View the details of the Independent Scientific Review Panel (ISRP) assessment for this project as part of the Resident Fish, Regional Coordination, and Data Management Category Review.
Assessment Number: | 1995-027-00-ISRP-20120215 |
---|---|
Project: | 1995-027-00 - Lake Roosevelt Sturgeon Recovery |
Review: | Resident Fish, Regional Coordination, and Data Management Category Review |
Proposal Number: | RESCAT-1995-027-00 |
Completed Date: | 4/13/2012 |
Final Round ISRP Date: | 4/3/2012 |
Final Round ISRP Rating: | Meets Scientific Review Criteria (Qualified) |
Final Round ISRP Comment: | |
Most of the responses to ISRP questions were adequate. Positive responses from the sponsors included summary updates for project results (2009-2011) and a description of expertise and roles of existing project personnel. The sponsors provided detailed information, including a good diagram, of how this project relates to and coordinates with project #200811600. It has now been made clearer to the ISRP which entities are leading the work in various areas. The ISRP had requested additional information on criteria for identifying stock rebuilding. However, no additional information was provided. The objective is simply to stock plenty of fish, and if it turns out to be too many, fish can be thinned through harvest. This is one approach, but a more plausible scientifically-based rebuilding schedule needs to be formulated. The ISRP requested more detailed methods and approaches for several tasks outlined in the proposal, including methods for determining (1) if predation on juvenile sturgeon was cause for recruitment failure and (2) if lack of proper food was the cause of starvation and recruitment failure. These were not included in the response. Instead, the sponsor’s response was "The LRSRP appreciates that the ISRP recognizes the complexity of the recruitment failure issue in the transboundary reach and the difficulties associated with identifying the limiting factors. The LRSRP recognizes the importance of designing detailed study approaches in order to objectively answer recruitment failure questions. The LRSRP plans to retain a subcontractor with appropriate expertise to assist with study design including detailed methods and implementation of the predation and food habits components of this project. The completion of this work is contingent upon funding." The sponsor stated that it plans to hire a subcontractor when funded to assist with study design and methodology involving predation and food web components of the project. A specific subcontractor was not identified in the proposal. It is highly desirable for a scientific proposal to identify key individuals or groups that would be responsible for such a major contribution to the study, to indicate that that they had been contacted, and for them to perhaps provide some indication of hypotheses and appropriate methodologies used to test the hypotheses. |
|
Qualification #1 - Qualification #1 - identify and hire the subcontractor
For the predation and food web components of the project, the sponsor needs to identify and hire the subcontractor, identify qualified staffing additions to conduct the work, and develop detailed methodologies, including the starvation approach. The ISRP should review the specific objectives and methodologies prior to implementation.
|
|
Qualification #2 - Qualification #2 - develop a plausible rebuilding schedule for the stock
The sponsor needs to develop a plausible rebuilding schedule for the stock with production and cohort/age structure goals during contracting. Similar work by other entities, including the Kootenai Tribe, should be reviewed for applicability.
|
|
Qualification #3 - Qualification #3 - High quality annual reports need to be completed and updated.
High quality annual reports need to be completed and updated.
|
|
First Round ISRP Date: | 2/8/2012 |
First Round ISRP Rating: | Response Requested |
First Round ISRP Comment: | |
Despite a reasonably sound and useful overview of sturgeon problems in this portion of the basin, several items in the proposal are in need of clarification. First, although the need for more understanding of the recruitment failure is well articulated and on target, designing studies to address this issue is an extremely difficult process and requires a more detailed, critical approach than is outlined here. The goals of the recruitment failure work are laudable. However, because of a lack of detail provided in the proposal, the ability of the proposed work to answer the key questions and meet those goals is very questionable. For example, it is not enough to answer if some sturgeon are eaten; it must be shown that this is a cause of the recruitment failure. Similarly, it is not enough to look at some fish food habits and invertebrates; it needs to be shown objectively that these factors are a cause for the recruitment failure. These are difficult questions to answer. Detailed study approaches are needed.
1. Purpose: Significance to Regional Programs, Technical Background, and Objectives The Lake Roosevelt Sturgeon Recovery Project (LRSRP) is an ongoing project implemented to monitor population status and conduct recruitment failure research on white sturgeon in the Roosevelt Reach of the upper Columbia River. The primary goal of the project is to conserve and restore white sturgeon in Lake Roosevelt and the upper Columbia River. Their hypothesis is that sturgeon cannot get from the hiding to rearing life stage due to contaminants and predation. This project is closely linked to project 200811600. The significance of this project is described as responding to many regional plans and programs including: the Lake Roosevelt Guiding Document and Management Plan, the Upper Columbia White Sturgeon Initiative and Recovery Plan (2002), the NPPC Fish and Wildlife Program (2009), the Spokane Subbasin Plan, the MERR Plan, and others. The technical background is described in the Problem Statement and is extensive, contains many good references, and provides sufficient detail regarding the current status and problems of white sturgeon populations in the Columbia River Basin. Past work is well described. The objectives are listed as: OBJ-1: Prevent further reduction in upper Columbia River sturgeon distribution, density and genetic diversity by implementing LRSRP/UCWSRI long-term measures. OBJ-2: Implement research examining hypotheses to determine the cause of upper Columbia River white sturgeon recruitment failure. These objectives are too general. The deliverables which follow, however, are the work elements and are at a reasonable level of detail. Even though the importance of rebuilding white sturgeon in the Upper River is well documented and well-justified in this proposal, some details remain vague. The intent to "rebuild the natural age-class structure lost during the recruitment failures of the last 30 years (UCWSRI 2002, Recovery Plan Measure 5.5.3)" raises the question of whether this historical age structure is well-known, or is it just assumed that recruitment occurred every year or nearly every year? The data in Figure 2 demonstrate the lack of recruitment very clearly but do not necessarily indicate steady recruitment of the past. It is also not separated by sex, so with sexual size dimorphism of sturgeon, it gives little indication of actual yearly recruitment. Care must therefore be used in designing the stock composition target that is part of the rebuilding effort. There seems to be no evidence that recruitment in this section of the river was necessarily a yearly event or even very consistent. Regarding the goal of 1000 mature individuals in an approximately 1:1 sex ratio at maturity, there will of course not be such a sex ratio at maturity because the males will mature several years before the females so there will be more mature males from a cohort starting at a 50-50 sex ratio. 2. History: Accomplishments, Results, and Adaptive Management (ISRP Review of Results) Overall, the sponsors did a nice job providing the status of sturgeon populations for this part of the system. Accomplishments are adequately summarized in the proposal. However, there has been a serious lack of reporting since the last ISRP review of this project. Completed reports are lacking for three recent years. Although the ISRP lauded the reporting in the previous review, the major lapse in reporting since then is cause for concern about project direction. Regarding growth, the authors reported that "the estimate of growth co-efficient, K, was substantially greater in magnitude, and resulting growth trajectories predicted that sturgeon in the Roosevelt Reach attain larger sizes at younger ages than observed in other areas of the Columbia River (Howell and McLellan in prep; Figure 5)." It is confusing as to why the growth of the Lake Roosevelt fish would be faster than others but from Keenleyside slower than the others. This does not appear to make sense. Clarification would be helpful. ISRP Retrospective Evaluation of Results The sponsors present a thorough review of sturgeon activities to date and do a reasonable job of focusing in on the knowledge limitations remaining. There does not seem to be a 2009, 2010, or 2011 Annual report, and there does not seem to be much, if any, history of refereed publications resulting from this long project. The ISRP will expand on its retrospective analysis following the response. 3. Project Relationships, Emerging Limiting Factors, and Tailored Questions for Type of Work (hatchery, RME, tagging) The sponsors provided a good description of how this project relates to and coordinates with many other BPA projects plus state and Canadian programs. Potential limiting factors are identified in the review section. There is a general description of how the sponsors are aware of emerging limiting factors such as non-natives and predators. Adequate answers were provided to the tailored questions. Tagging descriptions were provided in good detail. Database development and sharing is described in reasonable detail. It is not clear exactly who is doing the work. There is no recent annual report to clarify this issue. The sponsors state, "Stock assessment study design, analysis, and implementation will be led by the Spokane Tribe Lake Roosevelt Sturgeon Recovery Project (1995-027-00) in Washington and by BC Hydro in British Columbia. The CCT, under the White Sturgeon Enhancement Project (BPA project 2008-116-00), will provide a support role in population monitoring by providing a field crew, equipment, and technical advice. CCT participation will help increase sampling effort, and sample sizes, to improve precision of capture-recapture abundance and survival estimates, as well as indices of growth and condition to facilitate inferential statistical analysis." The one person mentioned as involved in this study is from neither entity but from WDFW, even though this is a Spokane Tribe proposal. The roles of the participants in achieving each objective are unclear and need to be clarified. Who exactly is doing various portions of this work? Some duplication appears to occur in 200811600 with regard to database management. Both agencies have sizeable budgetary resources dedicated to this effort, although it is the lead of 2008-116-00. The roles here need to be more clearly defined. We would request a diagram showing how work elements proposed under 200811600 and this project are to be divided up. There seems to be some duplication, and a diagram may show otherwise. 4. Deliverables, Work Elements, Metrics, and Methods Details of several work elements are not clearly articulated. For example, regarding predation: "Under the recruitment failure hypotheses assessment completed by the UCWSRI, several potential proximate mechanisms have been identified as potentially limiting survival of white sturgeon in the recovery area. The LRSRP proposes to examine predation on white sturgeon early life-history stages (ELS) by conducting diet analyses on predators collected from the transition zone from July through October using a combination of short duration gill net sets and by trawling with sampling being stratified by depth and by time of day." This approach intends to sample fish and look at stomachs, but it is not clear how the presence or absence of sturgeon will be translated into a quantitative assessment of the effect of predation and therefore on recruitment failure. The collection of the data is much more direct than the translation of the results into a predation effect on recruitment failure, and the approach should be described in more detail. Similarly, with regard to food limitation, the sponsors state, "We also intend to compare histology of post-feeding stages of white sturgeon collected during field surveys with reference specimens to identify starvation effects in wild fish, thus determine the role food availability plays as a limiting factor in sturgeon survival." Has this approach been used successfully elsewhere? Please provide background and references. Although year class strength has at least tentatively been associated with higher flows, it does not seem that any recommendations have been forwarded to test flow augmentation during late spring early summer to improve natural reproduction and recruitment. Has this topic been adequately investigated? Have recommendations been made? There may be some value in using otoliths to find hatching dates for larval fish. More details are needed of the contaminant work to be performed and the protocols and methods. 4a. Specific comments on protocols and methods described in MonitoringMethods.org The protocols and methods have been entered in MonitoringMethods.org. The basic protocols and methods were fairly complete and the level of detail for the methods is almost sufficient to be able to replicate the study data collection. Methods outlined in these sections often do not greatly exceed in detail that presented in the proposal. More details of the proposed experimental designs for predation should be provided. What might be the role of sculpins and how might it be evaluated? Under food resource availability, no details of methods are provided. How can sampling some stomachs clearly lead to conclusions regarding possible "starvation"? Have the sponsors clearly visualized and laid out how the starvation hypothesis can be evaluated? Modified by Dal Marsters on 4/13/2012 4:39:12 PM. |
|
Documentation Links: |
|
Proponent Response: | |
SPONSOR RESONSE TO ISRP COMMENTS: Resident Fish Categorical Review. Lake Roosevelt Sturgeon Recovery Project (BPA 1995-02-700) Sponsor: Spokane Tribe ISRP recommendation: Response Requested 1. Comments: Results are lacking for 2009-2011. Annual reports are behind schedule. Two or three years of data may change the direction and rationale of the study. The sponsors should provide a summary of recent results to the ISRP for review. RESPONSE: Data reports for years 2009-2011 are currently in preparation. Activities and project results through 2009 were largely summarized in the proposal. The proposal also included partial results for the period 2010-2011 (e.g. the development of larval collection as the preferred basis for the conservation aquaculture program, VPS pilot study). Information collected 2010-2011 that was not included in the proposal does not affect proposal direction and rationale. Brief summaries of activities and results of note that were not outlined in the proposal follow below. Review of female length at maturity: We reviewed sturgeon sex and maturity data collected during various LRSRP surveys 2003-2009 for assessment of FL at maturity i.e., the FL at which 50% of a cohort spawn for the first time. Generally, sex and reproductive status of sturgeon was assessed macroscopically in the field via otoscopic gonad inspection through surgical incision, and classified according to Conte et al. (1988). The analysis was restricted to females since few fish < 150 cm FL were examined for maturity, and average length at maturity in males is likely less than 150 cm FL. Thus, all fish unambiguously identified as males were excluded from the analysis. For cases where sex and maturity state could not be determined, fish were classified as immature (non-reproductive) females for the analysis recognizing that a small proportion of these fish could actually be males. All females classified as stages 13 (early vitellogenic) through 16 (post spawn) were considered mature (i.e. had spawned at least once previous to examination or, if virgin, were expected to spawn relatively soon after examination). Because during macroscopic examination of the ovaries it is difficult to differentiate pre-vitellogenic (stage 12; moderate sized ovary with small eggs present) females that have previously spawned and are in a resting state from virgin fish that will spawn for the first time at some unknown time in the future, females classified as stage 12 were alternately treated as mature or immature in two separate analyses. Female maturity was modeled as a logistic function of fork length using ratios of immature and mature fish:
Where M = proportion mature, x = fork length, x0 = fork length at 50% maturity, and a and b are model coefficients. Model parameters were estimated using non-linear least squares in SigmaPlot (v11.1.0.102; Systat Software, Inc). LRSRP staff conducted 277 field assessments of sex/maturity on 266 individual sturgeon captured in the Roosevelt Reach from 2003-2009. Sturgeon ranged in size from 94.0-260.5 cm FL (mean=191.3; n=277). Males ranged in size from 100.0-259.5 cm FL (mean=181.9; n=84), females ranged 148.0-260.5 cm FL (mean = 204.9; n=127), and fish where sex could not be determined ranged 94.0-237.4 cm FL (mean=177.0; n=66) (Figure 1). One fish of uncertain sex was excluded from the female length at maturity regression analysis based on its size (94 cm FL – i.e. a juvenile). Thus the total number of observations that were used in the analysis was 192 representing fish that were 141.0-260.5 cm FL (mean=195.9) at time of examination. Estimates for length at female maturity where stage 12 (pre-vitellogenic) females were treated as either mature or immature were similar at 183.3 and 189.8 cm FL, respectively (Figure 2). These estimates are similar to those published for female sturgeon in other impounded areas of the Columbia River (Beamesderfer et al 1995). Also, when applied to length-at-age growth models based on Roosevelt sturgeon fin-ray aging and mark-recapture data, estimates of length-at-maturity correspond to an age of approximately 30 years.
Figure 1. Length frequency plots for sturgeon examined for sex/maturity in the Roosevelt Reach 2003-2009.
Figure 2. Proportion mature female sturgeon by length (10 cm length bins) in the Roosevelt Reach 2003-2009 showing fitted logistic curves and parameter estimates. Blue lines denote 95% confidence bands and red lines denote 95% prediction bands. The upper plot assumes all pre-vitellogenic fish are mature, the lower plot assumes them to be immature.
Brief review of acoustic telemetry 2003-2009 From 2002-2005, LRSRP staff and researchers in Canada collectively outfitted 78 white sturgeon (104.0-256 cm FL; mean = 189.7) in the Transboundary Reach with Vemco V16 acoustic tags (nominal four year tag life). These tagged fish were tracked with a longitudinal array of Vemco VR2 and VR2W receivers spaced at varying intervals from HLK Dam downstream to the Columbia-Spokane confluence (the number of array nodes and node placement has developed over time). From 2002-2009, 76 of these fish were detected by the transboundary array. The duration of detection histories for another three of these fish were < 1 year suggesting fish death or premature tag failure. Of the remaining 73 fish, detection history durations ranged from 1.84-6.32 years (mean = 3.58). While at large, the detected longitudinal range of these fish within the Transboundary Reach ranged 0.0-186.4 km (mean = 71.4)(Table 1). Excluding movements during apparent spawn migrations, most fish habituated to either the Keenleyside or Roosevelt Reach for the entire duration of their tag’s life; only four (5.5%) fish utilized habitat in both reaches for substantial (months to years) periods of time outside of spawning timeframes. The longitudinal range of fish (including spawn migrations) that habituated to the Keenleyside Reach was substantially less than observed for fish habituating to the Roosevelt Reach (Table 1).
Table 1. Detected longitudinal ranges of movement for 73 sturgeon outfitted with Vemco V-16 acoustic tags in the Transboundary Reach 2002-2005. Fish ranged in size from 104 to 256 cm FL (mean=189.3). Detection history durations ranged from 1.84 to 6.32 years (mean = 3.58).
While at large, only two fish (2.7%) were detected as far downstream as the Columbia-Spokane confluence (rkm 1,029) and only 12 (16.4%) were detected downstream from Gifford (rkm 1,093). Thus telemetry observations are consistent with setline catch data that indicates sturgeon use of the, approximately, lower half of the Roosevelt Reach is limited (Figure 3, Figure 4).
Interpretation of spawning movements is under review as of this writing and detailed data will not be presented here. However, a preliminary assessment suggests forty nine fish made clear migrations to known or suspected spawning areas during spawn timeframes and twenty two were observed to make multiple spawning migrations while their tags were active. Of these, fourteen were females (169-256 cm FL at tagging) that undertook migrations at intervals ranging from two to four years (two year interval, n=4; three year interval, n=8; four year interval, n=2). A female that undertook spawn migrations at two year intervals is illustrated in Figure 5. Most fish that did not apparently undertake clear spawn migrations (primarily fish habituating to the Keenleyside Reach) were reproductively mature when tagged suggesting they may spawn at locations close to their general area of residence. Spawning intervals for these fish can therefore not be readily determined.
Figure 3. Summertime sturgeon relative abundance by river kilometer in the Roosevelt Reach during spatially balanced random setline surveys, 2007-2009. CPUE is the mean (SE error bars) catch per overnight set and Ep is the proportion of sets that captured sturgeon. Data are arranged in 10 km bins and are relative to the Highway 395 Bridge near Kettle Falls, WA (rkm 1,136). Note differences in CPUE scaling among individual plots. Labeled vertical dotted lines denote locations of landmarks for reference purposes.
Figure 4. Frequency plots (black bars) of furthest downstream detection for 73 sturgeon outfitted with V16 acoustic tags in the Transboundary Reach 2002-2005. The grey line represents the cumulative frequency of the count data moving upstream and roughly indicates what proportion of the tag group were detected downstream from any particular location.
Figure 5. Detected movements of acoustic tagged female ID 1848 in the Transboundary Reach 2005-2009. This fish was 238.5 cm FL and in pre-spawn condition when tagged at Waneta Eddy 24 June 2005. Subsequent migrations to the Waneta during the spawning timeframe suggest this female spawns, or is capable of spawning, on a two year cycle and shows fidelity to the Waneta spawning area. 2010 LRSRP activities and results (report in prep.):
Annual fall gill net sub-yearling recruitment survey:
Pilot contaminants experiment: In 2010 we wished to gain experience in incubating sturgeon embryos in preparation for calcein marking experiments slated for 2011. Out of this grew a pilot study looking at the gross effects of exposing sturgeon embryos and free-embryos to industrial slag of the type found in the substrates of the upper portions of the Roosevelt Reach. We collected (ponar dredge, towed plankton net) samples of industrial slag from the river thalweg at China Bend where post-hatch life stages are known to occur. Water used in the experiment was sourced from the river at China Bend. Experiments were conducted at the WDFW region 1 fish lab. Sturgeon used for the experiments were sourced from broodstock collected in 2010 and were excess to the aquaculture program requirements.
Two experiments were conducted (see Figure 6):
Figure 6. Modified Heath trays used as experimental enclosures for LRSRP pilot contaminant experiments in 2010. Left: to limit density dependent effects, a perforated plastic sheet was employed to evenly distribute embryos during incubation. Right: Heath tray filled with industrial slag in preparation for introduction of recently hatched free-embryos. 2011 LRSRP activities and results (report in prep.): Conservation aquaculture
Table 2. Summary of D-ring plankton net post-hatch sturgeon catch in the upper China Bend area of the Roosevelt Reach in 2010 and 2011.
Pilot calcein marking experiment:
Northport spawning area Vemco Positioning System (VPS) pilot study:
Figure 7. Map of VPS receiver locations as deployed in the Northport area in 2011.
Figure 8. Depth detection plots from six adult sturgeon outfitted with pressure sensor acoustic tags that were present in the vicinity of the Northport VPS array during the spawning timeframe in 2011.
Fall trawling for sub-yearling juveniles
Annual fall gill net sub-yearling recruitment survey:
The lack of YOY captures in the fall of 2011 during the targeted trawl and gill net efforts is particularly interesting since the spring discharge peak at the international border that year was the highest observed since 1997 – a year which produced a detectable recruitment event. Although spring discharge was high, water temperatures at the international border lagged behind normal (including 1997) in 2011 and this apparently delayed spawning by approximately two weeks based on the results of spawn monitoring efforts in Canada. This was confirmed by LRSRP D-ring larval collection efforts (for the aquaculture program) where first feeding larvae were not observed until 20-21 July and were not present in numbers until 25 July. Previous LRSRP early life history D-ring surveys found first feeding larvae to be present in numbers by early-mid July (see plots in proposal). By the time large numbers of first feeding larvae were present in 2011, discharge had declined substantially from peak and was not substantially greater than observed during the larval migration period in “normal” water years. The secondary peak in discharge in mid-July 1997 that likely coincided with peak larval dispersal (the water temperatures profile at the border in 1997 was similar to “normal” years) exceeded 200 kcfs. By contrast, river discharge in 2011 had declined to 145-160 kcfs by the time first-feeding larvae were becoming abundant. Thus, the apparent lack of recruitment to sub-yearling juveniles in 2011 as indicated by the fall gill net and trawl surveys adds weight to our contention that the primary determinant of recruitment in the transboundary sturgeon population is hydrologic conditions during the larval dispersal phase, not the spawning phase as has been suggested for other areas of the Columbia.
2. Comments: More rigorous and detailed methods and approaches are needed for several of the tasks outlined, specifically describing how the field work will translate into answered questions. Refer to the next paragraph for issues to address in the response.
Despite a reasonably sound and useful overview of sturgeon problems in this portion of the basin, several items in the proposal are in need of clarification. First, although the need for more understanding of the recruitment failure is well articulated and on target, designing studies to address this issue is an extremely difficult process and requires a more detailed, critical approach than is outlined here. The goals of the recruitment failure work are laudable. However, because of a lack of detail provided in the proposal, the ability of the proposed work to answer the key questions and meet those goals is very questionable. For example, it is not enough to answer if some sturgeon are eaten; it must be shown that this is a cause of the recruitment failure. Similarly, it is not enough to look at some fish food habits and invertebrates; it needs to be shown objectively that these factors are a cause for the recruitment failure. These are difficult questions to answer. Detailed study approaches are needed. RESPONSE: The LRSRP appreciates that the ISRP recognizes the complexity of the recruitment failure issue in the transboundary reach and the difficulties associated with identifying the limiting factors. The LRSRP recognizes the importance of designing detailed study approaches in order to objectively answer recruitment failure questions. The LRSRP plans to retain a subcontractor with appropriate expertise to assist with study design including detailed methods and implementation of the predation and food habits components of this project. The completion of this work is contingent upon funding.
3. Comments: Information that details how work elements proposed under 200811600 and this project are to be divided up should be provided in the response.
RESPONSE: As described in our proposal, the white sturgeon recovery effort in the upper Columbia River is an international endeavor with loose oversight by the Upper Columbia White Sturgeon Recovery Initiative. The Washington component is comprised of the collective efforts of two BPA funded projects: the White Sturgeon Enhancement Project (WSEP) (BPA 200811600) sponsored by the Colville Confederated Tribes (CCT) and the Lake Roosevelt Sturgeon Recovery Project (LRSRP) (BPA 199502700) sponsored by the Spokane Tribe of Indians (STOI) (Figure 9). There are three main categories of work that are proposed to be completed under the Washington component of the recovery effort, which are the core program, habitat assessment, and recruitment failure research. Within the core program, there is population monitoring, which consists of stock assessment with setlines, recruitment monitoring with gill nets, and telemetry, conservation aquaculture, and data management (Figure 9). The habitat assessments include the hydrographic survey to produce the bathymetric and sediment facies maps, velocity profiling, and water quality monitoring. The recruitment failure research includes studies to investigate the proximate mechanisms of recruitment failure (mortality), such as predation, starvation, and contaminant toxicity. The recruitment failure research will also examine the larval transport hypothesis by conducting habitat modeling using the habitat assessment data and by empirical testing by conducting larval release experiments. The larval release experiments are dependent on the ability to mass mark larval sturgeons, which will be studied prior to implementation of the release experiments (Figure 9).
Figure 9. Overview of proposed work to be completed by the LRSRP (199502700) and WSEP (200811600) projects including specific tasks. Diagrams divided by each of the major work categories (core program, habitat assessment, and recruitment failure research) that indicate the responsibilities of each sturgeon project are provided in Figures 10-12. Under the core program, the LRSRP (BPA 199502700) will lead the population monitoring and conservation aquaculture, and the WSEP (BPA 200811600) will lead development and maintenance of the data management system (Figure 10). The LRSRP proposal describes the level of effort needed to conduct mark-recapture experiments to estimate survival and abundance of wild and hatchery sturgeons. The LRSRP has adequate equipment and staff to complete two thirds of the level of effort necessary. Thus, the CCT through the WSEP will support the stock assessment by supplying a boat and crew to complete approximately one third of the setline and gill net effort (see green arrows, Figure 10). Study design, analysis, and reporting will be completed by the LRSRP. Due to the large amount of data, wide variety of data types, and involvement of multiple parties collecting and accessing data, the development of the data management system will require a development team greater financial support than can be provided by a single project. While the CCT through the WSEP will lead this effort, support in the form of funding, data transfer, and participation on the development team is to be provided by the STOI through the LRSRP (see black arrow, Figure 10).
Figure 10. Core program tasks proposed to be completed by the LRSRP (199502700) and WSEP (200811600) projects.
The LRSRP will complete the water quality monitoring component of the habitat assessment (Figure 11). The WSEP will be responsible for the hydrographic survey and resulting bathymetric and sediment facies maps, as well as the velocity profiling.
Figure 11. Habitat assessment tasks proposed to be completed by the LRSRP (199502700) and WSEP (200811600) projects. As previously indicated, there are two ways that the projects propose to research recruitment failure: 1) examination of proximate causes of mortality (i.e. predation, starvation, contaminant toxicity), and 2) determining if larvae are being transported to appropriate habitats. In order to assess proximate causes of mortality, the LRSRP (BPA 199502700) will investigate predation on sub yearling sturgeon, as well as food availability for sub yearling sturgeon (Figure 12). The WSEP (BPA 200811600) will examine the effects of contaminants on sub yearling sturgeon. We proposed two approaches to investigate the larval transport hypothesis: habitat modeling and empirical testing. The WSEP will be responsible for conducting the habitat modeling to determine if there are differences in the quantity of suitable habitat during years with and without detectable recruitment (Figure 12). The WSEP will also be responsible for empirically testing the larval transport hypothesis by conducting larval release experiments. The LRSRP will experiments to determine if sturgeon larvae can be mass marked with Calcein (Figure 12). The mass marking experiment will be completed before the release experiments, because the release groups need to be differentially marked (see green arrow, Figure 12). Following the completion of mass marking experiment, and assuming positive results, the WSEP will apply the marking procedure and conduct the release experiments.
Figure 12. Recruitment failure tasks proposed to be completed by the LRSRP (199502700) and WSEP (200811600) projects.
4. Comments: A description of the expertise and specific roles of personnel should be provided so responsibility for every project activity is clear. RESPONSE: Historically, the LRSRP subcontracted Washington Department of Fish and Wildlife (WDFW) to lead the project as they had biologists with sturgeon expertise available out of the Region 1 offices. However, staffing changes at the co-management agencies has created a need to restructure, so the project is currently experiencing a transitional phase in regards to staff. During the 2013-2017 funding cycle, the LRSRP will provide a qualified full-time biologist, one FT technician and one part-time technician out of the Spokane Tribe of Indians Department of Natural Resources. WDFW will continue to provide subcontracted support with one FT biologist and one PT technician. In addition, the CCT will continue to provide support through the White Sturgeon Enhancement Project (200811600). 1. Purpose: Significance to Regional Programs, Technical Background, and Objectives Even though the importance of rebuilding white sturgeon in the Upper River is well documented and well-justified in this proposal, some details remain vague. The intent to "rebuild the natural age-class structure lost during the recruitment failures of the last 30 years (UCWSRI 2002, Recovery Plan Measure 5.5.3)" raises the question of whether this historical age structure is well-known, or is it just assumed that recruitment occurred every year or nearly every year? The data in Figure 2 demonstrate the lack of recruitment very clearly but do not necessarily indicate steady recruitment of the past. It is also not separated by sex, so with sexual size dimorphism of sturgeon, it gives little indication of actual yearly recruitment. Care must therefore be used in designing the stock composition target that is part of the rebuilding effort. There seems to be no evidence that recruitment in this section of the river was necessarily a yearly event or even very consistent. RESPONSE: Attempts at identifying past levels of annual recruitment are limited by well documented ageing errors (which reduces the accuracy of age-based models) and highly variable growth rates (which reduces the accuracy of length-frequency analyses). Include other poorly described variables such as age-at first maturity, spawning intervals, number of available spawners, total annual egg production, mortality rates, etc. and the likelihood of being able to accurately predict past recruitment rates becomes even less likely. Due to these difficulties, the approach taken by the UCWSRI in the first decade of hatchery supplementation has been to ensure that stocking rates are conservative; we can always reduce the population in the future (through selective harvesting) if we have overstocked but we can’t go back in time and stock more fish if we have understocked. As stated in the proposal, we do plan on monitoring hatchery juveniles annually to 1) track density dependent effects on growth as a means to assess our stocking levels and 2) obtain more accurate estimates of post-release survival. Therefore, although the reviewer’s concerns are valid, we currently don’t have sufficient data to address this issue. It is possible that we could look at more natural population structures in the lower Columbia or the Fraser as a means to assess the variability in annual recruitment but those are much different systems than the upper Columbia River, so how valid any extrapolations might be is unclear. Ageing data for Roosevelt Reach sturgeon is limited in extent. During a setline survey of the Roosevelt Reach in 1998, fin rays from 124 sturgeon were aged (Devore et al 2000). No sex information was collected from these fish. A year class distribution and approximated recruitment index for this data are shown in the figure below. The plot suggests that for the period between the completion of Grand Coulee Dam (1942) and commencement of mainstem dam construction in Canada (beginning with HLK dam in 1969) recruitment was somewhat variable (likely largely function of the sparseness of the data) but relatively consistent.
Figure 13. Year class histogram and recruitment index based on fin-ray ageing of 124 white sturgeon collected during a summertime setline survey of the Roosevelt Reach in 1998 (Devore et al 2000). The recruitment index is for the period 1941-1998 where apparent relative recruitments, Rt , for years t prior to the 1998 sampling event were back-calculated by assuming that Rt-a = Pa x eaM where Pa is the proportion of the sample fish that were age a at t = 1998 (McAdam et al. 2005). A natural mortality rate of 0.08 was assumed. Actual mortality rates are unknown as the population was exploited up until prohibition of harvest in the mid-1990s.
2. History: Accomplishments, Results, and Adaptive Management (ISRP Review of Results) Regarding growth, the authors reported that "the estimate of growth co-efficient, K, was substantially greater in magnitude, and resulting growth trajectories predicted that sturgeon in the Roosevelt Reach attain larger sizes at younger ages than observed in other areas of the Columbia River (Howell and McLellan in prep; Figure 5)." It is confusing as to why the growth of the Lake Roosevelt fish would be faster than others but from Keenleyside slower than the others. This does not appear to make sense. Clarification would be helpful. RESPONSE: The VBG parameters reported in the proposal were derived from a preliminary analysis of mark-recapture (M-R) data collected in the Roosevelt Reach from 1998-2009. In retrospect, the direct comparison of the M-R derived growth curve for the Roosevelt Reach to those reported for other sturgeon populations that were based on fin ray ageing was likely not a fair one. Juvenile fish grew rapidly in the Roosevelt Reach 1998-2009 likely reflecting the low densities of this size class (due to recruitment failure and limited hatchery introductions) relative to other populations and is therefore not representative of historical norms. Following on from this, a single von Bertalanffy growth (VBG) model based on 1998-2009 M-R data for the Roosevelt Reach may not provide an adequate description of sturgeon growth across all size classes. A distinct change in the growth trajectory is apparent between approximately 100-150 cm FL in plots of growth rate at length (see figure below). While this change could represent the transition to sexual maturity or ontogenetic changes in behavior it is, again, more likely a density dependent effect. The initial M-R VBG analysis also utilized a substantial amount of data from hatchery fish that were only captured once following release. We now consider inclusion of these fish into the analysis as inappropriate since it likely incorporates a post-release growth phase that is potentially not representative of normal growth patterns due to factors such as high fish condition at release and an unknown period of post-release adjustment to the natural environment.
Figure 14. Mean annual growth in length for 178 sturgeon marked and subsequently recaptured in the Roosevelt Reach 1998-2009. Time at large for individual fish ranged 0.86 to 11.03 years (mean=3.62). Lines indicate growth rate trajectories estimated from VBG parameters derived from the mark-recapture data using Fabens (1965) growth increment equation. VBG parameters were variously estimated for: 1) a single analysis using all M-R data combined; 2) a composite analysis where parameters were estimated separately for the <100 cm FL, and >150 cm FL size classes (parameters for 100-150 cm FL size class were estimated via linear interpolation and thus represent a crude approximation); and 3) females.
We re-analyzed the M-R data (Fabens 1965) by considering the <100 cm FL (“juveniles”; hatchery single captures excluded) and >150 cm FL (“adults”) separately and assuming VBG for each stanza. VBG parameters for the 100-150 cm FL size class (“sub-adults”) were then crudely approximated via linear interpolation as data for this size class was sparse. We also subjected fin ray ageing data collected from sturgeon (n=124) in the Roosevelt Reach in 1998 to re-analysis by constraining L∞ to a more realistic value of 282 cm FL (previous unconstrained value = 255 cm). This value was derived by applying the observed maximum size of sturgeon in the Roosevelt Reach to date, Lmax, (279 cm FL) to Froese and Binholan’s (2000) invariate relationship: L∞=Lmax/0.985. Plots of growth rate at length derived from estimates of L∞ and K from the M-R analysis showed growth for the <100 cm FL size class was much greater than in other areas of the Columbia; again likely a reflection of the low densities of this size class in the Roosevelt Reach during the period of interest (Figure 15). The high rates of growth for this size class suggest that supplementation efforts are not yet overseeding available habitat. In contrast, growth rates of Roosevelt Reach sturgeon >150 cm FL were comparable to similar sized sturgeon in Bonneville reservoir and were greater than similarly sized Keenleyside Reach fish (Figure 15). Growth rates at length derived from 1998 fin ray ageing with L∞ constrained to 282 were comparable to M-R estimates for the >150 cm FL size class and, hence, Bonneville reservoir also (Figure 15).
Figure 15. Comparisons of approximate annual growth at length for the Roosevelt Reach and various other Columbia River populations (LCR = lower Columbia River below Bonneville Dam; BON = Bonneville reservoir; TDA = The Dalles reservoir; JDA = John Day reservoir; KR=Keenleyside Reach). Plots were derived from published VBG parameter values that were based on fin ray ageing (LCR, BON, TDA, JDA; [Beamesderfer et al. 1995]; KR [RL&L 2007], Roosevelt Reach 1998 [Devore et al. 2000], Roosevelt Reach 1998 (Linf constrained) and 1998-2009 Roosevelt Reach mark-recapture data. Data for the 1998-2009 Roosevelt Reach mark-recapture 100-150 cm FL stanza is omitted for the purposes of clarity but essentially connects the <100 cm FL and >150 cm FL plots. Disparities in sturgeon growth between the Roosevelt (Grand Coulee Dam to border) and Keenleyside (border to HLK Dam) Reaches likely reflect differences in habitat coupled with the tendency of fish to habituate to localized areas for relatively long periods of time (years). The international border marks the approximate upper boundary of the river-reservoir transition zone of the Transboundary Reach (Grand Coulee to HLK Dam) of the Columbia River, and thus the Keenleyside Reach is riverine year-round whereas the areas of the Roosevelt Reach in which sturgeon are found principally represents transition-zone and upper reservoir habitat. Conventional tagging (PIT) and telemetry (acoustic, radio) studies conducted in the transboundary reach to date indicate that, with the exception of apparent spawning migrations, sturgeon tend to reside in one or other of these general areas for long periods. For example, Hildebrand et al (1999) found that during the period 1990-1995 approximately 98% of conventionally tagged fish in the Keenleyside Reach moved less than 10km between successive captures. In the Roosevelt Reach, Brannon and Setter (1992) acoustic tagged 16 sturgeon between 1988-1990, of which none apparently moved upstream into the Keenleyside Reach. During the period 2004-2007, only four of 75 wild fish outfitted with acoustic tags (primarily sub-adults and adults) spent extended periods (several months to years) of time in both the Keenleyside and Roosevelt Reaches (Howell and McLellan 2011). Further, of the approximately 733 individual wild sturgeon handled by the LRSRP in the Roosevelt Reach through 2009, only 21 (~3%) were found to have been tagged in the Keenleyside Reach. Using movement information from mark-capture data collected since 1990, van Poorten and McAdam (2010) suggested that sturgeon in the Keenleyside Reach were spatially segregated into two groups (HLK and Waneta groups) and found differences in growth and metabolism between them, observing that the Waneta group exhibited lower growth which the authors attributed to food limitation. If there are observable differences in growth and metabolism within the Keenleyside Reach, then it is perhaps not surprising apparent differences exist between the Keenleyside and Roosevelt Reaches generally. Sturgeon growth in the Roosevelt Reach may be greater for various reasons that could include greater productivity and wider prey base, and lower energetic costs than might be expected for fish inhabiting more riverine habitat.
4. Deliverables, Work Elements, Metrics, and Methods Similarly, with regard to food limitation, the sponsors state, "We also intend to compare histology of post-feeding stages of white sturgeon collected during field surveys with reference specimens to identify starvation effects in wild fish, thus determine the role food availability plays as a limiting factor in sturgeon survival." Has this approach been used successfully elsewhere? Please provide background and references. RESPONSE: Gisbert and Doroshov (2003) conducted histological analysis of larval green sturgeon digestive tracts to identify effects of food deprivation. Histological differences were identified in food-deprived larval fish as evidenced by progressive deterioration of the digestive system after 10-15 days of starvation. The LRSRP plans to compare histology of post-feeding stages of white sturgeon to identify starvation effects on wild fish. The LRSRP histology study will be modeled after the Gisbert and Doroshov (2003) study. Food limitation (including quantity and quality) has been identified as a possible limiting factor in recruitment failure of white sturgeon in the transboundary reach. This analysis may provide important insight into possible habitat food limitation and the role it plays in sturgeon survival and recruitment.
4. Deliverables, Work Elements, Metrics, and Methods Although year class strength has at least tentatively been associated with higher flows, it does not seem that any recommendations have been forwarded to test flow augmentation during late spring early summer to improve natural reproduction and recruitment. Has this topic been adequately investigated? Have recommendations been made? The topic of flow augmentation during the late spring and early summer has not been thoroughly investigated at this time. The upper Columbia white sturgeon managers feel it is perhaps premature to develop recommendations for flow augmentation prior to completion of critical components of the Upper Columbia White Sturgeon Recovery plan. Comprehensive habitat assessments (proposed under the CCT project BPA 200811600) and other proposed work will provide a better understanding of dispersal patterns of sturgeon and habitat availability in the transboundary reach. In addition, costs associated with testing flow augmentation due to loss of power generation are potentially prohibitive. Given the critical data gaps associated with habitat availability, and likely high costs associated with flow augmentation, recommendations have not been made at this time.
4. Deliverables, Work Elements, Metrics, and Methods More details are needed of the contaminant work to be performed and the protocols and methods. RESPONSE: The contaminant work is not proposed under this project, but rather under the CCT White Sturgeon Enhancement Project (BPA 200811600).
Literature Cited This page has citations that were not included in the proposal. All other citations can be found in the literature cited section of the Lake Roosevelt Sturgeon Recovery Project (BPA 1995-027-00) proposal. Conte, F.S. Doroshov, S.I., Lutes, P.B., and E.M. Strange. 1988. Hatchery manual for the white sturgeon, Acipenser transmontanus Richardson. Cooperative Extension University of California, Division of Agriculture and Natural Resources, Publ. 3322, 104 pp. Froese, R. and C. Binohlan. 2000. Empirical relationships to estimate asymptotic length, length at first maturity and length at maximum yield per recruit in fishes, with a simple method to evaluate length frequency data. J. Fish Biol. 56:758-773. van Poorten, B.T. and S.O. McAdam. 2010. Estimating differences in growth and metabolism in two spatially segregated groups of Columbia River white sturgeon using a field-based bioenergetics model. The Open Fish Science Journal 3:132-141. |