The responses to ISRP's concern were not very concise and leave a lingering concern that the proponents are doing work that might be more suitable for agencies that are directly concerned with harvest management. As the proponent states:

"Generating reliable harvest estimates within the Columbia River is problematic but is critical to NMFS, Pacific Salmon Commission (PSC), TAC, States and Tribes for effective management and recovery of salmon/steelhead stocks" and if so, these agencies should step up to the plate. Perhaps they are already involved in funding or in-kind support, but this was not clear from the response.

While the radio tagging studies will no doubt provide some very interesting and useful data on straying, pre spawn mortality, fall back etc, some of these questions could be addressed with specific hypotheses and investigations. Perhaps a subset of the large array of devices could be used in such studies.

The project sponsor did a reasonable job of elaborating on the justification for radio tagging adult salmon to assess certain survival, harvest, and straying questions.

There was insufficient explanation of the sample sizes and the number of stocks that would be evaluated. The 600-800 transmitters for spring-summer Chinook and steelhead, respectively, seems to be a rather small sample of the total number of fish passing Bonneville. If they don't achieve a 25% recovery rate of deployed tags as anticipated, they may be looking at a transmitter shortage. There is a lack of specificity about which stocks they will monitor and why. Specifically, sponsors suggest that 600 to 800 tags are needed per stock to evaluate straying, pre-spawn mortality, etc. Yet they are only asking for only 1300 tags, and hope to re-use 200 or so from early in the season. This means only two stocks will be evaluated each year. With eight or so listed ESUs, migrating above Bonneville, they don't justify how this effort will be sufficient, how the monitoring will be sequenced by stock over the years. This certainly cannot be sufficient to support a basinwide estimate of adult survival. It is not clear that this will have direct linkage to management decisions that provide benefits to fish.

In addition to responding to ISRP questions, the ISRP recommends that sponsors clearly articulate and prioritize the objectives and try to narrow down the objectives to 2 or 3 that may be manageable.

As a first comment, however, this proposal reads as though the past projects that had been radio tagging adult salmon at Bonneville and monitoring their upstream migration to various points in the hydrosystem were winding down, and the sponsors of those past projects were searching for a use of the antenna array. It would be unfortunate if the array fell into disrepair and would be unavailable in the future. In the end, a decision to maintain the array when there is not an immediate use for a management purposes is an administrative decision. In the proposal it is not clear who is going to use the data or who is calling for the data. Past projects (or agencies) that have used data on radio-tagged salmon are identified, but not by project. Only one project 200714400 is clearly identified as needing radio telemetry monitoring of salmon, at this time. The proposal does not do a good job of justifying the need to gather the radio telemetry observations. The uses of past data are not well covered.

A response should indicate who wants the data and how it is going to be used in management.

A response should also explain the statistical basis for the design for testing escapements into different subbasins. The design for testing escapements and straying into various tributaries and reservoirs is descriptive. A discussion on a design for testing differences (e, g., ANOVA or an alternative) is needed.

There are multiple options for collecting data to estimate the various parameters that the sponsors indicate need to be determined - abundance, harvest rate, straying, fallback. The sponsors propose using the existing telemetry array and radio tagging fish to accomplish the task. There is inadequate justification of the benefits and costs of various options to collect the data - PIT tags versus radio tags, etc.

The budget includes the cost of 1,300 radio transmitters but no additional antennas, so the comment in the proposal (page 8) that "fixed aerial antennas will be installed in all major Columbia River tributaries..." was confusing.

Regarding tagging and sampling effort, the numbers to be tagged and the choice of fish were vague. For example, "known-source" fish with PIT tags (but unmarked so presumably natural) are preferred, but some hatchery-origin fish may be tagged. "Exact numbers of tagged fish for these studies will be determined by research needs and resources available." It seems like the purpose for radio tagging these fish is not yet established. The critical research questions (needs above) are not decided, so the numbers of fish to be tagged is not yet determined. This does not convince a reviewer of the essential need to use the array and radio tagging to obtain data on fish abundance and estimate vital fish population statistics from it.

The methods for biotelemetry work seem appropriate and established. The proposal would benefit from a back up plan if low escapements result in fewer fish to work with. How would smaller number affect variances?

Studies proposed under Objectives 4 (spill effects) and 5 (removable spillway weir, marine mammals) require specific experimental designs. These are complex topics and the present proposal gives only simplistic designs to study them. Methods for objective 4, regarding spill effects are not explained. More details on the design of experiments are needed as well as more information on the "innovative modeling" (proportional hazards regression). Were the results of the past research on this topic subjected to peer review?

Assignment of fish stock to "unknown-source" radio-tagged fish - Lundrigan et al. 2004 - was not in the citations. NOAA Fisheries is identified in the proposal as a contract provider to genotype fish and perform assignment of individuals. There is no evidence that NOAA knows it is a contract service provider for this proposal.

Regarding calculation of vital rates, the biggest question is the magnitude of the unknown losses and whether those overwhelm the precision of the estimates of real interest. There is little supporting documentation from the earlier investigations to put confidence limits on the estimates. Data from earlier work could be used to provide an idea about the sample sizes required and the quality of the data.