The ISRP believes this proposal is not fundable for the following reasons:

The symptoms of this watershed problem seem to be a mile-long channelized length of Rock Creek with excessive sediment and with poor in-channel and riparian habitat. The existing conditions are not clearly described for reviewers who are not familiar with the site. No maps, drawings or photos are provided, but they are needed. This proposal characterizes the problem as inadequate sediment transport capacity, with an over-widened, shallow channel. It does not provide a description of the channel: floodplain dimensions, channel planform/sinuousity conditions or the condition of bank and floodplain vegetation.

The technical background section adequately describes in general terms what is proposed, but it does not provide enough information about the status of the fish using Rock Creek or the extent to which the current habitat conditions have departed from pre-development conditions. The proposal is to rehabilitate about one mile of stream. Approximately how many fish used this reach in the past, and how does that compare to its current capacity? A high degree of accuracy is not needed, but the proposal does not really address the issue of how Rock Creek productivity will benefit from the restoration efforts. Additionally, it would have been helpful to cite other projects that have used a similar hard engineering approach to habitat restoration and have been able to demonstrate a significant increase in salmon production.

The proposal mentions that the role of tributaries as important spawning and rearing areas is recognized in the Lower Columbia subbasin plan, but it is not clear if Rock Creek was specifically identified as a stream in need of significant restoration. Nearby Hamilton Creek is targeted as an important chum salmon spawning site in several plans, but particular references to Rock Creek are missing from this proposal. If chum salmon are the primary focal species for this project (they are identified as such in Section 3 of the cover pages), how much potential chum production can be achieved by rehabilitating a mile of stream?

The planting plan is vague; without reference to the engineering works, it gives the impression that planting is unrelated to the stream engineering works. The prescriptions are large rock and large wood jam structures, with no mention of use of woody vegetation for bank stabilization or channel narrowing. The relatively hard engineering approach proposed here is unlikely to achieve the stated objectives.

The concern is that this proposal will result in inappropriate alteration of a riparian corridor to a narrowed channel, with a resulting lack of spawning gravels in the channel owing to excessive sediment transport capacity. In this situation you might want to give more room for flood flows, not less; the flushing of sediments approach could scour the channel bed, resulting in headcutting.

This proposal does not provide enough information to justify the proposed actions. Of specific concern is that the project does not seem to address the cause of the problems. If flood flows are given proper access to the floodplain, then lower velocity flows will result in sediment sorting that will replenish spawning gravels by hydraulic forces. This can be achieved best by the use of bioengineered structures such as willow baffles for bank protection and finer sediment retention, connected with brush mattress design on the floodplain banks. This kind of approach will significantly increase habitat functions sustainably and cost-effectively, but a level of expertise is needed that is not evident on the team.

This project is essentially a band-aid on some other problems. In fact, the following quotes are from sections of the CGTB watershed plan and indicate an entirely different approach:

1. Restoration of degraded channel habitat in Rock Creek may require action outside the targeted reach, often extending into riparian and hillslope (upland) areas that are believed to influence the condition of aquatic habitats.

2. Sediment conditions in Rock Creek will remain moderately impaired to impaired until headwaters sediment sources are addressed.

The County should be complimented for their intent to increase fish habitat and asked to ensure the more watershed-based, passive approach indicated in 1 and 2 above.

In more detail, Table 8. "Prioritized measures for the Columbia Gorge Tributaries Basin" lists measures to improve fish habitat conditions. The 1st location is the lower mainstem Rock Creek up to Rock Creek Falls (RM 1), for anadromous access. However, the prioritized submeasures for Measure #1, Protect stream corridor structure and function are:
A. Protect floodplain function and channel migration processes
B. Protect riparian function
C. Protect access to habitats
D. Protect instream flows through management of water withdrawals
E. Protect channel structure and stability
F. Protect water quality
G. Protect the natural stream flow regime

Floodplain function and channel migration processes must be maintained together with increasing riparian structure and function. Protecting channel structure and stability should follow achievement of the previous priorities, especially by addressing the excessive sediments being delivered to the lower mainstem from upstream logging-related roads and landslides.

Rather than concentrate flows by altering channel structure in the best habitat reach in order to flush sediments through this lower reach, the published priorities would imply instead some type of a revegetation/ soil bioengineering approach to restore the stream systems' sediment sorting and storage capabilities. This can be achieved while decreasing the lower channel width/depth ratio and rebuilding the lost floodplain by trapping sediments. In addition, watershed restoration should address upstream logging- related landslides and road crossings.

Table 7 summarizes the Limiting Factors for habitat conditions on Rock Creek. The list does not include the proposal proponents' assertion that channel instability is the primary problem.

Considering the relatively high cost of the project, the likelihood of continued maintenance, and some questions about (1) applying these particular restoration procedures to this channel type, and (2) whether a passive restoration approach would be more appropriate and cost-effective in the long term, this proposal needs a better scientific justification and some prediction of the effects of the project on salmonid productivity.

Other review comments:

Relationships to other projects are inadequately described; the proposal does not seem to be integrated with fish and wildlife programs. More details, and references to ongoing salmon enhancement projects of similar intent in the Columbia Gorge, are needed.

The other work described is for maintenance and replacement of a bridge, already initiated by Skamania County. There does not appear to be any interdependency or collaboration between the two proposed projects.

The objectives are clear in terms of the number of instream structures that will be placed and streambanks that will be replanted with native trees, but there are a number of unanswered questions.

It appears that this proposal will attempt to turn a plane bed channel into a forced pool-riffle channel (using the classification terminology of Montgomery and Buffington). If, in fact, the Rock Creek watershed continues to experience episodic high intensity erosion events, it seems very likely that the hard engineered structures will require frequent, expensive maintenance as the stream naturally tries to return to a sediment-rich plane bed channel. Is the high initial cost and possibly frequent maintenance the best use of restoration dollars?

The types of structures described in the proposal are typical of a Rosgen-type restoration project where the objective is to create better salmonid rearing habitat but it seems the real target species is chum salmon, which may not really benefit from the pool habitat that may be created. The proposal mentions potential benefits for coho, Chinook, and steelhead but do we know whether these species actually use Rock Creek?

The objective of getting rid of the non-native riparian plant species and replacing them with native species is a good one. What will be done to ensure these native plants survive?

To achieve the objectives stated, the equipment and personnel are probably adequate, but again the size and number of large machines required are such that there will be a substantial impact on the local environment, compared with the alternative suggested above.

Plans appear to be limited to annual reports to BPA, with occasional presentations to the public. There is no mention of data archiving, or storage of time-series photos, and possible volunteer actions outside the scope of the project (photographs to be supplied to the client by schools taking part).

Unless funding for long-term maintenance of the engineered features of Rock Creek is guaranteed, benefits are likely to be short term and harm may possibly result. If the stream sinuousity is indeed "fixed" by this proposal, it will persist over the long-term, but this may be deleterious to the fish population.