B08_Attachment 18 - SAMPLE_Small Review_DraftAMCR2020-ISAP-Review_20April2021Final.pdf

ISAP review of the 2020 AMCR Page 1 of 29

Independent Science Advisory Panel Evaluation of the Draft 2020 ESA Adaptive Management Compliance Report for Endangered Species Act Compliance, Adaptive Management Implementation, and Fish and Wildlife Mitigation

Submitted on 20 April 2021

Introduction

The Independent Scientific Advisory Panel (ISAP)1 was tasked with reviewing the draft 2020 ESA Adaptive Management Compliance Report for Endangered Species Act Compliance, Adaptive Management Implementation, and Fish and Wildlife Mitigation (AMCR) in a memo from the USACE on 7 February 2020. Attached to the memo, the Corps provided the ISAP with four documents for review – 1) the Draft 2020 ESA Adaptive Management Compliance Report for Endangered Species Act Compliance, Adaptive Management Implementation, and Fish & Wildlife Mitigation, 2) Appendices to the report, 3) Appendix attachment 1a (in reference to RPM #3), and 4) Appendix attachment 3a re: IRC monitoring. This is the third review of an annual report on USACE activities in support of adaptive resource management (AM) on the Missouri River in compliance with requirements under the U.S. Fish and Wildlife Service’s 2018 Biological Opinion concerning the Operation of the Missouri River Mainstem Reservoir System, the Operation and Maintenance of the Bank Stabilization and Navigation Project, the Operation of Kansas River Reservoir System, and the Implementation of the Missouri River Recovery Management Plan.

Consistent with previous report-review practices, the US Army Corps of Engineers (USACE) received input from external sources in crafting five task questions that are intended to guide this ISAP review. The questions provide the organizing structure to this review, however, the ISAP does not strictly hew to these questions, offering additional observations and recommendations pertinent to the maturing adaptive management program targeting pallid sturgeon and piping plovers. The questions include --

1) Does the draft Adaptive Management and Compliance Report (AMCR) provide the information necessary to evaluate the effectiveness of each management action given each action’s (current) status in the adaptive management cycle (that is, in plan and

1 For this review, the ISAP included Drs. Steve Chipps (co-chair), Dennis Murphy (co-chair), Steve Dinsmore, Melinda Daniels, John Loomis, and Bill Warren-Hicks. Steve Bartell served as Third Party Science Neutral.

ISAP review of the 2020 AMCR Page 2 of 29 design, implement, or monitoring phases of the cycle)? For each individual management action please identify any analyses, modeling outputs, or data summaries that should receive more or less emphasis or explanatory detail in the report, given the current status of the management action in the AM cycle.

2) Does the draft AMCR adequately forecast expected (future) outcomes of ongoing or potential management actions for purposes of decision making in the adaptive management process?

3) Does the draft AMCR describe ongoing and potential management actions and their implications, including challenges and uncertainties likely to be encountered, as they pertain to: 1) status and trends of the listed species and their habitats; 2) agency staff and partners who will implement selected actions and monitor them; and 3) stakeholders who may have concerns regarding the effects of those actions on “human considerations”?

4) Does the draft AMCR provide an updated review of the (current) status of the listed species and their habitats, and the adequacy of information from research, modeling, and monitoring to make real time decisions in an adaptive management framework (including information from published and unpublished sources)? Does the draft AMCR describe how the available scientific information has been and will be used in programmatic adaptive management decision-making?

5) Does the draft 2020 AMCR adequately address issues that were daylighted in ISAP’s review comments on the draft 2019 AMCR? Are there further refinements to the structure of the 2020 report which would better serve the information needs of programmatic adaptive management targeting the listed species and their habitats? (Note that the bird chapter in the draft 2020 AMCR does not incorporate all of the ISAP’s recommendations in its review of the 2019 AMCR; they will be addressed in next year’s

AMCR.)

The ISAP recognizes that the AMCR should 1) provide analyses of available monitoring, research, and habitat-condition data and relate the available data to the performance of management actions that have been undertaken, 2) relate findings drawn from those analyses to whether program targets and objectives are being achieved, and provide a forecast of outcomes of potential future management-action scenarios, thereby informing the adaptive management component of the program’s decision-making process, 3) describe candidate management options under consideration and their implications for the listed species and their habitats, for agency staff who will implement selected actions, and for stakeholders who may have concerns regarding the effects of those actions on “human considerations.” The ISAP understands that the evolving adaptive management program is not yet fully operational and has not advanced to full implementation for either of the targeted species, hence the 2020 AMCR cannot be expected to meet those three criteria in full. Accordingly, the ISAP in this review is looking for progress by the USACE toward a transparent, structured adaptive resource management program that is responsive to the species’ immediate and long-term ecological needs, with presentation of new information developed under the Missouri River Recovery Program, reflecting where appropriate the concerns of the many stakeholders in the Missouri River basin.

The ISAP in its task statement was reminded by the USACE of the sourcing and objectives of the annual AMCR -- “The Technical Team performs necessary data analysis and synthesis to evaluate action effectiveness,” conducting “additional analyses as directed by the ISP Manager

ISAP review of the 2020 AMCR Page 3 of 29 and as approved by the Senior Program Manager,” with results intended to meet four primary report objectives. They are to “(1) provide an analysis of monitoring data, especially pertaining to performance of actions relative to the targets, objectives and goals of the MRRP; (2) provide a forecast of outcomes of future management scenarios; (3) outline recommendations for managers and stakeholders; and (4) provide a review of the status of the science, including current published and unpublished research results relevant to management.”

That noted, consequential information beyond the documents provided to the ISAP for review has been presented in multiple web-based meetings that focused on the ongoing adaptive management efforts. The draft 2020 Adaptive Management Compliance Report was accompanied by three webinars held on 9 and 12 March 2021, focusing on lower Missouri River basin pallid sturgeon, upper Missouri River basin pallid sturgeon, and piping plover, which provided opportunities “to hear about the draft 2020 AMCR structure, selected findings with strategic implications, and possible strategic questions for the AM Workshop; and to raise clarification questions on the draft AMCR content and identify additional potential key topics and strategic questions for consideration during the AM Workshop” (from RESOLVE emails announcing the webinars). Those webinars were helpful to the ISAP in pointing to essential components of the AMCR, identifying points of emphasis, and providing an opportunity for the ISAP to get clarification on a number of AMCR issues. Questions posed to the webinar presenters by engaged MRRIC participants were valuable in conveying stakeholder interests, which were considered in this review.

Additional information related to adaptive resource management under the Missouri River Recovery Program was offered in three MRRIC-hosted workshops labeled Upper Basin Fish and HC Discussions, Lower Basin Fish and HC Discussions Targets, PSPAP, and Tradeoffs/Priorities Discussions convened as the Adaptive Management Workshop, 7-9 April 2021. Those meetings were intended to “Explore management options building on the draft 2020 Adaptive Management Compliance Report (AMCR); assess implications for the MRRP's strategic direction (particularly for 23 and beyond), and for defining the priority information that would support future decision-making; discuss tradeoffs and priorities for implementation and strategic direction given current budget context; and provide opportunity for ISAP (and other participants as time permits) to ask technical questions of USACE and Technical Team based on review of draft 2020 AMCR” (from the Resolve workshops agenda).

That set of meetings was preceded by an AM Workshop Pre-Session: IRC Conceptual Effects Pathways Discussion on 29 March 2021, described as “One of several responses to ongoing stakeholder concerns about potential impacts of IRC projects on HCs”… in particular “2019 MRRIC HC WG Strategic Plan Recommendations 16, 17 from MRRIC and ISAP to explore hypothesized effects of IRC projects on HCs.” The meeting’s purpose was to allow the USACE to “more explicitly describe the links between monitoring activities and specific HC concerns,” with the intent of providing “transparency and improved quality of discussions with HC Work Group,” and consistent with adaptive management promising “periodic improvements over time following ongoing stakeholder engagement and technical discussions.”

The USACE set the draft 2020 AMCR in context for the 2021 review – “Following the ISAP’s review of the 2019 AMCR, the Technical Team and agency staff restructured the AMCR and

ISAP review of the 2020 AMCR Page 4 of 29 discussed the revised structure with the ISAP and MRRIC and those changes are reflected in the draft 2020 AMCR. Revisions included organizing the report by management action, integrating HC information by management action, increasing clarity on current location in the AM cycle for each action, and in clearly distinguishing information specifically related to adaptive management needs from the broader ESA compliance reporting. In addition, beginning with the 2020 AMCR, reporting has an alternating year focus beginning with pallid sturgeon. Therefore, the draft 2020 AMCR will be focused on information to support programmatic AM decision-making for pallid sturgeon and information to support strategic plan level decision-making for management actions for the piping plover; the reviews will consider the full report each year in this context.”

Whereas the 2019 AMCR challenged the ISAP’s review process with its formatting and emphasis on compliance, the 2020 AMCR organizationally is on point. The 2020 AMCR is arranged into three sections -- a compliance section, and “adaptive management for birds” and “adaptive management for fish” sections. The two species sections are organized by management actions, catalogue ongoing scientific efforts, offer model predictions, and describe the “state of the science” and learning that is pertinent to the adaptive management effort. The information and order that is conveyed in the 2020 AMCR made addressing the USACE task questions reasonably straightforward. The USACE has made the 2020 AMCR accessible and readily reviewable. And, as noted above, the 2020 AMCR initiates an every-other-year emphasis on each of the two species – this year the pallid sturgeon is emphasized. The ISAP supports the annual bifurcation of adaptive management reporting but acknowledges that it is not yet clear how the material presentations should accommodate species information transfer differently in the “on years” and the “off years.” Moving ahead, the ISAP requires a clearer understanding of what information to expect (and not expect) in the report for each of the species for “on” and “off” years.

The ISAP has not met in person as a panel in more than a year and a half. Given the lack of face-to-face interaction on this work product and time constraints in providing this AMCR review, the ISAP acknowledges that this report is not a fully resolved consensus document. It should be understood that some of the material observations in both species sections of the review may not have been agreed to by the panel as a whole. In that light, the observations herein are offered as points of departure for further discussion in this pathbreaking and evergreen adaptive management planning process under the Missouri River Recovery Program (MRRP). The issues advanced in this review should be considered as part of an ongoing dialogue between the ISAP and the USACE and its technical teams. In that light, material observations in his review report, as in previous reports, should be viewed as advisory, not prescriptive. The ISAP understands that at least a few of the observations and recommendations below may not be actionable, at least in the near term, due to staffing and/or budgetary constraints.

ISAP review of the 2020 AMCR Page 5 of 29

Piping plover – questions and responses

1. Does the draft Adaptive Management and Compliance Report (AMCR) provide the each action’s (current) status in the adaptive management cycle (that is, in plan and design, implement, or monitoring phases of the cycle)? For each individual management action please identify any analyses, modeling outputs, or data summaries that should receive more or less emphasis or explanatory detail in the report, given the current status of the management action in the AM cycle.

The 2020 AMCR, a product of even-numbered-year reporting, is limited to summary information on piping plover (in its Chapter 2) accompanied by monitoring results and related analyses (in appendices). The prescription for adaptive management for piping plover is for a process, not specific actions, because, the program is itself adaptive. To evaluate adaptive management efforts targeting the plover, we need to know what was done, why it was done (linking to the SAMP), and what benefits were realized from actions undertaken. The authors of the AMCR have not clearly stuck to that tasking in the bird sections of the report; we encourage additional discussion with the ISAP to make this AMCR, and importantly the next AMCR, even more on task. The listing of hypotheses found in the Appendices to the 2020 AMCR (section 2.1.3.2) is useful. The hypotheses are referred to in the data analysis sections of the document and provide an essential link from the data analysis section to explanations of hypotheses under consideration. In the same light, the section on uncertainties (Table 2.5), and linking actions with certainty, are also useful and provide the reader with an informative narrative describing the difficulties of assessing AM approaches for plovers and terns. The summary of performance metrics (Table 2-1) is likewise helpful in summarizing the basic habitat, population growth, and fledge ratio compliance metrics.

Those text sections stop short of making a clear link to adaptive management needs. Specifically, linkages between Emergent Sandbar Habitat (ESH), population trends, and management actions are not fully developed in the AMCR text and no integrated analyses are presented. The relationships between ESH, population growth, and fledge ratio are not explored in an AM context. Furthermore, those three metrics suggest that population growth rate is responsive to ESH via immigration/emigration rather than local reproductive success (manifested as low fledge ratio). For example, the ESH targets are met and that indicates that habitat management is not needed in the near future. The population growth rate was met in 2020, but is still below the 3-year target, and the notion that lambda more than doubled in a single year suggests that immigration played a substantial role. The fledge-ratio was not met during any year in the reporting period, which indicates a need for some management action -- perhaps predator control -- to reverse trending low ratios. The findings suggest immediate AM attention should be directed to local reproductive support in the forms of nest predation control and human access restrictions to prevent nest disturbance. Uncertainties for both predator control and human restrictions are substantial given large gaps in knowledge resulting from the limited monitoring/monitoring effectiveness in both cases. A single table where benchmarks are clearly linked to AM needs is needed. That link is addressed in Table 2-2.

ISAP review of the 2020 AMCR Page 6 of 29

The adaptive management material for piping plover (Section 2.3) summarizes the basic AM process, actions completed in 2020, details about each management action, and impacts to Human Considerations (HCs). The primary function of the AMCR is to describe how monitoring data specifically link to one or more planned management actions. That essential summary in Table 2-3 is useful but is incomplete and does not clearly link each specific management action to a species need that is identified during monitoring. Note that that summary is redundant with the somewhat more useful summary in Table ES-1 that is presented in the report introduction.

More specifically, the lack of need for sandbar construction and augmentation actions is justified because ESH targets were exceeded. But justification for vegetation management in two river segments and for nest and chick relocation and predator removal is missing. Also missing are the specific monitoring findings that prompted those management actions. An effective and transparent AM plan should describe specific links between monitoring data and management actions. The 2020 report doesn’t meet that objective.

The individual management actions for the piping plover are unevenly described in the report.

The information on ESH is covered well, including the current state of ESH, projections for future ESH needs (from models), and an evaluation of ESH future conditions under different hydrological scenarios. However, comparable levels of detail are missing for other management actions, including mechanical and herbicide treatments to maintain ESH, nest relocation, and predator management. A balanced presentation of all management actions is needed in the report. The main AMCR report should include essential information necessary to understand the AM actions that were taken and why they were taken. Some essential material information on piping plover that is relevant to AM reporting is currently sequestered in the appendices.

The ISAP observes that monitoring of potential HC implications received minimal specific coverage in the report. The primary management action undertaken for birds in 2020 was vegetation management using herbicides. Associated monitoring with HC implications focused on related water quality. This monitoring is done annually and shared with stakeholders. The monitoring this year, as in past years, found no significant difference in water quality before and after herbicides were sprayed. The post-herbicide application water quality metrics were substantially below the water quality standards developed to protect human health and aquatic life. Continuing that water quality monitoring and reporting is recommended.

2. Does the draft AMCR adequately forecast expected (future) outcomes of ongoing or potential management actions for purposes of decision making in the adaptive management process?

Expected future outcomes of management actions are addressed in the AMCR in Section 2.5.

The focus is on the need for ESH construction using period of record (POR) hydrological conditions to determine the amount (in acres) of ESH construction needed during a seven-year period. Forecasting under four water scenarios indicates no immediate need for ESH construction; however, some construction may be needed in the (near) future if dry regional conditions persist. The AMCR might provide a clearer link between the need to create degraded ESH and when to create new ESH. If models indicate that conditions do not require creation of new ESH -- as in 2020 -- what conditions call for herbicide or other treatments to be used on degraded ESH? Can conditions be identified (decision criteria) when mechanical treatments to

ISAP review of the 2020 AMCR Page 7 of 29 restore degraded ESH might be used instead of creating new ESH, and incorporate that information into the report? Forecasting for other management actions is not summarized in this section, leaving relevant questions unaddressed – for example, what motivates the need to manage flows for plovers? What key parameters are being monitored and discussed during weekly calls? And, what are the decision points that result in (a) a decision to leave flows as they are, versus (b) a decision to move nests? Effective AM can only occur when there are clear and objective decision criteria; in the absence of clearly articulated decision criteria the management can only be described as ad hoc.

Implementation of the “hybrid monitoring plan” (Section 2.4.3 in Appendix) has been delayed with a revised plan to implement it in 2022. That delay in planned implementation is unfortunate, because a well-designed monitoring plan is the cornerstone of adaptive management. The ISAP reviewed the Monitoring Plan for Piping Plovers (Charadrius melodus) on the Missouri River (dated 15 October 2019) and received no response to any of the panel’s observations, suggestions, and recommendations in that review. Furthermore, no methodological changes were made to the tern and plover monitoring protocol in 2020. The panel is concerned that a monitoring re-design is underway that is not addressing concerns registered by the ISAP more than a year ago. That ISAP review needs a response in some MRRIC program venue. In the meantime, the panel notes a large variance component associated with fledging counts that is inconsistent over measures of plover and tern density (adult pairs/acre), possibly resulting in some underlying statistical issues with model fitting and prediction. The panel is uncertain if the hybrid monitoring plan will address the inconsistency. The ISAP looks forward to a technical discussion of the planned hybrid monitoring approach and questions in advance whether the approach is likely to increase the sample of fledging pairs at high density. The ISAP further observes that the plover and tern AMCR section does not provide a discussion related to the effectiveness and efficacy of current monitoring methods in supporting AM goals. A discussion of survey design aspects that can be improved, data needs to support the AM process, and sample-size sufficiency belongs in the AMCR.

Without delving deeply into the data-collection methods, the panel has discussed several points pertinent to monitoring that would offer potentially valuable subjects for discussions with the piping plover technical team. As an example of issues of concern, the panel observes that the plover and tern negative binomial models in AMCR Figures 2-16 and 2-17 show a decreasing relationship of production versus density. Examining the graphics, it appears that that decreasing trend may be due to 3 or 4 observations in the high-density range. Removing these influential observations could result in a relatively flat relationship between density and production, and therefore would change the conclusions in the text. From a statistical perspective, when the data set has a highly varying variance over the range of the independent variable (in this case density) there could be issues with quantifying parameter uncertainty (see the confidence limits on the graphics) and accuracy of parameter estimation. Examining the graphics, clearly there are more observations at lower density, which display a large variance. At issue is one of sample size at high density, and whether the actual variance in fledging production is captured with a small number of observations. We understand that the model parameter Pi is useful in that it addresses

ISAP review of the 2020 AMCR Page 8 of 29 the number of pairs observed at each site, but the parameter does not account for a lack of sufficient sample size or possibly representative samples at high density. The technical team should address these issues as a monitoring design challenge and examine the ability to obtain representative and consistent adult and fledging counts over the full range of bird densities.

Another challenge in managing piping plovers under the MRRP is the disconnect between the species’ life history attributes and the program’s planning area. Life-history theory provides a valuable lens through which species can be examined, including its basic survival and reproductive strategies, and its potential for persistence. The plover is a relatively long-lived shorebird (the mean lifespan of an adult is >5 years), thus it has multiple opportunities to reproduce during its lifetime. The cumulative breeding attempts of each adult require that the number of young reaching adulthood is ≥1 to maintain or increase the size of the population.

During the lifespan of an adult there can be multiple years when breeding is unsuccessful, or not attempted, and yet birds still replace themselves, a strategy that is well adapted to a stochastic riverine sandbar system like the MRRP region. Such a view of the species and its persistence on the landscape has not been well represented in discussions about bird-focused AM actions and that has led to the view that static population benchmarks (fledge ratio, lambda) are sufficient to characterize status of piping plovers on the Missouri River. Future conversations on this topic, including new information on the species’ movements in the Northern Great Plains (NGP) region, would prove insightful.

The ability of the draft AMCR to forecast expected future outcomes and needs for potential management actions remains limited by the lack of future climate-change and land-use-change-driven flow scenarios. The five modeled management scenarios in Section 2.5.2 provide useful ESH management need predictions for categorical flow conditions (wet, dry), but the probable occurrence of these categorical conditions is unknown given the lack of any climate change/land use change driven flow scenarios for the basin. In essence, the AMCR forecasts what to do if a year is wet or dry but does not forecast the relative frequency or magnitude of these categorical states. The forecasting does not consider scenarios, such as (a) a basin-wide shift to increased flow magnitudes that require less frequent ESH intervention, (b) conditions with prolonged droughts forecast that would require more ESH interventions, and (c) how either scenario would be offset by or interact with flow regime consequences for lake shoreline habitat versus in-channel ESH. The lack of future-casting prohibits longer-term strategic management resource positioning. Section 2.5.3 outlines how management resource planning could happen if a better understanding of flow regime probabilities was available. For example, no sandbar construction, augmentation or modification was planned for 2020 given the ESH by high flows (a wet scenario) in 2018 and 2019 (see Section 2.3.1). What is needed to support this decision is the probable frequency and duration of dry intervals that would trigger ESH interventions. It is important to emphasize that the POR is largely representative of conditions no longer present in the basin, given dramatic land use changes (including tile drainage) and climate changes that have occurred during the POR, and particularly because it excludes observed flow regimes from 2012 onwards, the time period most representative of current and likely future conditions.

3. Does the draft AMCR describe ongoing and potential management actions and their implications, including challenges and uncertainties likely to be encountered, as they

ISAP review of the 2020 AMCR Page 9 of 29 pertain to: 1) status and trends of the listed species and their habitats; 2) agency staff and partners who will implement selected actions and monitor them; and 3) stakeholders who may have concerns regarding the effects of those actions on “human considerations”?

Response comments regarding program ongoing and potential management actions and their implications are presented as follows:

(1) Regarding the status and trends of the listed species and their habitats -- the report includes a reasonably detailed summary of the status and trends of the plover and also for ESH; however, there are no clear links between them. Performance metrics are presented in Section 2-2, while management actions are in Section 2-3. The only mention of a link is in Table 2-4 under “Current state of knowledge,” where the general relationship between a management action and a population metric is stated. Uncertainties are not addressed.

(2) Concerning agency staff and partners who will implement selected actions and monitor them, the topic does not appear to be addressed in the AMCR or supporting appendices.

(3) In reference to stakeholder concerns regarding the effects of those actions on “human considerations” -- the topic is only briefly addressed (Section 2.3.8) in the context of each management action and implications for HCs, although this is possibly due to the lighter reporting for birds in an even-numbered year. As outlined in this section, in the absence of adequate monitoring, adaptive management cannot proceed effectively. There appears to be a significant disconnect between recreation interests and species management strategies. Greater coordination between agencies is required to develop adequate AM strategies regarding human-generated disturbance/destruction of nests and to assess the effectiveness of alternative strategies and impacts on both birds and recreational access. Signage restricting human entry into tern and plover nesting areas was posted in all river sections in 2020, but its effectiveness in preventing or discouraging visitor use was not formally monitored. Anecdotal information based on bird and fish work-crew observations indicates that signs had little effect on visitor use – signs are ignored. Thus, human impacts are likely to occur, but their magnitude remains to be quantified.

Model predictions can be presented in a number of forms. Incorporating uncertainty in model predictions, including using the probability of meeting or exceeding a target, is a defensible approach. However, presenting in the AMCR additional detail on how the probability of exceedance is computed would be valuable. That presentation should include information on (a) what is represented by the probability, and does it include model parameter uncertainty, (b) whether the probability is derived from simple scenario testing, and (c) whether Monte Carlo approaches are used. While using the probability of exceedance is useful and informative, it is difficult to understand without further explanation what is behind the concept of a “chance exceedance.” The AMCR should identify the variance components that are included and specifically excluded in these computations. The technical team should identify the key model parameters that are dominating the resulting exceedance probabilities and indicate whether those parameters can be accurately estimated in an AM framework.

ISAP review of the 2020 AMCR Page 10 of 29

4. Does the draft AMCR provide an updated review of the (current) status of the listed species and their habitats, and the adequacy of information from research, modeling, and monitoring to make real time decisions in an adaptive management framework (including information from published and unpublished sources)? Does the draft AMCR describe how the available scientific information has been and will be used in programmatic adaptive management decision-making?

The AMCR provides a brief account of the species’ current status, as well as updates on recent research activities. The most important new finding is that plover movements exceed previously estimated rates. Until recently the Missouri River has been assumed to involve a mostly “closed” population of piping plovers with limited rates of emigration and immigration from adjacent alkaline wetlands. As such, the adequacy of information to guide management actions on the river largely depend on local (riverine) monitoring data. A recent mark-recapture study of plovers has revealed greater rates of immigration and emigration than previously documented.

Those findings have important implications for AM actions that might be undertaken on the river and for interpreting piping plover monitoring data from the river and reservoirs. Data indicate that plovers may emigrate from the Missouri River during periods of poor habitat conditions and immigrate to the river when habitat becomes abundant. It will be critical for the technical team to determine how to best account for this net migration in annual trend data. For example, what are the relative contributions of immigration and in situ reproduction to population growth in the years following a major flood event? Similarly, how will plover population “declines” several years after major flood events be interpreted – as a true decline in the local population, or as a temporary decline until the next flood event? The implications of the USGS piping plover movement study in the NGP are evident throughout the report. The 35% increase in adult plovers in river habitat from 2019 cannot easily be explained on the basis of local reproduction, and a similar argument can be made for the more than 3-fold increase in population growth rate during the same time period. This indicates that future AMCR reports will need to consider the implications of plover movements to and from elsewhere in the NGP when interpreting demographic changes on the Missouri River. Note that if habitat metrics are good, as represented in Table 2-1, why then is the population not meeting recovery goals for both lambda and fledge ratio?

The choice of a negative binomial distribution for the sampling distribution of observed numbers of fledglings might be reconsidered. Generally, a negative binomial is used for highly variable count data, as noted in the AMCR, or to address issues associated with a high proportion of zeros in the data set. Samples resulting in zero fledglings are not seen frequently in the illustrations (e.g., Figures 2-16 and 2-27), which stimulates the question of selecting the negative binomial. Is there evidence that the expected mean and variance are not equal (that is, assumptions from the Poisson distribution)? The technical team should provide the biological reasoning for this conclusion. The choice of the underlying distribution could change the resulting model and possibly change model predictions and their interpretation.

5. Does the draft 2020 AMCR adequately address issues that were daylighted in ISAP’s review comments on the draft 2019 AMCR? Are there further refinements to the structure of the 2020 report which would better serve the information needs of programmatic adaptive management targeting the listed species and their habitats?

ISAP review of the 2020 AMCR Page 11 of 29

(Note that the bird chapter in the draft 2020 AMCR does not incorporate all of the ISAP’s recommendations in its review of the 2019 AMCR; they will be addressed in next year’s AMCR.)

Ideally, the AMCR should be a concise, stand-alone report that presents all essential information for a reader to readily access including (a) the current status of the affected populations, (b) recent trends in those populations, (c) the suite of management actions that was considered, and, importantly, (d) what management actions are planned on the basis of specific population criteria. The current AMCR is still too lengthy to meet this goal, and some essential information is buried in appendices. Continued refinements to the structure of the report are needed.

The panel’s 2019 AMCR review provided some explicit suggestions for the 2020 AMCR (see page 4 of the review for details). The 2020 AMCR structure reflects some, but not all, of the ISAP suggestions. The current report organizes the adaptive management reporting by management action -- eight of them documented in Table 2-3 -- although as noted previously, that table is missing essential information for some actions. This review recognizes that the 2020 AMCR is an even-numbered year report and lacks certain information on the AM cycle for the piping plover that should be more fully reported in 2021. Material information on adaptive management for piping plover was found in two sections of the report, which was confusing and seems unnecessary. The primary material was in Section 2, especially on pages 31-37, while a concise summary of the AM actions was found earlier in the report in Table ES-1. This presentation should be consolidated to a single section in future reports (executive summary information might be repeated, leading off the bird section itself). The AMCR should include a single table that links basic reporting metrics -- ESH, fledge ratio, population growth -- to planned AM actions and activities in the coming year. If this had been done in Table 2-1, the need to create ESH could have been eliminated and the focus of the text shifted to management actions to maintain habitat and meet predator control needs in 2021. The Panel recommends meeting with the authors of the AMCR to revise, refine, and streamline future AMCRs.

Pallid sturgeon – questions and responses

1. Does the draft Adaptive Management and Compliance Report (AMCR) provide the each action’s (current) status in the adaptive management cycle (that is, in plan and design, implement, or monitoring phases of the cycle)? For each individual management action please identify any analyses, modeling outputs, or data summaries that should receive more or less emphasis or explanatory detail in the report, given the current status of the management action in the AM cycle.

An emerging concern among the ISAP involves the implementation design of IRCs and their spatial juxtaposition to spawning habitat; spatial considerations in the placement of Interception and Rearing Complexes (IRCs) have not been fully integrated with the life-history attributes of age-0 pallid sturgeon. IRC related hypotheses should be revisited in light of recent scientific studies and ongoing hydro-geomorphic and biological sampling efforts that identified natural

ISAP review of the 2020 AMCR Page 12 of 29

“hot spots” for pallid embryo collection. For example, Hypothesis 19 (H19), states that IRCs “… serve specifically to intercept and retain drifting free embryos in areas with sufficient prey for first feeding.” Recent studies, in both the field and laboratory, show that survival and growth during the free-embryo stage is strongly dependent on remaining in the drift during growth and development, prior to exogenous feeding. During the negative rheotactic phase of larval development, larvae orient head-first as they drift downstream at rates congruent with water velocities (Braaten et. 2008). As larvae develop feeding functions and (actively) settle to the river bottom, they change their behavior by orienting head-first upstream (in positive rheotaxis) to begin searching for and feeding on prey (Mrnak et al. 2020). In fact, in extremely low- to no-velocity environments, free embryos will increase their swimming activity in an effort to re-enter (or maintain their position in) the drift. The acts of drifting and settling are distinct behavioral phases of larval sturgeon and by linking spawning and IRC channel habitats to the life-history attributes of age-0 pallid sturgeon, these important management actions will be better served.

The functional role provided by IRCs and how IRCs should be oriented or placed adjacent to known natural or constructed upstream spawning areas requires further discussion. Information highlighted in the AMCR and discussions during the AM workshop seem to be steering IRCs in a different and better direction as management actions. With just 2 of 12 planned IRC structures implemented at this point in the MRRP, discussions toward strategic refinement of spatial considerations for this management action is timely. Efforts to learn more about the hydrologic circumstances and habitat conditions where exogenously feeding age-0 sturgeon are collected (so-called “hot spots”) will provide foundational information toward improving IRC design and help elucidate why IRCs should be included as an important, targeted management action that is spatially linked to known or constructed spawning habitat. The challenge will be in determining where and how to construct them and how to refine the implementation design. IRCs should be located so they align with [known] upstream spawning areas where embryos first emerge and begin their downstream drift phase. Information on temperature-dependent development can be combined with mean water velocity downstream of spawning areas to estimate drift distance

(km) as a means for identifying approximate settling areas for exogenous-feeding larvae (Bratten et al. 2008). Similarly, upper basin and lower basin larval drift studies can provide important insights into the spatial relationship between spawning locations (e.g., in this case, release sites of 1-2 dph fish) and downstream rearing locations where EFL are collected. ISAP also notes that ongoing advection/dispersion modeling efforts has high potential to inform effective IRC placements.

While anticipating further discussion of spatial issues in identifying IRC locations, the ISAP recognizes that the programmatic assumptions, enabling equations, model descriptions, and approach to assessing the utility of IRCs as currently considered are well written and readily understood. Many of the issues that arose during the Fall Science Meetings in 2020 are discussed and addressed in the AMCR narrative. Table 3-16 is informative, displaying the overall survival rate from EFL to age 1 (S1). It should be possible to take a successful scenario (defined as achieving an S1 of 22%), and expert judgement (noting that data are not currently available) to establish a confidence range on the value. At issue is whether the estimates are conservative or not. While a relatively high interception rate is required to achieve the 22% overall survival-rate goal with 12 IRCs, depending on uncertainties and based on best professional judgement, a smaller number of IRCs might be sufficient to meet program goals. The AMCR mentions that it would be valuable to reduce uncertainty in the estimated fraction of EFL intercepted at each IRC, ISAP review of the 2020 AMCR Page 13 of 29 possibly through additional modeling or monitoring. The ISAP agrees that such an effort could be useful and could aid in an uncertainty analysis of this key demographic parameter.

Just two of the 12 planned IRC sites – at Moberly Bend and Searcys Bend -- have been constructed and no construction occurred in 2020 given the prohibition in Section 1226 of America’s Water Infrastructure Act of 2018. No new spawning habitat projects were planned or constructed in the 2020 reporting year. The USACE and USFWS are in discussions to determine whether spawning habitat construction remains a priority given evidence of successful, albeit limited, pallid reproduction in the lower Missouri River. However, the AMCR provides a detailed description of the plan for evaluating IRC effectiveness (see Section 3.3.3 Channel Reconfigurations for IRCs, Lower River) as well as HC impact evaluation using a study design with control-treatment comparison reaches. Regardless, evaluating the effectiveness of management actions in the Missouri River is likely to be presently limited and may require using shovelnose sturgeon as a surrogate species in assessing IRC performance, although the validity of a surrogate relationship between the two sturgeon species has not been fully tested (but see Gosch et al. 2018). This surrogate issue requires an examination of the existing survey design, with the objective of optimizing sampling effort to provide the greatest return per investment.

Statistically, this means exploring survey designs that maximize the ability to catch 0/1 age fish and adults, if they are in fact located in the sampling area.

The AMCR would be strengthened with discussion of the sufficiency of current monitoring and modeling activities to support the AM decision-making process. The monitoring in support of the AM program attempts to provide data in support of statistical descriptions of population status and to some extent trends; the monitoring also provides data to support the development and implementation of population models. It remains to be demonstrated that current survey designs are equally effective in supporting both statistical descriptions and modeling. If the emphasis is on statistical characterization of population dynamics, current monitoring should further consider and possibly revise and refine survey design in relation to sampling uncertainty, minimizing non-detects (where, for example, fish occupy the sampled area, but are not collected by current sampling procedures), representative sampling of known habitat, and alterations in environmental conditions that can influence sampling effectiveness.

If the emphasis of monitoring is to provide data to support the population modeling effort, further attention needs to be given to survey designs informed by specific parameter estimation requirements and directed towards increasing accuracy and precision in estimated parameters.

The main point is that a single survey design might not prove sufficient to support statistical and modeling descriptions of population dynamics, particularly dynamics in relation to implemented management actions. The monitoring might realistically require multiple sampling approaches and methods of data analysis, each focused on specific questions including for example trend detection or spatial distribution and abundance.

Continuing this line of inquiry, the ISAP has previously suggested that stratified random sampling may be a useful approach for improving pallid sturgeon monitoring. However, after extended reflection, the ISAP believes the stratified design may have underlying statistical properties that should be considered and studied prior to implementation. Stratification serves to minimize overall sampling variance associated with the population mean. However, if appropriate strata are not selected, and the statistical weighting of these strata is in error, the resulting population estimated mean will be biased and the estimated population variance will be

ISAP review of the 2020 AMCR Page 14 of 29 incorrect. For example, the Missouri River CPUE sampling could stratify on the lower and upper basins. But that stratification implies complete random sampling over a large span of river miles.

The current approach of sampling bends is inefficient, and can produce zero-dominated data sets, either by incorrect location selection or inadequate sampling methods. Resulting population estimates are difficult to interpret; for example, is a zero representative of no catch, but the fish are in fact present, or is zero an accurate measure of true fish absence? For stratified random sampling, underlying theory concerning unbiased conversion of the sample mean to the population mean does not hold. In stratified random sampling, each fish in a stratum must have the same probability of detection, or again the population estimates will be statistically biased.

For a rare event (e.g., catching an age-0 pallid sturgeon), the underlying theory of a stratified random sample again does not hold. A panel survey was outlined as a possible alternative approach in the AMCR. While this approach reduces effort and cost, the panel survey does not directly overcome the above statistical issues.

Finding an ideal survey design that is sufficient for all of the AM questions of interest will be difficult. It’s possible that each question of interest may require a separate approach, which may include surveys designed to estimate specific parameters in a population model(s). Or, for a specific statistical model, a formal model-based survey may be optimal. The ISAP recommends a technical working group be created to work through these design issues. To facilitate discussions, the ISAP recommends that monitoring data used to support population analyses and conclusions in the AMCR be made readily available, either as an appendix to the AMCR or electronically.

2. Does the draft AMCR adequately forecast expected (future) outcomes of ongoing or potential management actions for purposes of decision making in the adaptive management process?

The ability of the draft AMCR to forecast expected future outcomes of ongoing or potential management actions for purposes of decision making in the adaptive management process remains limited by the lack of flow scenarios that account for past and projected future climate and land use changes in the watershed. The 2020 AMCR begins with a discussion of basin conditions in the Missouri River noting that 2020 started with wetter-than-average conditions for the reporting year. Runoff into the reservoir system above Sioux City, Iowa was 31.3 million acre-feet (MAF) or about ~121% of average. Figure ES-1 illustrates that over the last 120 years, the four wettest years on record have occurred in the last nine years. Although not obvious from Figure ES-1, there is an important point to be made about the frequency and magnitude of high/low runoff events, which appear to be increasing in the Missouri River basin (Livneh et al.

2016; Figure 1). High flow conditions can constrain implementation of the MRRP and as pointed out in previous…