use values at wind projects shown in Figure 11 ............................................................................................ 53

ACKNOWLEDGMENTS

Funding for this study was provided by RES America Developments, Inc. (RES). We thank Sean Belland Theresa Webber at RES for assistance with project coordination and logistics. WeyerhaeuserCorporation provided site access. At ABR, Inc. Delee Spiesschaert and Susan Cooper provided logisticalsupport; Mike Davis and Michael Medina conducted field surveys; Rich Blaha and Pam Odom providedassistance with report preparation.

Skookumchuck Avian Study viii

Documents

CHANGED AND UNFORSEEN CIRCUMSTANCES 7.1 CHANGED …