04._Performance_Work_Statement_Avian_Predation_Analysis.pdf

Performance Work Statement

KBAO Avian Predation Analysis

1.0 SCOPE

1.1 OBJECTIVE The objective of this contract is to acquire contractor services to provide a comprehensive investigation of predator-prey interactions in Clear Lake Reservoir. This would be accomplished by two separate but related, tasks; (1) recover sucker PIT tags from bird colonies and conduct aerial photo censuses necessary to detect, enumerate, and guide PIT tag recovery efforts at American white pelican and double-crested cormorant colonies in the Upper Klamath basin and (2) determine PIT tag deposition probabilities for pelicans and cormorants, as feasible, to improve the accuracy and precision of predation rate estimates generated as part of Task 1.

The primary focus of this study will be to determine an independent on-colony PIT tag deposition probability for American white pelicans. However, if an isolated group of double-crested cormorants is accessible, similar methods would also be applied to estimate deposition probabilities for cormorants nesting in Clear Lake Reservoir. If cormorant-specific deposition trials cannot be conducted, an annual PIT tag deposition probability of 0.51 (95% c.i. = 0.34 - 0.70) established for cormorant nesting in the Columbia River estuary would be used (see Hostetter et al. 2015). Hostetter et al. (2015) used identical methods proposed herein, and exclusively fed ground-nesting cormorants (the same nesting habitat used in the Upper Klamath basin) hatchery-reared rainbow trout in the size range of smolts found in the estuary.

1.2 BACKGROUND

● Thousands of Lost River suckers and shortnose suckers are annually PIT-tagged (or recaptured) in

Upper Klamath Lake and Clear Lake Reservoir. These data are used to gather information on Lost River sucker and shortnose sucker behavior and survival (Janney et al. 2008; Hewitt and Hayes 2013; Hewitt et al. 2015; Burdick et al. 2015).

● In 2019, as part of efforts to recover imperiled sucker populations and to address concerns regarding juvenile recruitment, upwards of 3,000 hatchery-reared juvenile suckers will be PIT-tagged and released into Upper Klamath Lake as part of the U.S. Fish and Wildlife Service’s Sucker Assisted Rearing Program (SARP). PIT tags enable specific information to be linked to individual fish at release and recapture, including species, size, age-class (adult, juvenile), release/recapture location, rearing history (SARP, naturally reared), and other information. A portion of these PIT-tagged suckers are consumed by avian predators with their corresponding PIT tag deposited (regurgitated or defecated) at nesting sites in Upper Klamath Lake and Clear Lake Reservoir, where they can be detected by researchers following the breeding season (Evans et al. 2016).

● This study addresses the need for a comprehensive investigation of predator-prey interactions in

Clear Lake Reservoir. This would be accomplished by two separate but related, tasks; (1) recover sucker PIT tags from bird colonies and conduct aerial photo censuses necessary to detect, enumerate, and guide PIT tag recovery efforts at American white pelican and double-crested cormorant colonies in the Upper Klamath basin and (2) determine PIT tag deposition probabilities for pelicans and cormorants, as feasible, to improve the accuracy and precision of predation rate estimates.

2.0 REFERENCES - The following document may be referenced during the performance of this contract: No other documents are required.

Number

Title and Edition

Date

Source

Biological Opinions on the Effects of Proposed Klamath Project Operations from May 31, 2013, through March 31, 2023, on Five Federally Listed Threatened and Endangered Species

May 31, 2013 “Bureau of Reclamation”

3.0 TASKS - The contractor shall perform the following tasks:

Task 3.1: PIT Tag Recovery and Colonial Waterbird Surveys

In collaboration with U.S. Geological Survey – Klamath Field Station staff, conduct passive electronic recovery of sucker PIT tags at American white pelican and double-crested cormorant colonies detected in the Upper Klamath basin. Scanning efforts would be identical to those described in Evans et al. (2016), with multiple passes conducted for each pelican and cormorant colony identified during aerial surveys. Where possible, colony observations and geo-locations of recovered tags would be used to attribute a specific predator species to the mortality events.

Because pelicans and cormorants often nest communally (i.e., species-specific nesting areas can overlap from one season the next), and only sucker PIT tags recovered in the same year they were detected alive can be assigned a specific predator, efforts to do so have been limited to date. If segregated nesting areas are observed, the SARP releases present a unique opportunity to determine species-specific predation given that all individuals will be available to avian predators for the first time.

Although recovery efforts will primarily focus on American white pelican and double-crested cormorant colonies, other piscivorous waterbirds (e.g., Caspian terns, great blue herons) may also breed at some locations and their respective nesting habitat would also be scanned, where feasible. All data will be compiled in a database maintained in collaboration with the U.S.

Geological Survey - Klamath Field Station (POC David Hewitt).

Coordination with U.S. Fish and Wildlife Service – Klamath Basin National Wildlife Refuge Complex staff would occur prior to initiating this work, with any necessary special use or access permits requested during the Spring and Fall of 2018 and 2019, as necessary.

Completion of this work is also contingent on transportation to the remote colony locations in Upper Klamath Lake following the 2019 and 2020 breeding season via a Marsh Master owned and operated by U.S. Fish and Wildlife Service. Since 2009 the Refuge has supported this task by providing access to colonies in the marsh that are not accessible by conventional vehicles or boats, therefore, recommend requesting their support in 2019 and 2020.

The size of American white pelican and double-crested cormorant colonies would be determined using the methods of Adkins et al. (2014) by obtaining oblique digital photography during multiple aerial surveys of waterbird colonies in the Upper Klamath River basin. Surveys would be scheduled throughout the breeding season to detect early colony formation (late April), peak colony size (late May to early June), and nesting success (late July), to the extent possible (i.e., assuming colonies are active). Anticipated survey locations would include historical colonies in Upper Klamath Lake, Clear Lake Reservoir, Tule Lake, and Sheepy Lake, and other potential locations within the basin that are identified during the 2018 and 2019 breeding season. Aerial photographs would be imported into an existing GIS software platform developed by our research team to enumerate individual pelicans and cormorants at each colony, and to characterize colony productivity by identifying fledglings at each colony (see Evans et al. 2016 for additional details). These data are particularly important in understanding variation in predation rates that may be attributed to differences in colony attendance, colony size, and productivity at each colony.

Predation rate estimates for juvenile Upper Klamath Lake suckers would be based on the number of fish available to avian predators during the 2018 and 2019 breeding season. As such, all juvenile suckers released as part of the SARP prior to 31 August each year (presumed end of the nesting season) would be considered available to avian predators. If sample sizes are adequate, predation rates for sub-groups of fish will also be generated (e.g., repatriated vs.

salvaged) to determine possible differences in relative susceptibility. Additionally, calculate predation rate estimates for adult Lost River and shortnose suckers, to the extent possible (as sample sizes allow). To do so, a PIT-tagged sucker must have been either physically captured in nets, or passively encountered at remote antennas within the year prior to the tag being detected on a bird colony, but no later than August 31. For instance, all PIT-tagged suckers captured or encountered at remote antenna arrays between 1 September 2017 and 31 August 2018 would be considered available to avian predators in 2018 (Evans et al. 2016). To minimize spurious results that can arise from small sample sizes of tagged fish (Evans et al.

2012), limit analyses to groups of ≥ 100 PIT-tagged suckers per subset.

Following previously established methods (Hostetter et al. 2015; Evans et al. 2016), a Bayesian hierarchical model will be used to estimate predation rates based on the number of PIT-tagged suckers available and the number of these tags recovered on Upper Klamath Lake and Clear Lake Reservoir colonies in 2018 and 2019. This model simultaneously estimates and incorporates the uncertainty stemming from imperfect on- colony detection and deposition probabilities (Task 2; see Figure 2 for conceptual illustration), resulting in unbiased and precise estimates of predation rates (Hostetter et al. 2015).

Figure 2. Conceptual model of the tag recovery process in capture–mark–recovery studies of avian predation on fish populations. The probability of recovering a fish tag on a bird colony is the product of three probabilities: the probability that a tagged fish is consumed (predation probability), the probability that the tag is deposited on the nesting colony (deposition probability), and the probability that the tag is detected by researchers (detection probability). Predation rate calculations to-date at Upper Klamath basin bird colonies have not accounted for deposition probabilities, resulting in minimum estimates of bird predation on ESA-listed suckers.

At minimum, generate predation rates on juvenile suckers released into Upper Klamath Lake and species-specific (i.e., LRS, SNS) predation rates for adult suckers in Upper Klamath Lake and Clear Lake Reservoir for each colony location identified during aerial surveys. Predation rate estimates would also be generated for each sub-group of fish with adequate sample sizes.

To provide a comprehensive assessment of avian predation on Upper Klamath Lake juvenile suckers, generating predation estimates for colonies not located in Upper Klamath Lake, but within foraging range, is necessary given that prior research has documented foraging bouts by birds nesting at Clear Lake Reservoir in Upper Klamath Lake (Evans et al. 2016). Without a correction for deposition (see below), predation rate estimates generated as part of this task would remain minimum estimates of predation on Upper Klamath Lake and Clear Lake Reservoir suckers in 2018 and 2019 (see Evans et al. 2016 for additional details).

Task 3.2. PIT Tag Deposition Probabilities

To determine what fraction of PIT tags ingested by American white pelicans and double-crested cormorants are subsequently deposited on breeding colonies, follow methods established in Hostetter et al. (2015), by directly feeding breeding birds dead fish implanted with PIT tags throughout the breeding season in 2020. Since colonies in Upper Klamath Lake are not readily accessible for in-season monitoring, this study would be conducted at anticipated colonies in Clear Lake Reservoir. Alternatively, if colonies are not established in Clear Lake Reservoir due to environmental conditions or failed nesting, other potential colony locations would be considered if access is permitted (e.g., artificial islands in Sheepy and Tule lakes built by the U.S. Army Corps of Engineers). Deposition probabilities derived from this study would be applied to predation rate estimates for pelican and cormorant colonies located throughout the Upper Klamath basin.

For the deposition study, fish would be procured in a range of lengths representative of PIT-tagged suckers naturally present and depredated by American white pelicans and double-crested cormorants in Clear Lake Reservoir (72 mm – 730 mm; Evans et al. 2016).

Three independent feeding trials to identify any temporal variation in on-colony deposition probabilities will be attempted. These periods would be determined by colony-specific breeding chronology but generally would be during the early stages of colony establishment (late March), during peak colony attendance (late May), and during the chick rearing period (late June). Roughly equal sample sizes of PIT-tagged fish would be fed to birds each trial (n = 100 per trial) with each trial lasting no more than one week. Sample sizes are the same as those used by Hostetter et al. (2015), which were adequate for statistical comparisons of deposition probabilities by trial. Each fish would be measured, implanted with a PIT tag identical to those used in sucker tagging programs (134 kHz full-duplex 12-mm), and presented to breeding pelicans or cormorants, as feasible. The final fate of each PIT-tagged fish would be recorded, with only those fish positively ingested by birds included in subsequent analyses. To increase the likelihood of successful feeding events, feed birds on land, within their nesting area, an approach that has been successfully used with other waterbird species in the Columbia River basin (double-crested cormorants, California gulls; Hostetter et al. 2015) without colony disturbance. However, if breeding locations prevent this, consider methods similar to those described in Teuscher et al. (2015) whereby fish were injected with air and floated to foraging birds on the water. Other methods may also be feasible and would be explored during the experiment setup phase (see Deliverables and Schedule below).

To help ensure that only breeding birds associated with the Clear Lake Reservoir colonies are fed, birds targeted in feeding trials would be on or near colony locations. Feeding trials would be conducted in a manner as to not disturb nesting birds, with PIT-tagged fish tossed to birds from temporary blinds, boats, tunnels or similar. Additionally, trials would be designed to feed as many unique birds as possible to best represent the colony at large. Boat- or ground-based colony monitoring during each feeding period, and photography from aerial surveys (see Task

1) would be used to identify breeding birds and delineate accurate colony boundaries, information critical to isolate PIT tags that were deposited on-colony during subsequent scanning efforts. These observations would also determine in-season patterns of use and interactions between various piscivorous waterbird species nesting at Clear Lake Reservoir, information necessary to calculate predator-specific predation rate estimates.

4.0 PERFORMANCE REQUIREMENTS SUMMARY

Task No.

Task Description

Performance

Indicator

Performance

Standard

Minimum Acceptable Quality

Level 3.1-3.2 Reporting Timeliness Receipt of draft and final products by COTR no later than the delivery time as stated in Section 5.0

Adherence to schedule;

Schedule tracking by

COTR

3.1-3.2

Reporting

Quality of Deliverables (such as written documents)

Sufficiency review by COTR of deliverables

Successful review without major deficiencies (e.g., erroneous conclusions); Review by COTR

3.1-3.2

Informal Updates Scheduled meetings and/or conference calls

Participation in meetings and/or conference calls to progress

Attendance - 100% of all meetings and/or conference calls as requested; Review by COTR of conference calls and/or meeting attendance logs from all sources

5.0 DATA AND REPORTS The contractor shall provide the following data and reports as follows:

Item No.

Description

Content

Format

Medium

Delivery

1.Informal updates

2. Progress Report

3. Draft Scientific Report

Quarterly updates every three (3) months

Six (6) months after award date

Results of Investigation

Progress

Progress

Results

Telephone or email

Telephone or email

Printed

Same

Printed

*KBAO c/o

*KBAO c/o COTR / three

(3) hard copies and digital results by June 30,

4. Oral Presentation

5. Final

Report with Reviewer Comments addressed

Research Findings

Scientific Results with changes

Power Point

Printed/digital

Digital

Printed/digital

Conducted at KBAO or USFWS by Sep 30, 2020

COTR three

(3) hardcopies and digital results by Dec 30, 2020

*Hard-copy and electronic files (e.g., MS Word, Adobe Acrobat pdf) of the draft and final reports shall be submitted to the U.S. Bureau of Reclamation Klamath Basin Area Office attention:

Attention: Torrey Tyler

U.S. Bureau of Reclamation Klamath Basin Area Office 6600 Washburn Way Klamath Falls, OR 97603

6.0 DELIVERABLES - The contractor shall submit the following reports in accordance with paragraph 5.0, “DATA AND REPORTS”:

A Draft Scientific Report (detailed report including an introduction, Method, Results, and Discussion sections) of findings associated with Task 3.1-3.2 will be submitted to the Bureau of Reclamation and U.S. Fish and Wildlife Service by 31 June 2020. A Final Scientific Report, incorporating comments received from the BOR, USFWS and USGS, will be delivered no more than 45 days after receiving written comments on the draft report. After the Final Scientific Report is complete, an oral presentation of research findings will be made at the BOR or USFWS offices in Klamath Falls, OR.

A Draft Data Report (2-3-page summary of findings and associated data tables and graphs) will be submitted to the Bureau of Reclamation and U.S. Fish and Wildlife Service by 30 December 2020. A Final Data Report, incorporating comments received from the BOR, USFWS, and USGS, will be delivered no more than 45 days after receiving written comments on the draft report. The Data Report will provide an update of and summary of findings during the 2019 nesting season.

Although it is not a specific deliverable associated with this project, in collaboration with the USGS, attempt to submit the Scientific Report to a peer-reviewed scientific journal for publication. Any publications and presentations stemming from this project will gratefully acknowledge funding and participation from the Bureau of Reclamation and the U.S. Fish and Wildlife Service.

The contractor shall provide an interim report to provide an update on the status of the tasks associated with the contract to Reclamation through conference calls.

The contractor working collaboratively with USGS will complete performance work statement tasks according to the following schedule:

a. The contractor shall provide updates to Reclamation informally through telephone and email every three months after contract is awarded through to completion of the agreement.

b. The contractor will provide a progress report document to Reclamation KBAO six months after contract award date.

c. The contractor will provide Reclamation with a draft Scientific Report by June 30, 2020 of the results. Reclamation will provide comments back to the contractor by July 30, 2020.

d. The contractor shall revise the draft Scientific Report based on review comments and provide Reclamation with a hard-copy and an electronic file of a Final Scientific Report by

December 30, 2020.

e. The contractor will provide an oral presentation using PowerPoint software of research findings will be held at either the U.S. Fish and Wildlife or Reclamation office in Klamath

Falls, Oregon by September 30, 2020.